- Email: [email protected]
The Transfemoral Approach for Removal of Well-Fixed Femoral Stems in 2-Stage Septic Hip Revision
The Journal of Arthroplasty xxx (2015) 1e7
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Original article
The Transfemoral Approach for Removal of Well-Fixed Femoral Stems in 2-Stage Septic Hip Revision Bernd Fink, MD a, b, *, Damian Oremek, MD a a b
€ningen gGmbH, Markgro €ningen, Germany Department of Joint Replacement, General and Rheumatic Orthopaedics, Orthopaedic Clinic Markgro University Hospital Hamburg-Eppendorf, Hamburg, Germany
a r t i c l e i n f o
a b s t r a c t
Article history: Received 7 September 2015 Received in revised form 8 October 2015 Accepted 4 November 2015 Available online XXX
Background: The value of a transfemoral approach for removal of well-fixed infected hip arthroplasties in 2-stage revision is unclear, especially whether cerclages for closure of the flap in the first step lead to higher reinfection rates and whether reopening of the flap for reimplantation of a hip arthroplasty leads to a lower union rate of the bony flap. Methods: Seventy-six septic 2-stage revisions via a transfemoral approach with cerclages for closure of the flap in the first step and reopening of the flap for reimplantation were followed prospectively for a mean period of 51.2 ± 23.2 (24-118) months. Results: The union rate of the bony flap after reimplantation was 98.7%, and no recurrence of reinfection was recorded in 93.4% of all cases. Subsidence of the stem occurred at a rate of 6.6%, dislocation at a rate of 6.6%, and there was no aseptic loosening of the implants. The Harris Hip Score was 62.2 ± 12.6 points with the spacer and 86.6 ± 15.5 points 2 years after reimplantation. Nine fractures (11.8%) of the flap occurred during the operation because of osteolytic or osteoporotic weakness of the flap itself, but these all healed without further intervention. Conclusion: The transfemoral approach is a safe method for septic revision of well-fixed hip prostheses, and the use of cerclage wires for closing the osteotomy flap in the first stage does not appear to lead to a higher reinfection rate. Similarly, the reopening of the flap does not appear to decrease the union rate of the flap. © 2015 Elsevier Inc. All rights reserved.
Keywords: transfemoral approach extended trochanteric osteotomy septic two-stage revision hip revision arthroplasty stem removal
Periprosthetic infections associated with total hip arthroplasty are, with an incidence of approximately 1%, a rare but nevertheless serious complication of hip prosthetic implantation [1,2]. Late infections require revision surgery that involves removal of all foreign materials and radical debridement of the prosthetic bed [3,4]. The removal of all foreign materials can be a complex operation if a well-fixed cementless or cemented prosthetic stem is involved. The transfemoral approach has been shown to be of value in these cases during aseptic revision surgery [5e8]. However, the value of a transfemoral approach in septic 2-stage revision has not been adequately shown because there are only a few reports and these One or more of the authors of this paper have disclosed potential or pertinent conflicts of interest, which may include receipt of payment, either direct or indirect, institutional support, or association with an entity in the biomedical field which may be perceived to have potential conflict of interest with this work. For full disclosure statements refer to http://dx.doi.org/10.1016/j.arth.2015.11.008. * Reprint requests: Bernd Fink, MD, Clinic of Joint Replacement, General and €ningen gGmbH, KurtRheumatic Orthopaedics, Orthopaedic Clinic Markgro €ningen, Germany. Lindemann-Weg 10, 71706 Markgro http://dx.doi.org/10.1016/j.arth.2015.11.008 0883-5403/© 2015 Elsevier Inc. All rights reserved.
on small patient cohorts. Moreover, there are concerns that cerclages for closure of the flap in the first step as foreign materials may lead to higher reinfection rates, and reopening of the flap for reimplantation of a hip arthroplasty may influence the union rate of the bony flap. In addition, technical details of the surgical procedures described differ between the few reports. Morshed et al [9] carried out an extended trochanteric approach during 13 2-stage revisions. In these cases, the flap over the spacer implanted during the first stage was not closed because the surgeons wanted to avoid using foreign materials such as cerclage wires to fix the osteotomy. In contrast, Lim et al [10] fixed the flap with cerclage wires in 23 cases but, unlike Morshed et al [9], did not reopen the approach in the second stage and implanted the cementless revision stem via the endofemoral route. In a retrospective study carried out by Levine et al [11], the extended trochanteric osteotomy was reopened during the second stage in 12 of 23 cases and in 11 cases not. That study also involved a number of different revision stems (cementless modular, cementless monoblock, and cemented).
2
B. Fink, D. Oremek / The Journal of Arthroplasty xxx (2015) 1e7
For 2-stage, septic revision surgery of well-fixed implants, we favor the transfemoral approach and closure of the bony flap with cerclage wires to avoid migration of the flap, or its dislocation, as described by Morshed et al. [9]. We reopen the flap during the second stage by removing the cerclage wires so that we can carry out a second radical debridement of the prosthetic bed and ensure that the distally fixed, cementless, modular revision stem is correctly positioned in the isthmus of the femur with the fixation zone distal to the osteotomy. The lack of publications concerning the transfemoral approach to septic, 2-stage revision surgery means that there is little information regarding the healing ability of the osteotomy after it has been opened twice, the freedom from infection when cerclage wires are used to close the bony flap during the first stage of the procedure, the clinical outcome after placement of the spacer and the implantation of a new prosthesis, and the frequency of subsidence and loosening of the cementless revision stem after the implantation. The objective of the prospective study reported here was to provide answers to the following questions against a background of a systematic operative procedure: What is the frequency of complete union of the bony flap after reimplantation? How frequently does reinfection occur? What is the frequency of subsidence and loosening of the revision stem? What is the Harris Hip Score in the interim phase with the spacer, between the first and second step? What is the Harris Hip Score after the final reimplantation? What are the complications associated with this approach?
the periprosthetic infection, half were performed in other institutions in the patients with 2 operations and all in the patients with >2 operations. Eight patients exhibited fistulas in the hip region. The periprosthetic infection was diagnosed by aspiration of the hip joint, which is a standard procedure in our clinic before any revision of a hip prosthesis is carried out, and bacteriologic cultivation of the aspirated fluid was assessed for 14 days according to €fer et al [12]. According to the criteria of the Musculoskeletal Scha Infection Society, the prosthesis was declared as infected when a sinus tract was present or, in addition to the isolation of the microorganism, the serum C-reactive protein (CRP), the synovial white blood cell count, and the percentage of polymorphonuclear cells were elevated, or a purulence was present [13]. In 15 cases, an additional synovial biopsy was performed to isolate the microorganism. Bacteriologic and histologic examination according to the methods of Atkins et al [14], Virolainen et al [15], and Pandey et al [16] of the membrane at the site of loosening, which was removed during the operation, was carried out to confirm the original diagnosis. The microorganisms detected by these methods are listed in Table 1; it should be noted that 2 causative organisms were identified in 25 cases. The transfemoral approach was performed because the cemented or cementless stems were very well fixed. The transfemoral approach was carried out using a previous published modified Wagner technique [5,6,17e19]. With the patient in a lateral position, an extended posterolateral incision was made and the posterolateral edge of the femur ventral to the linea aspera was exposed in the lateral intermuscular septum after ligation of the perforating vessels. The lateral circumference of the femur was exposed in the area where the end of the osteotomy flap was going to be positioned and 3.2 mm holes drilled while cooling in the ventral and dorsal end of the intended flap. The ventromedial
Materials and Methods Two hundred fifteen patients with late periprosthetic infection of a hip endoprosthesis underwent septic, 2-stage, cementless prosthesis revision surgery between August 2004 and April 2013. In 81 patients, the transfemoral approach was used to remove a solid fixed stem, where the shape of the stem, the roughness of the stem surface or the cement mantle lead to a high risk of uncontrolled periprosthetic fractures during the revision. Two patients died from unrelated causes during the follow-up period, and 3 patients were lost for follow-up, so 76 patients were evaluated prospectively over a follow-up period of at least 2 years (51.2 ± 23.2; 24-118 months). The patient cohort consisted of 37 women and 39 men with an average age of 70.7 ± 9.8 years (43-90 years) and an average body mass index of 28.7 ± 5.5 (19.1-41.9). The original diagnosis that led to the primary arthroplasty was osteoarthritis in 73 cases, femoral head fracture in 2 cases, and 1 case of rheumatoid arthritis. There were 20 cemented and 56 cementless acetabular cups as well as 30 cemented stems (4 revision stems) and 46 cementless stems (16 revision stems). The average life span of the primary implant was 7.9 ± 7.8 years (0.2-26.5 years). In 30 cases, a primary implant was involved, but there were also 12 patients who had already undergone one revision operation. Ten patients had already undergone 2 revision operations (2 with one septic debridement without removal of the implants and 2 with 2-stage septic revision), 18 patients had had 3 operations (4 with one septic debridement without removal of the implants and 2 with additional 2-stage revision after one septic debridement), 4 patients had had 4 operations (one with septic debridement without removal of the implants and one with septic 2-stage revision), one patient 5 operations, and one patient had been given 8 operations with one septic 2-stage revision. Of the operations reported here for treating
Table 1 The Microorganisms Identified as the Cause of the Periprosthetic Infections. Microorganism
Number of cases
Staphylococcus epidermidis MRSE Staphylococcus aureus Propionibacterium acnes Staphylococcus capitis Staphylococcus hominis Propionibacterium granulosum Staphylococcus lugdunensis Staphylococcus warneri Staphylococcus caprae Actinomyces neuii Staphylococcus haemolyticus Streptococcus mitis Enterococcus faecalis Enterococcus faecium Peptostreptococcus micros Staphylococcus chromogenes Streptococcus agalactiae Staphylococcus simulans Corynebacterium striatum Streptococcus anginosus Rhizobium radiobacter Escherichia coli Streptococcus gordonii Dermabacter hominis Streptococcus oralis Corynebacterium minutissimum Corynebacterium jeikeium Corynebacterium amycolatum Propionibacterium propionicum Staphylococcus saccharolyticus Peptostreptococcus asaccharolyticus
34 9 9 7 6 4 4 4 3 3 2 2 2 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
MRSE, Methicilline-resistant staphylococcus epidermidis.
B. Fink, D. Oremek / The Journal of Arthroplasty xxx (2015) 1e7
trochanter region was osteotomized using a chisel at the vastogluteal border, and then, the dorsolateral osteotomy, the connecting osteotomy between the 2 drill holes, and the distal ventromedial osteotomy of about 3 cm were performed with an oscillating saw while cooling with water. The ventromedial osteotomy was completed with a chisel that was introduced into the already prepared distal, ventral osteotomy and then driven blind under the m. vastus lateralis to the proximal end of the osteotomy. Once the osteotomy was complete, the flap with the m. vastus medialis still attached was raised medially. After removal of all implants and debridement, the cup-shaped acetabulum spacer was formed out of antibiotic-loaded cement (with a specific mixture of antibiotics recommended by the microbiologist). The stem spacer component consisted of old prosthesis stem models, monoblock devices in most cases, no longer used for primary implantations, that were encased in antibiotic-supplemented cement and, just before implantation, coated in the patient's own blood to facilitate easier removal. Immediately after the cement-covered stem had been inserted, the bony flap was closed with double cerclage wires (1.5 mm diameter) and the excess cement was removed from the flap. The 2 components of the spacer were connected by a metal head (Fig. 1). According to the antibiotic susceptibility profile of the microorganisms, a specific mixture of antibiotics for use in the spacer cement and in the systemic treatment was recommended by our microbiologist (Table 2). A maximum of 10% of the total cement powder amount was added as an antibiotic to the industrially prepared Palacos G (4 times) and Copal cement (72 times; Heraeus, Darmstadt, Germany). The cement of the spacer contained 2 antibiotics in 22 cases and three in 54 cases (Table 2). The parenteral antibiotic therapy was designed specifically for each case by our microbiologist and initiated during surgery once the implant had been removed, the infected and ischemic tissues had been effectively debrided, and at least 5 samples of tissue had been obtained for the bacteriologic assessment (14 days of enrichment) from the joint capsule and from the membrane around the loosened region as well as from the purportedly infected tissues. The intravenously administered antibiotics are listed in Table 2; there were 36 cases in which 2 antibiotics were administered systemically. The high bioavailability of the antibiotics rifampicin and ciprofloxacin allowed their oral administration from the second day after surgery. The initial antibiotic regime was changed 3 times after final culture of the revision surgery samples because a difference in the susceptibility of detected microorganism was observed. The patient was discharged after 2 weeks of parenteral antibiotic therapy and mobilization with partial weight-bearing on the operated leg. Six weeks after the removal of the infected implant, the transfemoral approach was reopened for purposes of spacer removal and reimplantation of a cementless, press-fit acetabular component (Allofit S; Zimmer GmbH, Winterthur, Switzerland; 29 times), a trabecular metal cup with augments (Zimmer GmbH, Warsaw, IL; 8 times), a Ganz ring (33 times), or a Burch-Schneider cage (Zimmer GmbH, Winterthur, Switzerland; 6 times) and a modular cementless revision stem (Revitan curved; Zimmer GmbH). There were 11 cases in which destruction or widening of the isthmus required a static distal locking of the stem to improve its distal fixation as described in earlier reports [5,6,20,21]. This involved the attachment of 3 locking screws after the implantation of the distal Revitan component in each case, using the implantation alignment guide. After the implantation of the new prosthetic components, the transfemoral approach was closed again using new double cerclage wires of 1.5-mm thickness. During the reimplantation procedure, at least 5 samples of tissue were removed for bacteriologic examination, and then, the same parenteral antibiotic therapy was reinitiated. This was
3
continued for 2 weeks and then changed to the appropriate oral administration of the antibiotic for a period of 4 weeks. Rifampicin and ciprofloxacin were administered orally from the second day after surgery as before. The orally administered antibiotics are listed in Table 2; 2 antibiotics were administered in 54 cases. Osseous defects before revision of the infected prosthesis were classified according to the system of Paprosky et al [22e24] and for the femur were as follows: 7 occurrences of type 1, 29 occurrences of type 2, 29 of type 3A, 9 of type 3B, and 2 of type 4. Postoperatively, the leg was subjected to partial weight-bearing by loading with 10 kg for a period of 6 weeks. Thereafter, the weight-bearing was gradually increased to full weight-bearing 3 months postoperatively as described by other authors for other cementless revision stems [21,25e29]. The hip joint was not allowed to be flexed for >70 for 6 weeks after the operation to avoid movement of the bony flap. Patients were examined clinically and radiologically before the operation and then 6 weeks, 3 months, 6 months, 1 year, 2 years, and 5 years (in 15 cases) after the operation and at the latest followup. Inflammatory parameters (CRP) were also followed. According to Masri et al [30] and Zimmerli et al [31], a patient could be judged infection free at follow-up if he or she was free of clinical signs for infection (fever, local pain, redness, warmth, sinus tract infection) and had a CRP level <10 mg/dL. The clinical outcome was evaluated using the Harris Hip Score [32,33]. All radiographic examinations were assessed independently by the 2 authors. All measurements in a sequence of radiographs were corrected for magnification using the prosthetic head diameter as a reference as described by Nunn et al [34] The osteotomy site was considered to be radiologically healed if callus was seen bridging the site in both the anteroposterior and lateral planes in agreement with Chen et al [35] and Miner et al [36]. Subsidence was measured € hm using the technique of Callaghan et al [37] as described by Bo and Bischel [25,26] and McInnis et al [38]. According to van Houwelingen et al [29] and Pattyn et al [39], significant subsidence was defined as being >5 mm. Radiological evaluation of implant fixation of the femoral stem was assessed by the criteria of Engh et al [40] (bone ingrowth fixation, stable fibrous fixation, unstable fixation). The femoral score of Barnett and Nordin [41] was calculated immediately postoperatively and at the 2-year follow-up examination to analyze the postoperative bone formation of the proximal € hm et al [25,26] and McInnis et al [38]. The femur according to Bo difference between the Barnett and Nordin scores 2 years after the operation and the directly postoperative score were used to calculate the amount of proximal bone formation. Reliability for the radiographic examinations was high, with an intrarater/intraclass correlation coefficient of 0.99 and of 0.98 between raters, respectively, for the radiological parameters such as healing of the osteotomy, implant fixation, subsidence, and Barnett and Nordin score. The statistical analyses were conducted using the computer program SPSS for Windows (version 11.0; SPSS Inc, Chicago, IL). All subjects gave informed consent to participate in the study, and the protocol was approved by the research ethics boards of the respective institutions. Results None of the samples taken at the time of spacer removal and reimplantation were positive for bacterial infection. In one case, the inflammatory parameter still had not normalized by 6 weeks after surgery so a repeat debridement was carried out together with exchange of the spacer. Reimplantation of a cementless revision stem was carried out 6 weeks later after normalization of the inflammation parameter. There was a reoccurrence of the infection
4
B. Fink, D. Oremek / The Journal of Arthroplasty xxx (2015) 1e7
Fig. 1. Radiographs of a 62-year-old patient with 2-stage septic revision of an infected hip arthroplasty via a transfemoral approach of the right side. (A) Preoperative radiograph with an infected cementless hip arthroplasty with osteolysis in the region of the greater trochanter. (B) Postoperative radiograph with the interims prosthesis (spacer). (C) Radiograph 7 days postoperative after reimplantation. (D) Radiograph 3 months postoperative after reimplantation showing osseous consolidation of the bony flap. (E) Radiograph 2 years postoperative after reimplantation showing no subsidence of the stem and bone remodeling in the proximal region of the femur.
in 5 patients, 12, 13, 14, 38, and 60 months after reimplantation of the new implants with the same microorganism as in the first septic revision in 3 cases and with another microorganism in the 2 cases with the longer time period after the reimplantation (38 and 60 months). This leads to a success rate without reinfection of
93.4%. Three of these 5 patients with reinfection have had septic revision surgeries before. The length of the osteotomy flap of the transfemoral approach ranged from 13.0 to 30.9 cm with a mean of 18.9 ± 3.6 cm. Nine fractures (11.8%) of the flap occurred during the operation because
B. Fink, D. Oremek / The Journal of Arthroplasty xxx (2015) 1e7 Table 2 Local and Systemic Antibiotic Therapy. Antibiotic
Number Local
Flucloxacillin Rifampicin Vancomycin Ciprofloxacin Imipenem þ cilastatin Ampicillin Ampicillin þ sulbactam Amoxicillin þ clavulanic acid Gentamicin Gentamicin Levofloxacin Linezolid Penicillin G Clindamycin Cotrimoxazole Amoxicillin Lorafem Daptomycin Doxycycline Fosfomycin Ofloxacin
46
Number Intravenous 21 29 33
Number Oral
49 8
15 1
1 3 5
76 2 (add.) 1
72
7 1
33 9
for the rest of the study. Thus, significant subsidence occurred in 5 cases (6.6%), and all occurred during the first 6 months after revision surgery. Aseptic loosening of the implants was not observed during the study, and no breakage of, or radiolucency around, the interlocking screws in the cases in which they were used, was seen. According to Engh et al [40], 73 stems showed bone ingrowth fixation and 3 showed stable fibrous fixation. During the 2-year postoperative observation period, there was an average increase in the Barnett and Nordin femoral score [41] in the region of the osseous flap of 2.82 ± 10.04 (21.5 to 34.5) when compared to the values obtained immediately after the operation. This suggests that bone regeneration occurred in the proximal femur. However, the data were very heterogeneous. Discussion
4 7 11 2
1 1 1 4
5
1
add., additional.
of osteolytic or osteoporotic weakness of the flap itself. The individual pieces of the fractured flaps could be recovered intact and effectively joined together again using double-looped cerclage osteosynthesis. Examinations carried out 3 months after the operation revealed that the flap exhibited bony consolidation in 75 cases (98.7%). In one female patient with recurrent infection, there was no osseous consolidation of the flap of the osteotomy. Loss of blood during the operation amounted to 990 ± 580 mL on average with a range of 400-2600 mL. The Harris Hip Score improved from 46.9 ± 17.6 (5-88) points preoperative to 62.2 ± 12.6 (20-74) with the spacer before the second operation. All patients could walk without or with only mild pain with partial weightbearing. After reimplantation of the new implant, the Harris Hip Score increased continually during the follow-up period: it was 66.2 ± 11.5 (40-80) points 3 months postoperatively, 69.8 ± 15.1 (41-92) points 6 months postoperatively, 71.7 ± 16.4 (46-96) points 9 months postoperatively, 75.9 ± 15.8 (48-96) points 12 months postoperatively, 81.6 ± 16.2 (52-98) points 18 months postoperatively, and 86.6 ± 15.5 (59-100) points 24 months after the operation. Other complications included 2 dislocations that could be treated conservatively and 3 other dislocations that showed significant subsidence of 5 mm and had to be revised. Two of these were revised with a change of the head with an offset head system (Bio-Ball; Merete Medical, Berlin, Germany) and later showed a recurrence of the infection. The third one, with a bone defect of Paprosky IIIA, suffered an intraoperative fissure at the isthmus of the femur during implantation of the stem. He was revised with a change of the proximal component. Thus, the rate of dislocation was 6.6%. Additional significant subsidence of 5 mm was observed in 2 other cases, but these were without symptoms and persisted
Union of the bony osteotomy flap was 98.7% and had a comparable rate of success in all the studies of septic revision involving a transfemoral approach and does not appear to depend on whether the flap is fixed with cerclage wires during the first stage or not. Similarly, it does not appear to matter whether the flap is reopened during the second stage or not (Table 3). The deciding factor appears to be a sufficient vascular supply to the bone that, here, is assured by leaving intact the musculature attached to the flap; this was also observed by Meek et al [7]. The adequacy of the vascular supply may also be essential for the healing of fractures of the bony flap, which was seen in all 9 cases in our study. Use of cerclage wires to close the osteotomy flap during the first revision does not seem to lead to a higher rate of reinfection: after flap closure with cerclage wires, we and Lim et al [10] observed a rate of reinfection of 6.6% and 4%, respectively. These rates were lower than in the study of Morshed et al [9] who reported a reinfection rate of 23% without closure of the flap with cerclage wires. Overall, the success rates reported by Lim et al [10] and ourselves are comparable to those reported in other studies with a freedom from infection between 79% and 96% after an endofemoral 2-stage revision of cementless implants [30,42e45] and between 90% and 95% for 2-stage revisions with antibiotic-supplemented cementing [2,46,47]. Thus, the extended transfemoral approach for revision of well-fixed infected implants and cerclage fixation of the bony flap do not lead to a higher rate of reinfection. In fact, the transfemoral approach enables a reproducibly effective debridement of the infected prosthetic bed and of the osteolysis in the femur caused by the infection and the reopening of the flap a second radical debridement. This suggestion is supported by the high degree of freedom from infection noted in our study, despite the inclusion of 13 patients who had already undergone septic revision surgery. In our opinion, the results should be similar when using woven cables instead of cerclage wires. The frequency of subsidence and dislocation observed during our study is comparable to that reported by Lim et al [10] and Levine et al [11] and is similar to the frequencies reported for aseptic revisions with cementless, distally fixed, modular revision stems [20]. However, they are very much lower than the rates of 15% subsidence and 31% dislocation reported by Morshed et al [9].
Table 3 Results of 2-Stage Cementless Revision of Periprosthetic Infection of the Hip. Author
N
Follow-Up (mo)
Healing flap (%)
Success Rate (%)
Subsidence, >5 mm (%)
Dislocation (%)
Femoral Fracture (%)
Morshed et al [9] Lim et al [10] Levine et al [11] Current study
13 23 23 76
>24 63 (24-123) 49.1 (24-82) 51.2 (24-118)
100 100 96 98.7
77 96 87 93.4
15.4 4.3
30.8 4.3 8.7 6.6
23 8.7 8.7 1.3
HHS, harris hip score.
6.6
HHS 81.8 (59-93) 86.6 (59-100)
6
B. Fink, D. Oremek / The Journal of Arthroplasty xxx (2015) 1e7
On the one hand, this might be related to the fact that Morshed et al [9] used monoblock stems rather than the modular, cementless revision stems used in our studies. Modularity enables the 2 objectives of the operation to be considered separately: first, a firm, distal fixation of the stem is achieved in an adequately prepared osseous bed and, then, the variable proximal component is selected so that the correct level of antetorsion and the correct leg length are obtained [6,20,48]. On the other hand, the cerclage fixation of the bony flap could also contribute to the recovery of abductor tone that would then lead to a lower rate of dislocation than that reported by Morshed et al [9], who used delayed fixation of the flap. The patients with the interim prosthesis could be mobilized with partial weight-bearing, attained a Harris Hip Score of 62.2 ± 12.6, and could be discharged from the clinic until the second revision surgery was scheduled. The Harris Hip Scores assessed at the end of the 2-year follow-up period were similar to those reported in other studies with 1-stage and 2-stage septic revisions without a transfemoral approach [2,30,42,43,46,47,49,50]. Thus, our procedure, involving the transfemoral approach in both revisions, does not seem to have any negative influence on the functional and clinical outcomes in the long-term period. Although the present study represents the largest reported series evaluating the use of a transfemoral approach and/or an extended trochanteric osteotomy in treating prosthetic infections, it has some limitations. One is the relatively short period of followup observation. However, we suggest that a minimum of 2 years is sufficient for questions posed by the present study. Many other similar studies have shown that reinfection occurs within 2 years of reimplantation [30,51e53]. The 2 cases that exhibited an infection 38 and 60 months after reimplantation became infected with a different microorganism from that isolated at the first septic revision so these can be considered to be new infections rather than truly recurring infections with the same organism. Moreover, a subsidence of the stem and early loosening of the components would also be recognizable within this period [25,26,38]. In conclusion, the use of a transfemoral approach in the setting of infection allows a safe removal of a well-fixed cemented or cementless stem with a reproducible high healing rate of the bony flap, low reinfection rate, and good clinical results. It has the advantage of a wide exposure for complete removal, debridement, and direct preparation of the femoral canal. Acknowledgments The authors thank Lars Frommelt, MD, Service for Infectious Diseases, Clinical Microbiology and Infection Control, ENDO-Klinik, Hamburg, Germany, for his support in choosing the local and systemic antibiotic therapy for the patients of this study and his support in writing this article. References: 1. Fitzgerald Jr RH. Infected total hip arthroplasty: diagnosis and treatment. J Am Acad Orthop Surg 1995;3:249. 2. Garvin KL, Hanssen AD. Current concepts review: infection after total hip arthroplasty. J Bone Joint Surg 1995;77-A:1576. 3. Cui Q, Mihalko WM, Shields JS, et al. Antibiotic-impregnated cement spacers for the treatment of infection associated with total hip or knee arthroplasty. J Bone Joints Surg Am 2007;89:871. 4. Hanssen AD, Osmon DR. Evaluation of a staging system for infected hip arthroplasty. Clin Orthop Relat Res 2002;4003:16. 5. Fink B, Grossmann A. Modified transfemoral approach to revision arthroplasty with uncemented modular revision stems. Oper Orthop Traumatol 2007;19:32. 6. Fink B, Grossmann A, Schubring S, et al. A modified transfemoral approach using modular cementless revision stems. Clin Orthop Relat Res 2007;462:105. 7. Meek D, Greidanus NV, Garbuz DS, et al. Extended trochanteric osteotomy: planning, surgical technique, and pitfalls. Instr Course Lect 2004;53:119. 8. Paprosky WG, Sporer SM. Controlled femoral fracture. Easy in. J Arthroplasty 2003;18(Suppl 1):91.
9. Morshed S, Huffman R, Ries M. Extended trochanteric osteotomy for 2-stage revision of infected total hip arthroplasty. J Arthroplasty 2005;20:294. 10. Lim SJ, Moon YW, Park YS. Is extended trochanteric osteotomy safe for use in 2stage revision of periprosthetic hip infection? J Arthroplasty 2011;26:1067. 11. Levine BR, Della alle CJ, Hamming M, et al. Use of the extended trochanteric osteotomy in treating prosthetic hip infection. J Arthroplasty 2009;24:49. 12. Sch€ afer P, Fink B, Sandow D, et al. Prolonged bacterial culture to identify late periprosthetic joint infection: a promising strategy. Clin Infect Dis 2008;47: 1403. 13. Parvizi J, Zmistowski B, Berbari EF, et al. New definition of periprosthetic joint infection: from the Workgroup of the Musculoskeletal Infection Society. Clin Orthop Relat Res 2011;469:2992. 14. Atkins BL, Athanasou N, Deeks JJ, et al. Prospective evaluation of criteria for microbiological diagnosis of prosthetic-joint infection at revision arthroplasty. J Clin Microbiol 1998;36:2932. 15. Virolainen P, Lahteenmaki H, Hiltunen A, et al. The reliability of diagnosis of infection during revision arthroplasties. Scan J Surg 2002;91:178. 16. Pandey R, Drakouilakis E, Athanasou NA. An assessment of the histological criteria used to diagnose infection in hip revision arthroplasty tissues. J Clin Pathol 1999;52:118. €de 1989;18:438. 17. Wagner H. A revision prosthesis for the hip joint. Orthopa 18. Wagner H. Hip prosthesis revision with the non-cemented femoral revision stemd10 year experience. Med Orth Tech 1997;117:138. 19. Wagner M, Wagner H. The transfemoral approach for revision of total hip replacement. Oper Orthop Traumatol 1999;11:278. 20. Fink B, Urbansky K, Schuster P. Mid term results with the curved modular tapered, fluted titanium Revitan stem in revision hip replacement. Bone Joint J 2014;96B:889. 21. Fink B, Grossmann A, Fuerst M. Distal interlocking screws with a modular revision stem for revision total hip arthroplasty in severe bone defects. J Arthroplasty 2010;25:759. 22. Della Valle CJ, Paprosky WG. The femur in revision total hip arthroplasty evaluation and classification. Clin Orthop Rel Res 2004;420:55. guec P, Sieber HP, et al. Stem and osteotomy length are 23. De Menezes DFA, Le Be critical for success of the transfemoral approach and cementless stem revision. Clin Orthop Rel Res 2012;470:883. 24. Paprosky WG, Lawrence J, Cameron H. Femoral defect classification. Clinical application. Orthop Rev 1990;19(Suppl 9):9. €hm P, Bischel O. Femoral revision with the Wagner SL revision stem. Evalu25. Bo ation of one hundred and twenty-nine revisions followed for a mean of 4.8 years. J Bone Joint Surg Am 2001;83:1023. €hm P, Bischel O. The use of tapered stems for femoral revision surgery. Clin 26. Bo Orthop Relat Res 2004;420:148. €fer W, et al. Revision total hip arthroplasty: How do 27. Kessler S, Kinkel S, Ka metaphyseal onset, diaphyseal fill and three-point-stem-fixation influence the postoperative subsidence of a revision straight-stem? Z Orthop 2002;140:1. 28. Schuh A, Werber S, Holzwarth U, et al. Cementless modular hip revision arthroplasty using the MRP Titan Revision Stem: outcome of 79 hips after an average of 4 years' follow-up. Arch Orthop Trauma Surg 2004;124:306. 29. van Houwelingen AP, Duncan CP, Masri BA, et al. High survival of modular tapered stems for proximal femoral bone defects at 5 to 10 years follow-up. Clin Orthop Rel Res 2013;471:454. 30. Masri BA, Panagiotopoulos KP, Greidanus NV, et al. Cementless two-stage exchange arthroplasty for infection after total hip arthroplasty. J Arthroplasty 2007;22:72. 31. Zimmerli W, Widmer AF, Blatter M, et al. Role of rifampin for treatment of orthopedic implant-related staphylococcal infections: a randomized controlled trial. Foreign-Body Infection (FBI) Study Group. JAMA 1998;279:1537. 32. Harris WH. Traumatic arthritis of the hip after dislocation and acetabular fractures: treatment by mold arthroplasty. An end result study using a new method of result evaluation. J Bone Joint Surg Am 1969;51:737. 33. Klauser W, Bangert Y, Lubinus P, et al. Medium-term follow-up of a modular tapered noncemented titanium stem in revision total hip arthroplasty. A singlesurgeon experience. J Arthroplasty 2013;28:84. 34. Nunn D, Freeman MAR, Hil PF, et al. The measurement of migration of the acetabular component of hip prostheses. J Bone Joint Surg Br 1989;71-B: 629. 35. Chen WM, McAuley JP, Engh Jr CA, et al. Extended slide trochanteric osteotomy for revision total hip arthroplasty. J Bone Joint Surg Am 2000;82:1215. 36. Miner TM, Momberger NG, Chong D, et al. The extended trochanteric osteotomy in revision hip arthroplasty: a critical review of 166 cases at mean 3-year, 9month follow-up. J Arthroplasty 2001;16:188. 37. Callaghan JJ, Slavati EA, Pellicci PM, et al. Results of revision for mechanical failure after cemented total hip replacement, 1979 to 1982. A two to five-year follow-up. J Bone Joint Surg 1985;67-A:1074. 38. McInnis DP, Horne G, Dvane PA. Femoral revision with a fluted, tapered, modular stem: seventy patients followed for a mean of 3.9 years. J Arthroplasty 2006;21:372. 39. Pattyn C, Mulliez A, Verdonk R, et al. Revision hip arthroplasty using a cementless modular tapered stem. Int Orthop 2012;36:35. 40. Engh CA, Glassman AH, Suthers KE. The case of porous-coated hip implants: the femoral side. Clin Orthop Relat Res 1990;261:63. 41. Barnett E, Nordin BEC. The radiological diagnosis of osteoporosis: a new approach. Clin Radiol 1960;11:166.
B. Fink, D. Oremek / The Journal of Arthroplasty xxx (2015) 1e7 42. Fehring TK, Calton TF, Griffin WL. Cementless fixation in 2-stage reimplantation for periprosthetic sepsis. J Arthroplasty 1999;14:175. 43. Haddad FS, Muirhead-Allwood SK, Manktelow AR, et al. Two-stage uncemented revision hip arthroplasty for infection. J Bone Joint Surg 2000;82-B: 689. 44. Hofmann AA, Goldberg TD, Tanner AM, et al. Ten-year experience using an articulating antibiotic cement hip spacer for the treatment of chronically infected total hip. J Arthroplasty 2005;20:874. 45. Nestor BJ, Hanssen AD, Ferrer-Bonzalez R, et al. The use of porous prostheses in delayed reconstruction of total hip replacements that have failed because of infection. J Bone Joint Surg 1994;76-A:349. 46. Garvin KL, Evans BG, Salvati EA, et al. Palacos gentamicin for the treatment of deep periprosthetic hip infections. Clin Orthop Relat Res 1994;298:97. 47. Lieberman JR, Callaway GH, Salvati EA, et al. Treatment of the infected total hip arthroplasty with a two staged re-implantation protocol. Clin Orthop Relat Res 1994;301:205.
7
48. Fink B, Hahn M, Fuerst M, et al. Principle of fixation of the cementless modular revision stem Revitan. Unfallchirurg 2005;108:1029. 49. Choi HR, Kwon YM, Freiberg AA, et al. Comparison of one-stage revision with antibiotic cement versus two-stage revision results for infected total hip arthroplasty. J Arthroplasty 2013;28:66. 50. De Man FHR, Sendi P, Zimmerli W, et al. Infectiological, functional, and radiographic outcome after revision for prosthetic hip infection according to a strict algorithm. 22 One-stage and 50 two-stage revisions with a mean follow-up time of 5 (2-17) years. Acta Orthop 2011;82:27. 51. Kraay MJ, Goldberg VM, Fitzgerald SJ, et al. Cementless two-staged total hip arthroplasty for deep periprosthetic infection. Clin Orthop Relat Res 2005;441:243. 52. Lai KA, Shen WJ, Yang CY, et al. Two-stage cementless revision THR after infection. Acta Orthop Scand 1996;67:325. 53. Stockley I, Mockford BJ, Hoad-Reddick A, et al. The use of two-stage exchange arthroplasty with depot antibiotics in the absence of long-term antibiotic therapy in infected total hip replacement. J Bone Joint Surg 2008;90-B:145.
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Original article
The Transfemoral Approach for Removal of Well-Fixed Femoral Stems in 2-Stage Septic Hip Revision Bernd Fink, MD a, b, *, Damian Oremek, MD a a b
€ningen gGmbH, Markgro €ningen, Germany Department of Joint Replacement, General and Rheumatic Orthopaedics, Orthopaedic Clinic Markgro University Hospital Hamburg-Eppendorf, Hamburg, Germany
a r t i c l e i n f o
a b s t r a c t
Article history: Received 7 September 2015 Received in revised form 8 October 2015 Accepted 4 November 2015 Available online XXX
Background: The value of a transfemoral approach for removal of well-fixed infected hip arthroplasties in 2-stage revision is unclear, especially whether cerclages for closure of the flap in the first step lead to higher reinfection rates and whether reopening of the flap for reimplantation of a hip arthroplasty leads to a lower union rate of the bony flap. Methods: Seventy-six septic 2-stage revisions via a transfemoral approach with cerclages for closure of the flap in the first step and reopening of the flap for reimplantation were followed prospectively for a mean period of 51.2 ± 23.2 (24-118) months. Results: The union rate of the bony flap after reimplantation was 98.7%, and no recurrence of reinfection was recorded in 93.4% of all cases. Subsidence of the stem occurred at a rate of 6.6%, dislocation at a rate of 6.6%, and there was no aseptic loosening of the implants. The Harris Hip Score was 62.2 ± 12.6 points with the spacer and 86.6 ± 15.5 points 2 years after reimplantation. Nine fractures (11.8%) of the flap occurred during the operation because of osteolytic or osteoporotic weakness of the flap itself, but these all healed without further intervention. Conclusion: The transfemoral approach is a safe method for septic revision of well-fixed hip prostheses, and the use of cerclage wires for closing the osteotomy flap in the first stage does not appear to lead to a higher reinfection rate. Similarly, the reopening of the flap does not appear to decrease the union rate of the flap. © 2015 Elsevier Inc. All rights reserved.
Keywords: transfemoral approach extended trochanteric osteotomy septic two-stage revision hip revision arthroplasty stem removal
Periprosthetic infections associated with total hip arthroplasty are, with an incidence of approximately 1%, a rare but nevertheless serious complication of hip prosthetic implantation [1,2]. Late infections require revision surgery that involves removal of all foreign materials and radical debridement of the prosthetic bed [3,4]. The removal of all foreign materials can be a complex operation if a well-fixed cementless or cemented prosthetic stem is involved. The transfemoral approach has been shown to be of value in these cases during aseptic revision surgery [5e8]. However, the value of a transfemoral approach in septic 2-stage revision has not been adequately shown because there are only a few reports and these One or more of the authors of this paper have disclosed potential or pertinent conflicts of interest, which may include receipt of payment, either direct or indirect, institutional support, or association with an entity in the biomedical field which may be perceived to have potential conflict of interest with this work. For full disclosure statements refer to http://dx.doi.org/10.1016/j.arth.2015.11.008. * Reprint requests: Bernd Fink, MD, Clinic of Joint Replacement, General and €ningen gGmbH, KurtRheumatic Orthopaedics, Orthopaedic Clinic Markgro €ningen, Germany. Lindemann-Weg 10, 71706 Markgro http://dx.doi.org/10.1016/j.arth.2015.11.008 0883-5403/© 2015 Elsevier Inc. All rights reserved.
on small patient cohorts. Moreover, there are concerns that cerclages for closure of the flap in the first step as foreign materials may lead to higher reinfection rates, and reopening of the flap for reimplantation of a hip arthroplasty may influence the union rate of the bony flap. In addition, technical details of the surgical procedures described differ between the few reports. Morshed et al [9] carried out an extended trochanteric approach during 13 2-stage revisions. In these cases, the flap over the spacer implanted during the first stage was not closed because the surgeons wanted to avoid using foreign materials such as cerclage wires to fix the osteotomy. In contrast, Lim et al [10] fixed the flap with cerclage wires in 23 cases but, unlike Morshed et al [9], did not reopen the approach in the second stage and implanted the cementless revision stem via the endofemoral route. In a retrospective study carried out by Levine et al [11], the extended trochanteric osteotomy was reopened during the second stage in 12 of 23 cases and in 11 cases not. That study also involved a number of different revision stems (cementless modular, cementless monoblock, and cemented).
2
B. Fink, D. Oremek / The Journal of Arthroplasty xxx (2015) 1e7
For 2-stage, septic revision surgery of well-fixed implants, we favor the transfemoral approach and closure of the bony flap with cerclage wires to avoid migration of the flap, or its dislocation, as described by Morshed et al. [9]. We reopen the flap during the second stage by removing the cerclage wires so that we can carry out a second radical debridement of the prosthetic bed and ensure that the distally fixed, cementless, modular revision stem is correctly positioned in the isthmus of the femur with the fixation zone distal to the osteotomy. The lack of publications concerning the transfemoral approach to septic, 2-stage revision surgery means that there is little information regarding the healing ability of the osteotomy after it has been opened twice, the freedom from infection when cerclage wires are used to close the bony flap during the first stage of the procedure, the clinical outcome after placement of the spacer and the implantation of a new prosthesis, and the frequency of subsidence and loosening of the cementless revision stem after the implantation. The objective of the prospective study reported here was to provide answers to the following questions against a background of a systematic operative procedure: What is the frequency of complete union of the bony flap after reimplantation? How frequently does reinfection occur? What is the frequency of subsidence and loosening of the revision stem? What is the Harris Hip Score in the interim phase with the spacer, between the first and second step? What is the Harris Hip Score after the final reimplantation? What are the complications associated with this approach?
the periprosthetic infection, half were performed in other institutions in the patients with 2 operations and all in the patients with >2 operations. Eight patients exhibited fistulas in the hip region. The periprosthetic infection was diagnosed by aspiration of the hip joint, which is a standard procedure in our clinic before any revision of a hip prosthesis is carried out, and bacteriologic cultivation of the aspirated fluid was assessed for 14 days according to €fer et al [12]. According to the criteria of the Musculoskeletal Scha Infection Society, the prosthesis was declared as infected when a sinus tract was present or, in addition to the isolation of the microorganism, the serum C-reactive protein (CRP), the synovial white blood cell count, and the percentage of polymorphonuclear cells were elevated, or a purulence was present [13]. In 15 cases, an additional synovial biopsy was performed to isolate the microorganism. Bacteriologic and histologic examination according to the methods of Atkins et al [14], Virolainen et al [15], and Pandey et al [16] of the membrane at the site of loosening, which was removed during the operation, was carried out to confirm the original diagnosis. The microorganisms detected by these methods are listed in Table 1; it should be noted that 2 causative organisms were identified in 25 cases. The transfemoral approach was performed because the cemented or cementless stems were very well fixed. The transfemoral approach was carried out using a previous published modified Wagner technique [5,6,17e19]. With the patient in a lateral position, an extended posterolateral incision was made and the posterolateral edge of the femur ventral to the linea aspera was exposed in the lateral intermuscular septum after ligation of the perforating vessels. The lateral circumference of the femur was exposed in the area where the end of the osteotomy flap was going to be positioned and 3.2 mm holes drilled while cooling in the ventral and dorsal end of the intended flap. The ventromedial
Materials and Methods Two hundred fifteen patients with late periprosthetic infection of a hip endoprosthesis underwent septic, 2-stage, cementless prosthesis revision surgery between August 2004 and April 2013. In 81 patients, the transfemoral approach was used to remove a solid fixed stem, where the shape of the stem, the roughness of the stem surface or the cement mantle lead to a high risk of uncontrolled periprosthetic fractures during the revision. Two patients died from unrelated causes during the follow-up period, and 3 patients were lost for follow-up, so 76 patients were evaluated prospectively over a follow-up period of at least 2 years (51.2 ± 23.2; 24-118 months). The patient cohort consisted of 37 women and 39 men with an average age of 70.7 ± 9.8 years (43-90 years) and an average body mass index of 28.7 ± 5.5 (19.1-41.9). The original diagnosis that led to the primary arthroplasty was osteoarthritis in 73 cases, femoral head fracture in 2 cases, and 1 case of rheumatoid arthritis. There were 20 cemented and 56 cementless acetabular cups as well as 30 cemented stems (4 revision stems) and 46 cementless stems (16 revision stems). The average life span of the primary implant was 7.9 ± 7.8 years (0.2-26.5 years). In 30 cases, a primary implant was involved, but there were also 12 patients who had already undergone one revision operation. Ten patients had already undergone 2 revision operations (2 with one septic debridement without removal of the implants and 2 with 2-stage septic revision), 18 patients had had 3 operations (4 with one septic debridement without removal of the implants and 2 with additional 2-stage revision after one septic debridement), 4 patients had had 4 operations (one with septic debridement without removal of the implants and one with septic 2-stage revision), one patient 5 operations, and one patient had been given 8 operations with one septic 2-stage revision. Of the operations reported here for treating
Table 1 The Microorganisms Identified as the Cause of the Periprosthetic Infections. Microorganism
Number of cases
Staphylococcus epidermidis MRSE Staphylococcus aureus Propionibacterium acnes Staphylococcus capitis Staphylococcus hominis Propionibacterium granulosum Staphylococcus lugdunensis Staphylococcus warneri Staphylococcus caprae Actinomyces neuii Staphylococcus haemolyticus Streptococcus mitis Enterococcus faecalis Enterococcus faecium Peptostreptococcus micros Staphylococcus chromogenes Streptococcus agalactiae Staphylococcus simulans Corynebacterium striatum Streptococcus anginosus Rhizobium radiobacter Escherichia coli Streptococcus gordonii Dermabacter hominis Streptococcus oralis Corynebacterium minutissimum Corynebacterium jeikeium Corynebacterium amycolatum Propionibacterium propionicum Staphylococcus saccharolyticus Peptostreptococcus asaccharolyticus
34 9 9 7 6 4 4 4 3 3 2 2 2 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
MRSE, Methicilline-resistant staphylococcus epidermidis.
B. Fink, D. Oremek / The Journal of Arthroplasty xxx (2015) 1e7
trochanter region was osteotomized using a chisel at the vastogluteal border, and then, the dorsolateral osteotomy, the connecting osteotomy between the 2 drill holes, and the distal ventromedial osteotomy of about 3 cm were performed with an oscillating saw while cooling with water. The ventromedial osteotomy was completed with a chisel that was introduced into the already prepared distal, ventral osteotomy and then driven blind under the m. vastus lateralis to the proximal end of the osteotomy. Once the osteotomy was complete, the flap with the m. vastus medialis still attached was raised medially. After removal of all implants and debridement, the cup-shaped acetabulum spacer was formed out of antibiotic-loaded cement (with a specific mixture of antibiotics recommended by the microbiologist). The stem spacer component consisted of old prosthesis stem models, monoblock devices in most cases, no longer used for primary implantations, that were encased in antibiotic-supplemented cement and, just before implantation, coated in the patient's own blood to facilitate easier removal. Immediately after the cement-covered stem had been inserted, the bony flap was closed with double cerclage wires (1.5 mm diameter) and the excess cement was removed from the flap. The 2 components of the spacer were connected by a metal head (Fig. 1). According to the antibiotic susceptibility profile of the microorganisms, a specific mixture of antibiotics for use in the spacer cement and in the systemic treatment was recommended by our microbiologist (Table 2). A maximum of 10% of the total cement powder amount was added as an antibiotic to the industrially prepared Palacos G (4 times) and Copal cement (72 times; Heraeus, Darmstadt, Germany). The cement of the spacer contained 2 antibiotics in 22 cases and three in 54 cases (Table 2). The parenteral antibiotic therapy was designed specifically for each case by our microbiologist and initiated during surgery once the implant had been removed, the infected and ischemic tissues had been effectively debrided, and at least 5 samples of tissue had been obtained for the bacteriologic assessment (14 days of enrichment) from the joint capsule and from the membrane around the loosened region as well as from the purportedly infected tissues. The intravenously administered antibiotics are listed in Table 2; there were 36 cases in which 2 antibiotics were administered systemically. The high bioavailability of the antibiotics rifampicin and ciprofloxacin allowed their oral administration from the second day after surgery. The initial antibiotic regime was changed 3 times after final culture of the revision surgery samples because a difference in the susceptibility of detected microorganism was observed. The patient was discharged after 2 weeks of parenteral antibiotic therapy and mobilization with partial weight-bearing on the operated leg. Six weeks after the removal of the infected implant, the transfemoral approach was reopened for purposes of spacer removal and reimplantation of a cementless, press-fit acetabular component (Allofit S; Zimmer GmbH, Winterthur, Switzerland; 29 times), a trabecular metal cup with augments (Zimmer GmbH, Warsaw, IL; 8 times), a Ganz ring (33 times), or a Burch-Schneider cage (Zimmer GmbH, Winterthur, Switzerland; 6 times) and a modular cementless revision stem (Revitan curved; Zimmer GmbH). There were 11 cases in which destruction or widening of the isthmus required a static distal locking of the stem to improve its distal fixation as described in earlier reports [5,6,20,21]. This involved the attachment of 3 locking screws after the implantation of the distal Revitan component in each case, using the implantation alignment guide. After the implantation of the new prosthetic components, the transfemoral approach was closed again using new double cerclage wires of 1.5-mm thickness. During the reimplantation procedure, at least 5 samples of tissue were removed for bacteriologic examination, and then, the same parenteral antibiotic therapy was reinitiated. This was
3
continued for 2 weeks and then changed to the appropriate oral administration of the antibiotic for a period of 4 weeks. Rifampicin and ciprofloxacin were administered orally from the second day after surgery as before. The orally administered antibiotics are listed in Table 2; 2 antibiotics were administered in 54 cases. Osseous defects before revision of the infected prosthesis were classified according to the system of Paprosky et al [22e24] and for the femur were as follows: 7 occurrences of type 1, 29 occurrences of type 2, 29 of type 3A, 9 of type 3B, and 2 of type 4. Postoperatively, the leg was subjected to partial weight-bearing by loading with 10 kg for a period of 6 weeks. Thereafter, the weight-bearing was gradually increased to full weight-bearing 3 months postoperatively as described by other authors for other cementless revision stems [21,25e29]. The hip joint was not allowed to be flexed for >70 for 6 weeks after the operation to avoid movement of the bony flap. Patients were examined clinically and radiologically before the operation and then 6 weeks, 3 months, 6 months, 1 year, 2 years, and 5 years (in 15 cases) after the operation and at the latest followup. Inflammatory parameters (CRP) were also followed. According to Masri et al [30] and Zimmerli et al [31], a patient could be judged infection free at follow-up if he or she was free of clinical signs for infection (fever, local pain, redness, warmth, sinus tract infection) and had a CRP level <10 mg/dL. The clinical outcome was evaluated using the Harris Hip Score [32,33]. All radiographic examinations were assessed independently by the 2 authors. All measurements in a sequence of radiographs were corrected for magnification using the prosthetic head diameter as a reference as described by Nunn et al [34] The osteotomy site was considered to be radiologically healed if callus was seen bridging the site in both the anteroposterior and lateral planes in agreement with Chen et al [35] and Miner et al [36]. Subsidence was measured € hm using the technique of Callaghan et al [37] as described by Bo and Bischel [25,26] and McInnis et al [38]. According to van Houwelingen et al [29] and Pattyn et al [39], significant subsidence was defined as being >5 mm. Radiological evaluation of implant fixation of the femoral stem was assessed by the criteria of Engh et al [40] (bone ingrowth fixation, stable fibrous fixation, unstable fixation). The femoral score of Barnett and Nordin [41] was calculated immediately postoperatively and at the 2-year follow-up examination to analyze the postoperative bone formation of the proximal € hm et al [25,26] and McInnis et al [38]. The femur according to Bo difference between the Barnett and Nordin scores 2 years after the operation and the directly postoperative score were used to calculate the amount of proximal bone formation. Reliability for the radiographic examinations was high, with an intrarater/intraclass correlation coefficient of 0.99 and of 0.98 between raters, respectively, for the radiological parameters such as healing of the osteotomy, implant fixation, subsidence, and Barnett and Nordin score. The statistical analyses were conducted using the computer program SPSS for Windows (version 11.0; SPSS Inc, Chicago, IL). All subjects gave informed consent to participate in the study, and the protocol was approved by the research ethics boards of the respective institutions. Results None of the samples taken at the time of spacer removal and reimplantation were positive for bacterial infection. In one case, the inflammatory parameter still had not normalized by 6 weeks after surgery so a repeat debridement was carried out together with exchange of the spacer. Reimplantation of a cementless revision stem was carried out 6 weeks later after normalization of the inflammation parameter. There was a reoccurrence of the infection
4
B. Fink, D. Oremek / The Journal of Arthroplasty xxx (2015) 1e7
Fig. 1. Radiographs of a 62-year-old patient with 2-stage septic revision of an infected hip arthroplasty via a transfemoral approach of the right side. (A) Preoperative radiograph with an infected cementless hip arthroplasty with osteolysis in the region of the greater trochanter. (B) Postoperative radiograph with the interims prosthesis (spacer). (C) Radiograph 7 days postoperative after reimplantation. (D) Radiograph 3 months postoperative after reimplantation showing osseous consolidation of the bony flap. (E) Radiograph 2 years postoperative after reimplantation showing no subsidence of the stem and bone remodeling in the proximal region of the femur.
in 5 patients, 12, 13, 14, 38, and 60 months after reimplantation of the new implants with the same microorganism as in the first septic revision in 3 cases and with another microorganism in the 2 cases with the longer time period after the reimplantation (38 and 60 months). This leads to a success rate without reinfection of
93.4%. Three of these 5 patients with reinfection have had septic revision surgeries before. The length of the osteotomy flap of the transfemoral approach ranged from 13.0 to 30.9 cm with a mean of 18.9 ± 3.6 cm. Nine fractures (11.8%) of the flap occurred during the operation because
B. Fink, D. Oremek / The Journal of Arthroplasty xxx (2015) 1e7 Table 2 Local and Systemic Antibiotic Therapy. Antibiotic
Number Local
Flucloxacillin Rifampicin Vancomycin Ciprofloxacin Imipenem þ cilastatin Ampicillin Ampicillin þ sulbactam Amoxicillin þ clavulanic acid Gentamicin Gentamicin Levofloxacin Linezolid Penicillin G Clindamycin Cotrimoxazole Amoxicillin Lorafem Daptomycin Doxycycline Fosfomycin Ofloxacin
46
Number Intravenous 21 29 33
Number Oral
49 8
15 1
1 3 5
76 2 (add.) 1
72
7 1
33 9
for the rest of the study. Thus, significant subsidence occurred in 5 cases (6.6%), and all occurred during the first 6 months after revision surgery. Aseptic loosening of the implants was not observed during the study, and no breakage of, or radiolucency around, the interlocking screws in the cases in which they were used, was seen. According to Engh et al [40], 73 stems showed bone ingrowth fixation and 3 showed stable fibrous fixation. During the 2-year postoperative observation period, there was an average increase in the Barnett and Nordin femoral score [41] in the region of the osseous flap of 2.82 ± 10.04 (21.5 to 34.5) when compared to the values obtained immediately after the operation. This suggests that bone regeneration occurred in the proximal femur. However, the data were very heterogeneous. Discussion
4 7 11 2
1 1 1 4
5
1
add., additional.
of osteolytic or osteoporotic weakness of the flap itself. The individual pieces of the fractured flaps could be recovered intact and effectively joined together again using double-looped cerclage osteosynthesis. Examinations carried out 3 months after the operation revealed that the flap exhibited bony consolidation in 75 cases (98.7%). In one female patient with recurrent infection, there was no osseous consolidation of the flap of the osteotomy. Loss of blood during the operation amounted to 990 ± 580 mL on average with a range of 400-2600 mL. The Harris Hip Score improved from 46.9 ± 17.6 (5-88) points preoperative to 62.2 ± 12.6 (20-74) with the spacer before the second operation. All patients could walk without or with only mild pain with partial weightbearing. After reimplantation of the new implant, the Harris Hip Score increased continually during the follow-up period: it was 66.2 ± 11.5 (40-80) points 3 months postoperatively, 69.8 ± 15.1 (41-92) points 6 months postoperatively, 71.7 ± 16.4 (46-96) points 9 months postoperatively, 75.9 ± 15.8 (48-96) points 12 months postoperatively, 81.6 ± 16.2 (52-98) points 18 months postoperatively, and 86.6 ± 15.5 (59-100) points 24 months after the operation. Other complications included 2 dislocations that could be treated conservatively and 3 other dislocations that showed significant subsidence of 5 mm and had to be revised. Two of these were revised with a change of the head with an offset head system (Bio-Ball; Merete Medical, Berlin, Germany) and later showed a recurrence of the infection. The third one, with a bone defect of Paprosky IIIA, suffered an intraoperative fissure at the isthmus of the femur during implantation of the stem. He was revised with a change of the proximal component. Thus, the rate of dislocation was 6.6%. Additional significant subsidence of 5 mm was observed in 2 other cases, but these were without symptoms and persisted
Union of the bony osteotomy flap was 98.7% and had a comparable rate of success in all the studies of septic revision involving a transfemoral approach and does not appear to depend on whether the flap is fixed with cerclage wires during the first stage or not. Similarly, it does not appear to matter whether the flap is reopened during the second stage or not (Table 3). The deciding factor appears to be a sufficient vascular supply to the bone that, here, is assured by leaving intact the musculature attached to the flap; this was also observed by Meek et al [7]. The adequacy of the vascular supply may also be essential for the healing of fractures of the bony flap, which was seen in all 9 cases in our study. Use of cerclage wires to close the osteotomy flap during the first revision does not seem to lead to a higher rate of reinfection: after flap closure with cerclage wires, we and Lim et al [10] observed a rate of reinfection of 6.6% and 4%, respectively. These rates were lower than in the study of Morshed et al [9] who reported a reinfection rate of 23% without closure of the flap with cerclage wires. Overall, the success rates reported by Lim et al [10] and ourselves are comparable to those reported in other studies with a freedom from infection between 79% and 96% after an endofemoral 2-stage revision of cementless implants [30,42e45] and between 90% and 95% for 2-stage revisions with antibiotic-supplemented cementing [2,46,47]. Thus, the extended transfemoral approach for revision of well-fixed infected implants and cerclage fixation of the bony flap do not lead to a higher rate of reinfection. In fact, the transfemoral approach enables a reproducibly effective debridement of the infected prosthetic bed and of the osteolysis in the femur caused by the infection and the reopening of the flap a second radical debridement. This suggestion is supported by the high degree of freedom from infection noted in our study, despite the inclusion of 13 patients who had already undergone septic revision surgery. In our opinion, the results should be similar when using woven cables instead of cerclage wires. The frequency of subsidence and dislocation observed during our study is comparable to that reported by Lim et al [10] and Levine et al [11] and is similar to the frequencies reported for aseptic revisions with cementless, distally fixed, modular revision stems [20]. However, they are very much lower than the rates of 15% subsidence and 31% dislocation reported by Morshed et al [9].
Table 3 Results of 2-Stage Cementless Revision of Periprosthetic Infection of the Hip. Author
N
Follow-Up (mo)
Healing flap (%)
Success Rate (%)
Subsidence, >5 mm (%)
Dislocation (%)
Femoral Fracture (%)
Morshed et al [9] Lim et al [10] Levine et al [11] Current study
13 23 23 76
>24 63 (24-123) 49.1 (24-82) 51.2 (24-118)
100 100 96 98.7
77 96 87 93.4
15.4 4.3
30.8 4.3 8.7 6.6
23 8.7 8.7 1.3
HHS, harris hip score.
6.6
HHS 81.8 (59-93) 86.6 (59-100)
6
B. Fink, D. Oremek / The Journal of Arthroplasty xxx (2015) 1e7
On the one hand, this might be related to the fact that Morshed et al [9] used monoblock stems rather than the modular, cementless revision stems used in our studies. Modularity enables the 2 objectives of the operation to be considered separately: first, a firm, distal fixation of the stem is achieved in an adequately prepared osseous bed and, then, the variable proximal component is selected so that the correct level of antetorsion and the correct leg length are obtained [6,20,48]. On the other hand, the cerclage fixation of the bony flap could also contribute to the recovery of abductor tone that would then lead to a lower rate of dislocation than that reported by Morshed et al [9], who used delayed fixation of the flap. The patients with the interim prosthesis could be mobilized with partial weight-bearing, attained a Harris Hip Score of 62.2 ± 12.6, and could be discharged from the clinic until the second revision surgery was scheduled. The Harris Hip Scores assessed at the end of the 2-year follow-up period were similar to those reported in other studies with 1-stage and 2-stage septic revisions without a transfemoral approach [2,30,42,43,46,47,49,50]. Thus, our procedure, involving the transfemoral approach in both revisions, does not seem to have any negative influence on the functional and clinical outcomes in the long-term period. Although the present study represents the largest reported series evaluating the use of a transfemoral approach and/or an extended trochanteric osteotomy in treating prosthetic infections, it has some limitations. One is the relatively short period of followup observation. However, we suggest that a minimum of 2 years is sufficient for questions posed by the present study. Many other similar studies have shown that reinfection occurs within 2 years of reimplantation [30,51e53]. The 2 cases that exhibited an infection 38 and 60 months after reimplantation became infected with a different microorganism from that isolated at the first septic revision so these can be considered to be new infections rather than truly recurring infections with the same organism. Moreover, a subsidence of the stem and early loosening of the components would also be recognizable within this period [25,26,38]. In conclusion, the use of a transfemoral approach in the setting of infection allows a safe removal of a well-fixed cemented or cementless stem with a reproducible high healing rate of the bony flap, low reinfection rate, and good clinical results. It has the advantage of a wide exposure for complete removal, debridement, and direct preparation of the femoral canal. Acknowledgments The authors thank Lars Frommelt, MD, Service for Infectious Diseases, Clinical Microbiology and Infection Control, ENDO-Klinik, Hamburg, Germany, for his support in choosing the local and systemic antibiotic therapy for the patients of this study and his support in writing this article. References: 1. Fitzgerald Jr RH. Infected total hip arthroplasty: diagnosis and treatment. J Am Acad Orthop Surg 1995;3:249. 2. Garvin KL, Hanssen AD. Current concepts review: infection after total hip arthroplasty. J Bone Joint Surg 1995;77-A:1576. 3. Cui Q, Mihalko WM, Shields JS, et al. Antibiotic-impregnated cement spacers for the treatment of infection associated with total hip or knee arthroplasty. J Bone Joints Surg Am 2007;89:871. 4. Hanssen AD, Osmon DR. Evaluation of a staging system for infected hip arthroplasty. Clin Orthop Relat Res 2002;4003:16. 5. Fink B, Grossmann A. Modified transfemoral approach to revision arthroplasty with uncemented modular revision stems. Oper Orthop Traumatol 2007;19:32. 6. Fink B, Grossmann A, Schubring S, et al. A modified transfemoral approach using modular cementless revision stems. Clin Orthop Relat Res 2007;462:105. 7. Meek D, Greidanus NV, Garbuz DS, et al. Extended trochanteric osteotomy: planning, surgical technique, and pitfalls. Instr Course Lect 2004;53:119. 8. Paprosky WG, Sporer SM. Controlled femoral fracture. Easy in. J Arthroplasty 2003;18(Suppl 1):91.
9. Morshed S, Huffman R, Ries M. Extended trochanteric osteotomy for 2-stage revision of infected total hip arthroplasty. J Arthroplasty 2005;20:294. 10. Lim SJ, Moon YW, Park YS. Is extended trochanteric osteotomy safe for use in 2stage revision of periprosthetic hip infection? J Arthroplasty 2011;26:1067. 11. Levine BR, Della alle CJ, Hamming M, et al. Use of the extended trochanteric osteotomy in treating prosthetic hip infection. J Arthroplasty 2009;24:49. 12. Sch€ afer P, Fink B, Sandow D, et al. Prolonged bacterial culture to identify late periprosthetic joint infection: a promising strategy. Clin Infect Dis 2008;47: 1403. 13. Parvizi J, Zmistowski B, Berbari EF, et al. New definition of periprosthetic joint infection: from the Workgroup of the Musculoskeletal Infection Society. Clin Orthop Relat Res 2011;469:2992. 14. Atkins BL, Athanasou N, Deeks JJ, et al. Prospective evaluation of criteria for microbiological diagnosis of prosthetic-joint infection at revision arthroplasty. J Clin Microbiol 1998;36:2932. 15. Virolainen P, Lahteenmaki H, Hiltunen A, et al. The reliability of diagnosis of infection during revision arthroplasties. Scan J Surg 2002;91:178. 16. Pandey R, Drakouilakis E, Athanasou NA. An assessment of the histological criteria used to diagnose infection in hip revision arthroplasty tissues. J Clin Pathol 1999;52:118. €de 1989;18:438. 17. Wagner H. A revision prosthesis for the hip joint. Orthopa 18. Wagner H. Hip prosthesis revision with the non-cemented femoral revision stemd10 year experience. Med Orth Tech 1997;117:138. 19. Wagner M, Wagner H. The transfemoral approach for revision of total hip replacement. Oper Orthop Traumatol 1999;11:278. 20. Fink B, Urbansky K, Schuster P. Mid term results with the curved modular tapered, fluted titanium Revitan stem in revision hip replacement. Bone Joint J 2014;96B:889. 21. Fink B, Grossmann A, Fuerst M. Distal interlocking screws with a modular revision stem for revision total hip arthroplasty in severe bone defects. J Arthroplasty 2010;25:759. 22. Della Valle CJ, Paprosky WG. The femur in revision total hip arthroplasty evaluation and classification. Clin Orthop Rel Res 2004;420:55. guec P, Sieber HP, et al. Stem and osteotomy length are 23. De Menezes DFA, Le Be critical for success of the transfemoral approach and cementless stem revision. Clin Orthop Rel Res 2012;470:883. 24. Paprosky WG, Lawrence J, Cameron H. Femoral defect classification. Clinical application. Orthop Rev 1990;19(Suppl 9):9. €hm P, Bischel O. Femoral revision with the Wagner SL revision stem. Evalu25. Bo ation of one hundred and twenty-nine revisions followed for a mean of 4.8 years. J Bone Joint Surg Am 2001;83:1023. €hm P, Bischel O. The use of tapered stems for femoral revision surgery. Clin 26. Bo Orthop Relat Res 2004;420:148. €fer W, et al. Revision total hip arthroplasty: How do 27. Kessler S, Kinkel S, Ka metaphyseal onset, diaphyseal fill and three-point-stem-fixation influence the postoperative subsidence of a revision straight-stem? Z Orthop 2002;140:1. 28. Schuh A, Werber S, Holzwarth U, et al. Cementless modular hip revision arthroplasty using the MRP Titan Revision Stem: outcome of 79 hips after an average of 4 years' follow-up. Arch Orthop Trauma Surg 2004;124:306. 29. van Houwelingen AP, Duncan CP, Masri BA, et al. High survival of modular tapered stems for proximal femoral bone defects at 5 to 10 years follow-up. Clin Orthop Rel Res 2013;471:454. 30. Masri BA, Panagiotopoulos KP, Greidanus NV, et al. Cementless two-stage exchange arthroplasty for infection after total hip arthroplasty. J Arthroplasty 2007;22:72. 31. Zimmerli W, Widmer AF, Blatter M, et al. Role of rifampin for treatment of orthopedic implant-related staphylococcal infections: a randomized controlled trial. Foreign-Body Infection (FBI) Study Group. JAMA 1998;279:1537. 32. Harris WH. Traumatic arthritis of the hip after dislocation and acetabular fractures: treatment by mold arthroplasty. An end result study using a new method of result evaluation. J Bone Joint Surg Am 1969;51:737. 33. Klauser W, Bangert Y, Lubinus P, et al. Medium-term follow-up of a modular tapered noncemented titanium stem in revision total hip arthroplasty. A singlesurgeon experience. J Arthroplasty 2013;28:84. 34. Nunn D, Freeman MAR, Hil PF, et al. The measurement of migration of the acetabular component of hip prostheses. J Bone Joint Surg Br 1989;71-B: 629. 35. Chen WM, McAuley JP, Engh Jr CA, et al. Extended slide trochanteric osteotomy for revision total hip arthroplasty. J Bone Joint Surg Am 2000;82:1215. 36. Miner TM, Momberger NG, Chong D, et al. The extended trochanteric osteotomy in revision hip arthroplasty: a critical review of 166 cases at mean 3-year, 9month follow-up. J Arthroplasty 2001;16:188. 37. Callaghan JJ, Slavati EA, Pellicci PM, et al. Results of revision for mechanical failure after cemented total hip replacement, 1979 to 1982. A two to five-year follow-up. J Bone Joint Surg 1985;67-A:1074. 38. McInnis DP, Horne G, Dvane PA. Femoral revision with a fluted, tapered, modular stem: seventy patients followed for a mean of 3.9 years. J Arthroplasty 2006;21:372. 39. Pattyn C, Mulliez A, Verdonk R, et al. Revision hip arthroplasty using a cementless modular tapered stem. Int Orthop 2012;36:35. 40. Engh CA, Glassman AH, Suthers KE. The case of porous-coated hip implants: the femoral side. Clin Orthop Relat Res 1990;261:63. 41. Barnett E, Nordin BEC. The radiological diagnosis of osteoporosis: a new approach. Clin Radiol 1960;11:166.
B. Fink, D. Oremek / The Journal of Arthroplasty xxx (2015) 1e7 42. Fehring TK, Calton TF, Griffin WL. Cementless fixation in 2-stage reimplantation for periprosthetic sepsis. J Arthroplasty 1999;14:175. 43. Haddad FS, Muirhead-Allwood SK, Manktelow AR, et al. Two-stage uncemented revision hip arthroplasty for infection. J Bone Joint Surg 2000;82-B: 689. 44. Hofmann AA, Goldberg TD, Tanner AM, et al. Ten-year experience using an articulating antibiotic cement hip spacer for the treatment of chronically infected total hip. J Arthroplasty 2005;20:874. 45. Nestor BJ, Hanssen AD, Ferrer-Bonzalez R, et al. The use of porous prostheses in delayed reconstruction of total hip replacements that have failed because of infection. J Bone Joint Surg 1994;76-A:349. 46. Garvin KL, Evans BG, Salvati EA, et al. Palacos gentamicin for the treatment of deep periprosthetic hip infections. Clin Orthop Relat Res 1994;298:97. 47. Lieberman JR, Callaway GH, Salvati EA, et al. Treatment of the infected total hip arthroplasty with a two staged re-implantation protocol. Clin Orthop Relat Res 1994;301:205.
7
48. Fink B, Hahn M, Fuerst M, et al. Principle of fixation of the cementless modular revision stem Revitan. Unfallchirurg 2005;108:1029. 49. Choi HR, Kwon YM, Freiberg AA, et al. Comparison of one-stage revision with antibiotic cement versus two-stage revision results for infected total hip arthroplasty. J Arthroplasty 2013;28:66. 50. De Man FHR, Sendi P, Zimmerli W, et al. Infectiological, functional, and radiographic outcome after revision for prosthetic hip infection according to a strict algorithm. 22 One-stage and 50 two-stage revisions with a mean follow-up time of 5 (2-17) years. Acta Orthop 2011;82:27. 51. Kraay MJ, Goldberg VM, Fitzgerald SJ, et al. Cementless two-staged total hip arthroplasty for deep periprosthetic infection. Clin Orthop Relat Res 2005;441:243. 52. Lai KA, Shen WJ, Yang CY, et al. Two-stage cementless revision THR after infection. Acta Orthop Scand 1996;67:325. 53. Stockley I, Mockford BJ, Hoad-Reddick A, et al. The use of two-stage exchange arthroplasty with depot antibiotics in the absence of long-term antibiotic therapy in infected total hip replacement. J Bone Joint Surg 2008;90-B:145.