silver coated megaprostheses of the proximal tibia in patients with sarcoma

Accepted Manuscript Silver-coated megaprostheses of the proximal tibia in patients with sarcoma Jendrik Hardes, Marcel P Henrichs, Gregor Hauschild, Markus Nottrott, Wiebke Guder, Arne Streitbuerger PII: S0883-5403(17)30172-9 DOI: 10.1016/j.arth.2017.02.054 Reference: YARTH 55697 To appear in: The Journal of Arthroplasty Received Date: 20 September 2016 Revised Date: February 2017 Accepted Date: 20 February 2017 Please cite this article as: Hardes J, Henrichs MP, Hauschild G, Nottrott M, Guder W, Streitbuerger A, Silver-coated megaprostheses of the proximal tibia in patients with sarcoma, The Journal of Arthroplasty (2017), doi: 10.1016/j.arth.2017.02.054 This is a PDF file of an unedited manuscript that has been accepted for publication As a service to our customers we are providing this early version of the manuscript The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain ACCEPTED MANUSCRIPT Silver-coated megaprostheses of the proximal tibia in patients RI PT with sarcoma Jendrik Hardes*, Marcel P Henrichs*, Gregor Hauschild, Markus Nottrott, Wiebke Guder, Arne Streitbuerger SC From the Department of Orthopaedics and Tumour Orthopaedics, Münster University M AN U Hospital, Münster, Germany TE D *Both authors contributed equally to this study Address for correspondence: Jendrik Hardes, MD Department of Orthopaedics and Tumour Orthopaedics EP Albert Schweitzer Campus 48149 Münster, Germany AC C E-mail: hardes@uni-muenster.de Tel.: +49-251-8347995 Fax: +49-251-8345398 Silver-coated prostheses ACCEPTED MANUSCRIPT Silver-coated megaprostheses of the proximal tibia in patients with sarcoma Abstract Background: Proximal tibia replacement is associated with high rates of infection This study is the largest one that has compared the infection rates with titanium versus silver-coated megaprostheses in patients treated for sarcomas Methods: The infection rate in 98 patients with sarcoma or giant cell tumour in the proximal tibia who underwent placement of a titanium M AN U SC RI PT (n = 42) or silver-coated (n = 56) megaprosthesis (Mutars®) was 11 assessed, along with the treatments administered for any infection 12 Results: As the primary end point of the study, the rates of infection 13 were 16.7% in the titanium group and 8.9% in the silver group, resulting 14 in 5-year prosthesis survival rates of 90% in the silver and 84% in the 15 titanium group Whereas in the titanium group 37.5% of patients 16 ultimately had to undergo amputation due to infected proximal tibia 17 replacement, these mutilating surgical procedures were necessary in the 18 silver group in only one patient (14.3%) 19 Conclusion: The use of silver-coated prostheses reduced the infection 20 rate in a relatively large and homogeneous group of patients In addition, 21 less aggressive treatment of infection was possible in the group with 22 silver-coated prostheses AC C EP TE D 10 Silver-coated prostheses ACCEPTED MANUSCRIPT 23 Key words: prosthesis-related infections; bone neoplasms; silver; 24 implantation; proximal tibia; sarcoma AC C EP TE D M AN U SC RI PT 25 Silver-coated prostheses ACCEPTED MANUSCRIPT Introduction 27 Periprosthetic infection continues to be a frequent and very serious complication after 28 the placement of tumour endoprostheses.1–4 Particularly with proximal tibia 29 replacements, there is a high risk of infection due to the difficulty of muscle coverage 30 Grimer et al reported an infection rate of 36% before the introduction of routine use 31 of a gastrocnemius rotation flap, with the rate subsequently being reduced to 12% 32 thanks to the better muscle coverage.5 Treatment options for periprosthetic infection 33 range from general rinsing of the prosthesis and exchanging of the polyethylene 34 components, to a need for secondary amputation.2,6 Periprosthetic infection is thus 35 usually associated with a prolonged burden of suffering for the patient, and in the 36 worst case also with a permanent deterioration in quality of life Preventing 37 periprosthetic infection is therefore of the utmost importance M AN U SC RI PT 26 Among the metals known to have antimicrobial activity, silver has attracted interest 39 among many investigators due to its excellent level of antimicrobial activity and low 40 toxicity.7,8 In a previous study, we reported a reduction in the periprosthetic infection 41 rate when silver-coated tumour endoprostheses were used.9 However, the study only 42 included a small number of patients, and the follow-up period was short To the best 43 of our knowledge, there have been no studies investigating the use of silver-coated 44 tumour endoprostheses to prevent infection AC C EP TE D 38 45 The aim of the present study was therefore to compare the actual infection rate 46 among patients with titanium tumour prostheses with the infection rate in a larger 47 number of patients in whom silver-coated prostheses were implanted in the proximal 48 tibia as the primary implant Possible differences in the ways in which periprosthetic 49 infections were treated were also documented Silver-coated prostheses ACCEPTED MANUSCRIPT Patients and Methods 51 A total of 98 patients were treated with a proximal tibia replacement (Mutars®; 52 Implantcast, Buxtehude, Germany) Fifty-six patients (median age 19 years; range 11– 53 78) received a silver-coated prosthesis between 2005 and 2014, and 42 patients 54 (median age 16 years; range 11–69) received a titanium prosthesis between 1996 and 55 2004 Some of the patients were included in a previous study, but now have a longer 56 follow-up period.9 The silver-coated prosthesis used has been described in previous 57 studies.7–9 No silver coating was applied at the articulating surfaces or prosthetic 58 stems M AN U SC RI PT 50 Patients who received a silver-coated replacement had a median follow-up period of 60 38 months (mean 17 months; range 5–120) and 44 months for the surviving patients 61 At the final follow-up, 46 patients had no evidence of disease, four were alive with 62 disease, and five patients had died of the disease (median follow-up period 63 16 months) One patient died due to leukemia 79 months postoperatively Patients 64 with a titanium replacement had a median follow-up period of 128 months (mean 65 111 months; range 3–212) and 146 months for the surviving patients At the final 66 follow-up, 31 patients had no evidence of disease, two were alive with disease, and 67 nine patients had died of the disease (median follow-up period 51 months) AC C EP TE D 59 68 Only patients with bone or soft-tissue tumours with osseous infiltration and stage III 69 giant cell tumours were included Patients who had previously undergone treatment 70 with an intralesional procedure (e.g., curettage, osteosynthesis due to a pathologic 71 fracture) were also included In contrast, patients who had received their current 72 megaprosthesis after a failed megaprosthesis were excluded Patients with an extra- 73 articular knee resection were excluded The clinical charts for the 56 patients treated Silver-coated prostheses ACCEPTED MANUSCRIPT with a silver-coated megaprosthesis were assessed prospectively, with particular 75 attention being given to demographic data, diagnosis, preoperative intralesional 76 procedures, pathologic fracture, preoperative leucocyte count, adjuvant or 77 neoadjuvant therapy, reconstruction length, operating time, muscle flaps, skin 78 grafting, and complications Special emphasis was given to revision operations due to 79 mechanical failures (e.g., change of the bushing, aseptic loosening, periprosthetic 80 fracture) (Table I) SC RI PT 74 Surgery was carried out in both groups in the same operating rooms with laminar 82 air flow; staff did not use body-exhaust suits Postoperatively, all patients received an 83 intravenous third-generation cephalosporin for 3–7 days, followed by oral therapy 84 with a second-generation cephalosporin until wound healing was achieved The 85 proximal tibia replacements were routinely combined with a tube for soft-tissue 86 refixation.10,11 87 An 88 Musculoskeletal Infection Society (MSIS) criteria.12 However, using the MSIS criteria 89 two of our patients would have had no infection in the silver group (number and in 90 table II) Both patients had a clearly elevated CRP, a wound secretion and a redness of 91 the distal wound directly over the prosthesis – but without a clear fistula Therefore, 92 we evaluated this clinical scenario in immunocompromised patients as a 93 periprosthetic infection Table II summarises the data for patients with periprosthetic 94 infection In patients with periprosthetic infection, cure – with no clinical signs of 95 inflammation and negative C-reactive protein findings – was assessed by the treating 96 clinician at the date of the last available follow-up Treatment for periprosthetic 97 infection was documented at the final follow-up TE D M AN U 81 infection was diagnosed in accordance with the AC C EP implant-associated 98 99 100 Statistical analysis The primary end point of the study was periprosthetic infection without any previous revision surgery The secondary end point was the Silver-coated prostheses ACCEPTED MANUSCRIPT outcome of treatment for any periprosthetic infection that occurred Event-free 102 survival of the prosthesis relative to the parameter of infection was assessed using 103 Kaplan–Meier survivorship analysis Possible risk factors for periprosthetic infection 104 were screened using univariate analysis for the whole patient group and for the 105 titanium and silver subgroups Comparison of the infection rates between the silver 106 and titanium groups was done using the chi-square test 107 Results 108 Statistically 109 Univariate analysis of the whole study group (n = 98) identified only superficial 110 wound healing disturbances as a significant risk factor for infection (P = 0.04) 111 Significant influences on the risk of infection were noted only in the silver group, in 112 relation to prior intralesional surgery (P = 0.04) and operating time (P = 0.001) It 113 should be emphasised that revision surgery due to mechanical failures, which 114 occurred in 14.3% of cases in the silver group and 50% of those in the titanium group, 115 did not have a significant influence on the infection rate (Table I) 116 Incidence of periprosthetic infection 117 of 16.7% (seven of 42) was noted when the occurrence of infection was the primary 118 end point In the silver group, the infection rate was only 8.9% (five of 56) The 119 difference was not statistically significant (P = 0.247) (Table III) The 5-year event- 120 free survival rates for the prosthesis relative to the parameter of infection were 90% 121 (95% CI 81,5 to 98,7) in the silver group and 84% (95% CI 73,2 to 96,4) in the 122 titanium group The 10-year survival rate in the titanium group was 81% (95% CI 123 68,2 to 94,2), as only one patient developed an infection after years (Figure I) 124 (P = 0.415) factors for SC risk periprosthetic infection In the titanium group, an infection rate AC C EP TE D M AN U significant RI PT 101 Silver-coated prostheses ACCEPTED MANUSCRIPT Overall, seven of the 56 patients in the silver group (12.5%) developed 126 periprosthetic infections; two patients became infected after revision surgery 127 following mechanical failure of the prosthesis (Table II) In the titanium group, one 128 patient developed a periprosthetic infection after revision surgery (which was required 129 in 50% of patients) due to a mechanical failure of the prosthesis, resulting in an 130 overall infection rate of 19.0% (eight of 42) RI PT 125 None of the patients who died of disease during a median of 16 months 132 postoperatively had periprosthetic infections in the silver group, whereas two patients 133 (patients 12 and 13) with a titanium prosthesis developed periprosthetic infections in 134 the first two postoperative years and ultimately died due to the disease 135 Time to periprosthetic infection 136 led to a surgical intervention at a median of 18 months (range 1–71) after implantation 137 of the prosthesis, without any previous revision surgery In the silver group, 138 periprosthetic infection occurred at a median of months (range 3.5–19) after primary 139 implantation of the prosthesis in five patients without any previous revision surgery 140 (Table II) In two patients with periprosthetic infection after mechanical revision 141 surgery, infection developed months and 66 months postoperatively, respectively In 142 the titanium group, one patient developed infection after revision surgery months 143 postoperatively Overall, nine of 12 (75%) periprosthetic infections in the two groups 144 occurred within the first two postoperative years, if no later revision surgery due to 145 mechanical failure was necessary 146 Treatment of periprosthetic infection Whereas three of the eight patients in 147 the titanium group (37.5%) ultimately had to undergo amputation due to an infected 148 proximal tibia replacement, this type of mutilating surgical procedure was only M AN U SC 131 AC C EP TE D In the titanium group, periprosthetic infection Silver-coated prostheses ACCEPTED MANUSCRIPT necessary in one patient in the silver group (14.3%) (Table III) It must be borne in 150 mind here, however, that two of the three amputations in the titanium group were 151 carried out in patients who had previously undergone radiotherapy, resulting in poor 152 soft-tissue conditions Two-stage revision surgery with a temporary antibiotic- 153 impregnated cement spacer was ultimately successful in four of eight patients in the 154 titanium group (50.0%), but this was necessary in only one patient (14.3%) in the 155 silver group (Table III) In the silver group, by contrast, antibiotic treatment alone 156 (n = 1, 14.3%, follow-up 72 months) and minor revision surgery (n = 2, 28.6%, 157 follow-up and 82 months, respectively) with one-stage exchange of the prosthetic 158 body without removal of the stems were ultimately successful 159 Discussion 160 Periprosthetic infection of megaprostheses continues to be a common and serious 161 complication in orthopaedic oncology.2,3,13 In a systematic review of the literature 162 including a total of 4838 patients with a tumour endoprosthetic reconstruction of the 163 lower extremity, Racano et al.13 reported periprosthetic infection in 10% of cases 164 However, the authors not provide any data on the infection rate relative to the site 165 of endoprosthetic replacement However, it is well known that patients with proximal 166 tibia replacements in particular are at high risk for periprosthetic infection, with rates 167 ranging from 14% to 36%.5,14,15 AC C EP TE D M AN U SC RI PT 149 168 The high infection rates reported with the use of tumour endoprostheses in the area 169 of the proximal tibia, emphasise the importance of reducing the infection rate even 170 further Silver has been used as an antimicrobial agent for centuries.16 It has been used 171 successfully in the topical treatment of burns and chronic wounds and as a coating for 172 medical devices.17 However, negative results with silver-coated devices have also Silver-coated prostheses 10 ACCEPTED MANUSCRIPT comparison with 17.6% of the patients treated with a titanium prosthesis (P = 0.062) 198 Specifically for patients with a proximal tibia replacement, the overall infection rate – 199 infection of the primary implant, or infection after revision surgery of the primary 200 implant due to mechanical failure – was reduced from 17.1% in the titanium group to 201 6.9% in the silver group (P = 0.289) However, both due to the short follow-up period 202 and also the small numbers of patients, the results only provide initial evidence of the 203 potential effectiveness of silver coating SC RI PT 197 Within a period of years, our sarcoma centre has succeeded both in substantially 205 increasing the number of patients receiving silver-coated proximal tibia replacements 206 for a rare tumour location and also in achieving a much longer follow-up Although 207 the overall infection rate increased to 12.5% in the silver group in comparison with 208 our 2010 report,9 it is markedly below the value of 19% in the titanium group In 209 comparison, Puchner et al.15 noted an overall infection rate of 24.7% with uncoated 210 proximal tibia replacements With regard to the infection rate in the primary implant 211 alone (without including revision operations), silver coating can lead almost to a 212 halving of the figure, at 8.9% versus 16.7% in the titanium group By comparison, 213 Puchner et al.15 noted an infection rate in primary implants of 12.3% with implants 214 that were not silver-coated Wafa et al.22 reported on 18 patients with silver-coated 215 proximal tibia replacements However, only nine of the patients underwent primary 216 implantation, while the remainder received silver-coated prostheses during one-stage 217 or two-stage exchanges Infection occurred in one of nine patients (11.1%) AC C EP TE D M AN U 204 218 Although it was not possible to demonstrate a statistically significant reduction in 219 the periprosthetic infection rate, the present study confirms our initial results from 220 2010 indicating that silver-coated tumour prostheses are apparently able to reduce the Silver-coated prostheses 11 ACCEPTED MANUSCRIPT infection rate But why has the decline in the infection rate not been clearer? In our 222 view, this may be explained by the fact that active free silver ions bind to proteins and 223 become inactivated.23 This means that the surgeon has to avoid haematoma and poor 224 muscle coverage of the prosthesis, resulting in superficial wound healing problems, 225 which can result in bacterial colonisation In these areas, the silver coating is unable to 226 develop an adjuvant effect The silver coating can inhibit bacterial colonisation of the 227 prosthetic body, with subsequent periprosthetic infection, and can kill bacteria near 228 the prosthesis In our view, the development of periprosthetic infection cannot 229 normally be attributed to bacterial resistance to the silver – although this is in 230 principle possible with Gram-negative bacteria16 – but instead results from 231 inactivation of the free silver ions due to binding to proteins M AN U SC RI PT 221 Revision surgery for a tumour endoprosthesis may also be associated with a not 233 inconsiderable risk of periprosthetic infection,4,15 with bacterial contamination taking 234 place during the revision procedure Puchner et al.15 reported that ten out of 81 235 patients (12.3%) developed periprosthetic infection with proximal tibia replacements 236 after revision surgery performed for other reasons Overall, 10 of 20 infections (50%) 237 occurred after revision surgery In the present study, previous revision surgery might 238 have been responsible for periprosthetic infection in only three patients (two in the 239 silver group and one in the titanium group) among 15 patients with a periprosthetic 240 infection (20%) A significant risk factor is not evident in the present study, but it 241 cannot of course be excluded in the future AC C EP TE D 232 242 75% of periprosthetic infections in both groups in the present study occurred within 243 the first two postoperative years However, it is possible that further periprosthetic 244 infections may become clinically apparent in the silver group later, as the silver helps Silver-coated prostheses 12 ACCEPTED MANUSCRIPT to control infection but may dissociate from the prosthetic surface over time In the 246 case of periprosthetic infection, silver ions will go into solution, since the surrounding 247 tissue has a negative pH value If the infection cannot be controlled, the silver ions 248 can in principle degrade completely away from the prosthetic surface, and the 249 infection will become clinically apparent In clinical reality, however, periprosthetic 250 infection in the titanium group led to surgical interventions a median of 18 months 251 after implantation of the prosthesis without any previous revision surgery In contrast, 252 periprosthetic infection in the silver group occurred earlier after primary implantation 253 of the prosthesis We would therefore assume that in spite of the shorter follow-up 254 period in the silver group, no further substantial change can be expected in the group 255 – unless further revision operations become necessary due to mechanical 256 complications M AN U SC RI PT 245 In addition to the reduction in the rate of periprosthetic infection as the primary end 258 point of the study, attention also needs to be given as a secondary end point to the 259 implications of infections that occur nevertheless Despite the small group of patients, 260 at least a trend can be noted towards less invasive treatment measures in the silver 261 group in the presence of periprosthetic infection – such as debridement, antibiotics, 262 irrigation and retention (DAIR) and one-stage revision In the study mentioned above, 263 Wafa et al.22 also showed that in cases of periprosthetic infection of uncoated tumour 264 prostheses, one-stage exchanges of the prosthesis for a silver-coated one were only 265 followed by reinfection in two of 39 patients (5.1%) By contrast, repeat infections 266 occurred in five of 40 patients (12.5%) with one-stage exchanges of the prosthesis for 267 an uncoated one There are some studies in the literature that report a high percentage 268 of secondary amputations due to periprosthetic infection, in 19–46% of cases, with the 269 highest rate in proximal tibia replacements.2,6,14 Whereas in the titanium group 37.5% AC C EP TE D 257 Silver-coated prostheses 13 ACCEPTED MANUSCRIPT 270 of patients ultimately had to undergo amputation in the present study, these mutilating 271 surgical procedures were only necessary in the silver group in one patient (14.3%) The most important limitation of the present study is the retrospective nature of the 273 titanium group, whereas the data for the silver group were collected prospectively 274 The patient numbers are still small, of course, but in our opinion they are large enough 275 for such a rare procedure The advantage of this study lies in the fact that an implant 276 system was used that has been employed over many years, always in the same 277 surgical setting, and only by a few experienced surgeons 278 Conclusions M AN U SC RI PT 272 Silver represents a reasonable addition to the armamentarium for the treatment of 280 periprosthetic infection However, it would be unrealistic to assume that periprosthetic 281 infection can be completely avoided, since active free silver ions can only exert their 282 effects in close proximity to the prosthetic surface In cases of periprosthetic infection 283 despite the use of silver-coated prostheses, minor revision surgery or even adequate 284 antibiotic therapy alone may be successful In the future, studies with larger numbers 285 of patients and longer follow-up periods are warranted in order to confirm these 286 results EP AC C 287 TE D 279 Silver-coated prostheses 14 ACCEPTED MANUSCRIPT 288 References 289 in 250 patients with sarcoma Clin Orthop Relat Res 2006;450:164–171 Hardes J, Gebert C, Schwappach A, et al Characteristics and outcome RI PT 290 291 Gosheger G, Gebert C, Ahrens H, et al Endoprosthetic reconstruction 292 of infections associated with tumor endoprostheses Arch Orthop Trauma 293 Surg 2006;126:289–296 Jeys LM, Grimer RJ, Carter SR, et al Periprosthetic infection in SC 294 patients treated for an orthopaedic oncological condition J Bone Joint 296 Surg Am 2005;87:842–849 297 M AN U 295 Jeys LM, Kulkarni A, Grimer RJ, et al Endoprosthetic replacement for the treatment of musculoskeletal tumors of the appendicular skeleton and 299 pelvis J Bone Joint Surg Am 2008;90:1265–1271 Jeys LM, Grimer RJ, Carter SR, et al Risk of amputation following limb salvage surgery with endoprosthetic replacement in a consecutive series 303 of 1261 patients Int Orthop 2003;27:160–163 304 305 EP of the proximal tibia J Bone Joint Surg Br 1999;81:488–494 301 302 Grimer RJ, Carter SR, Tillman RM, et al Endoprosthetic replacement AC C 300 TE D 298 Gosheger G, Hardes J, Ahrens H, et al Silver-coated 306 megaendoprostheses in a rabbit model – an analysis of the infection rate 307 and toxicological side effects Biomaterials 2004;25:5547–5556 Silver-coated prostheses 15 ACCEPTED MANUSCRIPT 308 Hardes J, Ahrens H, Gebert C, et al Lack of toxicological side-effects in 309 silver-coated megaprostheses in humans Biomaterials 2007;28:2869– 310 2875 Hardes J, von Eiff C, Streitbuerger A, et al Reduction of periprosthetic RI PT 311 312 infection with silver-coated megaprostheses in patients with bone 313 sarcoma J Surg Oncol 2010;101:389–395 315 megaprostheses 316 2001;393:264–71 using a Trevira SC 10 Gosheger G, Hillmann A, Lindner N, et al Soft tissue reconstruction of tube Clin Orthop Relat Res M AN U 314 11 Hardes J, Ahrens H, Nottrott M, et al Attachment tube for soft tissue 318 reconstruction after implantation of a mega-endoprosthesis Oper Orthop 319 Traumatol 2012;24:227–234 320 12 Parvizi J, TE D 317 Gehrke T, International Consensus Group on Periprosthetic Joint Infection Defintion of periprosthetic joint infection 322 J Arthroplasty 2014;29:1331 324 325 326 13 Racano A, Pazionis T, Farrokhyar F, et al High infection rate AC C 323 EP 321 outcomes in long-bone tumor surgery with endoprosthetic reconstruction in adults: a systematic review Clin Orthop Relat Res 2013;471:2017– 2027 327 14 Myers GJ, Abudu AT, Carter SR, et al The long-term results of 328 endoprosthetic replacement of the proximal tibia for bone tumours J 329 Bone Joint Surg Br 2007;89:1632–1637 Silver-coated prostheses 16 ACCEPTED MANUSCRIPT 330 15 Puchner SE, Kutscha-Lissberg P, Kaider A, et al Outcome after 331 reconstruction of the proximal tibia – complications and competing risk 332 analysis PLoS One 2015;10:e0135736 16 Randall CP, Gupta A, Jackson N, et al Silver resistance in Grambacteria: a dissection of RI PT 333 334 negative endogenous and 335 mechanisms J Antimicrob Chemother 2015;70:1037–1046 exogenous 17 Politano AD, Campbell KT, Rosenberger LH, et al Use of silver in the 337 prevention and treatment of infections: silver review Surg Infect 338 2013;14:8–20 M AN U SC 336 18 Dahlberg PJ, Agger WA, Singer JR, et al Subclavian hemodialysis 340 catheter infections: a prospective, randomized trial of an attachable 341 silver-impregnated cuff for prevention of catheter-related infections Infect 342 Control Hosp Epidemiol 1995;16:506–511 TE D 339 19 Riley DK, Classen DC, Stevens LE, et al A large randomized clinical 344 trial of a silver-impregnated urinary catheter: lack of efficacy and 345 staphylococcal superinfection Am J Med 1995;98:349–356 AC C EP 343 346 20 Tokmaji G, Vermeulen H, Müller MC, et al Silver-coated endotracheal 347 tubes for prevention of ventilator-associated pneumonia in critically ill 348 349 350 patients Cochrane Database Syst Rev 2015;12;8:CD009201 21 Glehr M, Leithner A, Friesenbichler J, et al Argyria following the use of silver-coated megaprostheses Bone Joint J 2013;95:988–992 Silver-coated prostheses 17 ACCEPTED MANUSCRIPT 351 22 Wafa H, Grimer RJ, Reddy K, et al Retrospective evaluation of the incidence 353 endoprostheses in high-risk patients Bone Joint J 2015;97:252–257 354 23 Schierholz JM, Lucas LJ, Rump A, et al Efficacy of silver-coated 355 of early periprosthetic infection medical devices J Hosp Infect 1998;40:257–262 AC C EP TE D M AN U SC 356 with silver-treated RI PT 352 Silver-coated prostheses 18 ACCEPTED MANUSCRIPT AC C EP TE D M AN U SC RI PT 357 ACCEPTED MANUSCRIPT Data for 98 patients who received proximal tibia replacements Leucocyte count (median) Previous intralesional Titanium Silver (n = 42) (n = 56) 3.2 4.6 9.5% 16.1% operations (%) Osteosarcoma 78.5% Osteosarcoma 55.4% SC Diagnoses RI PT Table I Ewing sarcoma 19.6% Chondrosarcoma 4.7% GCT 8.9% M AN U Ewing sarcoma 11.9% Chemotherapy Radiotherapy TE D Median reconstruction length (mm) Median operating time (min.) MFH 7.1% 90.5% 80.0% 7.1% 3.6% 145 140 245 220 11.9% 10.7% 7.1% 7.1% 50.0% 14.3% Superficial EP Wound healing problems AC C Deep / hematoma Revision operations (mechanical failure) GCT, giant cell tumor; MFH, malignant fibrous histiocytoma ACCEPTED MANUSCRIPT Characteristics of the 15 patients with periprosthetic infection RI PT Table II Interval Interval to between Any revision primary surgery due to revision SC infection after surgery and Silver operation Patient (y) coating (months) before infection 74 Yes No – 38 Yes No – 16 Yes 3.5 No – 60 Yes Yes Change of the surgery TE D 77 mechanical failure Type of revision M AN U Age C-reactive infection (months) Isolated Fistula microorganism protein (mg/dL) – No No 5.1 – No MRSA 20.2 – No No 12.4 66 No P aeruginosa 15.6 Yes S caprae 0.5 – No S caprae, 28.7 femur stem due to 63 18 Yes Yes 69 19 Yes AC C EP periprosthetic No fracture Change of the femur stem due to aseptic loosening – Streptococcus agalactiae ACCEPTED MANUSCRIPT between Interval to infection after Any revision primary surgery due to revision surgery and Silver operation mechanical failure Type of revision Patient (y) coating (months) before infection 67 Yes No – 16 No No – 15 No 47 No – 10 10 No No – No 71 No 12 16 No 18 No 13 21 No No 14 60 No 18 No 99 No AC C 14 Yes M AN U TE D 14 EP 11 surgery C-reactive infection SC Age 15 RI PT Interval (months) Isolated Fistula microorganism protein (mg/dL) – No S epidermidis 8.5 – Yes No 25.7 – No S pyogenes 36.5 – No S epidermidis, 4.8 MRSA (at second infection) – – No S aureus 36.8 – – No S epidermidis 1.5 – – No S haemolyticus 0.5 – – No S epidermidis, E 14.7 faecalis Change of the bushing MRSA, methicillin-resistant Staphylococcus aureus; P., Pseudomonas; S., Staphylococcus No S lugdunensis 17.0 ACCEPTED MANUSCRIPT Infection rates and ultimately successful treatments for infection Patients with infection as primary Silver Titanium (n = 56) (n = 42) end point (n) Infection rate in patients with no 8.9 16.7 Secondary infection rate (%) Antibiotic treatment M AN U whole study period (n) SC revision surgery (%) Patients with infection over Total 12 RI PT Table III 12.5 19.0 11.2 15 15.3 (14.3%) (7.1%) (28.6%) (13.3%) (12.5%) (6.7%) Two-stage exchange of whole prosthesis (14.3%) (50%) (33.3%) Explantation of prosthesis and (28.6%) (13.3%) (14.3%) (37.5%) (26.7%) One-stage prosthesis exchange TE D without stem removal Two-stage prosthesis exchange EP without stem removal AC C permanent spacer Amputation ACCEPTED MANUSCRIPT Kaplan–Meier survival curve, showing infection-free survival of the primary implant in AC C EP TE D M AN U SC RI PT the silver and titanium group AC C EP TE D M AN U SC RI PT ACCEPTED MANUSCRIPT