Research Article

Use of a Silver Nylon DressingFollowing Total Hip and KneeArthroplasty Decreases thePostoperative Infection Rate

AbstractIntroduction: Prosthetic joint infection (PJI) is a potentiallycatastrophic complication of total joint replacement. Our purposewas to determine whether the use of a silver dressing reduces theincidence of superficial and deep PJI following primary total hipreplacement and total knee replacement.Methods: A case-control study of primary total hip replacementand total knee replacement was performed to compare theincidence of superficial and deep PJI in patients who received asilver nylon dressing with patients who received a standarddressing.Results: The incidence of infection was significantly lower in thestudygroup comparedwith the control group. Therewere nodeepPJIs in the silver dressing group. Twelve patients (2.3%) in thecontrol group developed PJI.Discussion: The use of a silver dressing significantly reduced theincidence of superficial and deep PJI following total jointreplacement (P = 0.010). Given the financial impact of PJI, theapplication of silver dressings may result in considerable costsavings, and a formal cost-benefit analysis could be investigated.

Prosthetic joint infection (PJI) is apotentially catastrophic compli-cation of total hip replacement(THR) and total knee replacement(TKR) surgeries that is associatedwith significant morbidity and mor-tality.1 Infection is the most commoncause of failure of TKR surgery andthe third most common cause offailure following THR surgery.2,3

The treatment of PJI often ne-cessitates prolonged antimicrobialregimens and repeat surgical proce-dures that may include temporaryor permanent removal of the

implant.4,5 Despite contemporarytreatment protocols, the incidence ofPJI continues to rise. In the setting ofmultidrug-resistant organisms andimmunocompromised hosts, theseinfections may result in loss of limbor life resulting from sepsis.Because the number of patients

having arthritis of the hip and kneecontinues to rise, the numbers ofjoint replacements are projected toincrease 673%and 174%by the year2030 for primary TKR and THR,respectively.6 This would amount toapproximately 4 million total joint

Ashley J. Tisosky, MD

Otatade Iyoha-Bello, BSc

Nicholas Demosthenes, BA

Giovanni Quimbayo, MD

Tara Coreanu, BSc

Ayesha Abdeen, MD, FRCSC

From the Department of OrthopaedicSurgery, Harvard CombinedOrthopaedic Residency Program,Massachusetts General Hospital,Boston, MA (Dr. Tisosky); theDepartment of Orthopaedic Surgery(Ms. Iyoha-Bello and Dr. Quimbayo)and the Department of Health CareQuality (Mr. Demosthenes), BethIsrael Deaconess Medical Center,Boston; the Faculty of HealthSciences, Clinical Research Center,Soroka Universtity Medical Center,Ben-Gurion University of the Negev,Beer-Sheba, Israel (Ms. Coreanu);and the Department of OrthopaedicSurgery, Harvard Medical School,Beth Israel Deaconess MedicalCenter, Boston (Dr. Abdeen).

JAAOS Glob Res Rev 2017;1:e034

DOI: 10.5435/JAAOSGlobal-D-17-00034

Copyright © 2017 The Author(s).Published by Wolters Kluwer Health,Inc. on behalf of the AmericanAcademy of Orthopaedic Surgeons.This is an open-access articledistributed under the terms of theCreative Commons Attribution-NonCommercial-No Derivatives License4.0 (CCBY-NC-ND), where it ispermissible to download and sharethe work provided it is properly cited.The work cannot be changed in anyway or used commercially withoutpermission from the journal.

richfosterText BoxPlease note that this study may involve findings that exceed the claims currently cleared by the FDA for the product. The various cleared uses for Silverlon® Dressings can be found in the “Cleared Indications Sheet” found in this information packet Full prescribing information can be found at:- www.silverlon.com/indications or call 1 888-551-0188

replacements (TJRs) performed eachyear. Although the rate of infectionreported following TJR is low(0.58%–2.4%), the increase in thevolume of TJR procedures has re-sulted in 5,310 to 21,240 infectionsper year.7 As the magnitude ofpatient suffering increases, so doesthe economic burden associatedwith prolonged antibiotic treat-ments, repeat surgical procedures,and extended admissions to hospitaland postacute care facilities, as wellas lost wages because of prolongedpatient disability.8 The financialimpact of PJI in the United States was$566 million in 2009 and is pro-jected to increase to $1.62 billion in2020.9 Strategies to reduce the inci-dence of deep PJI are imperative.The antimicrobial properties of

silver ion have been known for cen-turies. Historically, topical silverwas used in the 1800s in the treat-ment of infected traumatic battlewounds, burns, venereal diseases,and leg ulcers.10 The recent increasein infections resulting from resis-tant microorganisms has led to thedevelopment of contemporary silverion–containing dressings designed totreat and prevent infections. Silver-impregnated dressings have beenshown to be safe and effective inreducing the incidence of surgical siteinfections (SSIs) following colorectal,cardiothoracic, lumbar spine, andfoot and ankle surgeries.11–14 Aretrospective study has also shownsilver-impregnated dressings toreduce infection risk followingcesarean sections.15 Others haveshown silver dressings to be associ-ated with improved wound analgesiaand incisional cosmesis.16 A retro-spective review of patients whounderwent TJR with the use of asilver dressing (Aquacel Ag withHydrofiber; Convatec) demonstrated

a significant reduction in the inci-dence of acute PJI compared with astandard nonsilver dressing.17

To our knowledge, the performanceand efficacy of a silver nylon dressing(Silverlon) as it pertains specifically tothe incidence of PJI following primaryTHRandTKRsurgeries havenotbeenevaluated in a nonindustry–supportedstudy. We sought to determinewhether the use of a silver nylondressing reduces the incidence of bothsuperficial infection and deep PJIwithin 1 year following primary THRand TKR surgeries and whether thisdressing can be used safely in thispatient population.

Methods

A case-control study was conductedfor quality assurance purposes toevaluate the efficacy of a silver nylondressing (Silverlon IslandDressing) inpatients who presented to our insti-tution for primary THR and TKRsurgeries. The study group was fol-lowed prospectively and comparedwith historic controls. Permission toperform this quality assurance studywas granted by the institutionalreview board.Inclusion criteria included all male

and female patients aged 18 years andolder who were consecutively sched-uled to undergo primary TKR andTHR surgeries. Patients were treatedby two surgeons at a single institution.Exclusion criteria included revisionTHR and TKR, bilateral jointreplacement, and prior or active infec-tion of the operative extremity. Patientcharacteristics were collected, includ-ing age at surgery, sex, tobacco use,body mass index, and immunocom-promised status. Patients were consid-ered to be immunocompromisedwhenthey had uncontrolled diabetes, pres-

enceofHIV, liver disease, active cancerwith ongoing chemotherapy, and/orlatent tuberculosis, or were currentlyusing coritcosteroids or methotrexate.Current steroid use, diabetes, presenceof HIV, and liver disease were alsoevaluated as independent risk factorsfor infection. Patients’ surgical riskwas measured using the AmericanSociety of Anesthesiologists (ASA)physical status classification (ASA 1 or2 as low risk and ASA $3 as highrisk).18

This study consisted of two groups.The study group received a Silverlondressing at the time of surgery andconsisted of all patients who had pri-mary THR and TKR between 2013and 2014 after both surgeons startedto use silver-impregnated dressingsfor all TJR cases. The Silverlondressing was placed on the incision atthe time of surgery and removed onthe seventh postoperative day, asrecommended by the manufacturer.Dressings that were saturated prior tothis time frame were replaced with asecond Silverlon dressing that wasalso removed on the seventh post-operative day, as initially scheduled.The control group consisted of allpatients who had primary THR andTKR performed by the same twosurgeons from 2011 to 2013, prior tothe time the surgeons started to usethe silver-impregnated dressing. Thepatients in the control group receivedwhat will be referred to as a “stan-dard” dressing consisting of a non-adherent layer (Xeroform) and drygauze. The standard dressing waschanged to a second standard dress-ing on postoperative day 2 andthen changed daily until the woundremained dry.All surgical incisions were treated

with a primary closure using staplesin the skin and a sterile dressingplaced at the time of surgery. The

None of the following authors or any immediate family member has received anything of value from or has stock or stock options held ina commercial company or institution related directly or indirectly to the subject of this chapter: Dr. Tisosky, Ms. Iyoha-Bello,Mr. Demosthenes, Dr. Quimbayo, Ms. Coreanu, and Dr. Abdeen.

Use of a Silver Nylon Dressing

2 Journal of the American Academy of Orthopaedic Surgeons

postoperative protocol was consis-tent between both surgeons andcompliant with an institutional stan-dardized protocol for all patientsundergoing TJR. All patients receivedperioperative antibiotics, specificallycefazolin 2 g intravenous (IV) Q8H fora 24-hour period. Patients with a bodyweight greater than 100 kg were givencefazolin 3 g IV Q8H for a 24-hourperiod. Patients who had a penicillinallergy or a positive methicillin–resistant Staphylococcus aureus pre-operative nasal swab screening wereadministered vancomycin 15mg/kg IVQ12H for a 24-hour period. Venousthromboembolism (VTE) prophylaxiswas accomplished with enoxaparinsubcutaneously for 4 weeks post-operatively in all patients. Patientsreceived a dose of tranexamic acid10 mg/kg preoperatively unlesscontraindications were present.Contraindications for tranexamicacid administration at our institu-tion are as follows: a glomerularfiltration rate ,60 mL/min, knowncolorblindness, history of deepvein thrombosis or pulmonaryembolism, known thrombophilia(including, but not limited to, factorV Leiden and protein S or C defi-ciency), and anesthesiologist concernregarding specific coagulation issues,commonly including recent stroke,myocardial infarction, or multipleintracoronary stents. All patientsreceived low-molecular-weight hep-arin for prevention of VTE. Iflow-molecular-weight heparin wascontraindicated, VTE prophylaxiswas in the form of warfarin. Allpatients participated in physicaltherapy starting on postoperativeday 1 and continued a standardizedregimen postoperatively. Allpatients had staples removed 2weeks postoperatively.Study patients were followed pro-

spectively to determine the onset ofsuperficial infection (includingwoundinfection, stitch abscess, and cellulitis)and deep PJI. These data were ex-

tracted retrospectively for the controlgroup based on chart review. Datawere collected until the 12-monthannual follow-up visit for all studyand control subjects. Diagnosis fordeep PJI was based on the criteriadefined by the workgroup convenedby the Musculoskeletal InfectionSociety.19 Patients with deep PJI werefurther subcategorized into acutedeep PJI, defined as ,3 weeksbetween symptom onset and treat-ment, and chronic PJI, defined as .3weeks between symptom onset andtreatment.4 The diagnosis of superfi-cial infection was made by the treat-ing surgeon and included superficialincisional infections (wound drain-age, erythema, or dehiscence), stitchabscesses, and cellulitis (acute ery-thema extending beyond the limits ofthe surgical wound on the operativeextremity). The decision to aspiratethe knee to rule out the presence ofdeep PJI in the setting of a woundcomplication or cellulitis was left tothe discretion of the treating surgeonand was based on the presence ofjoint pain with range of motion,active drainage, fever, and/or elevatedserum inflammatory markers.

Statistical AnalysisThe primary outcomemeasure of thisstudy included the combined rate ofpostsurgical superficial and deepinfections that occurred within 1 yearfollowing surgery. Secondary out-comes included the individual inci-dence of deep and superficialinfections. Subgroup prespecifiedanalysis included analysis by immu-nocompetence status. Continuousvariables with normal distributionare presented as mean (6SD) andcompared using the Student t-test.Continuous variables with non-normal distributions are presented asmedians and interquartile ranges andcompared using the Mann-Whitneytest. Dichotomous or nominal cate-goric variables are compared with

the use of the chi-square test withnormal approximation or Fisherexact test, as appropriate.We used a generalized estimating

equation (GEE) approach for theunivariate and multivariate analysesof the outcomes to account for theindividual surgeon clustering. Weused an interchangeable correlationmatrix and logistic regression as alink function. The variables wereintroduced to the model based on theclinical and statistical significance(P , 0.1 in univariate analysis). Thefinal model included the followingvariables: patient age at the time ofsurgery, ASA class (ASA 1 or 2 aslow risk and ASA $3 as high risk),immunocompromised status, andthe use of a silver dressing.In a subgroup analysis among

those who had any infection, super-ficial infection and deep infectionwere stratified by the immunocom-petence status. A univariate logisticregression model using GEEs toaccount for the individual surgeonclustering was used to calculate oddsratios (ORs) and 95% confidenceintervals (CIs).Statistical analysis was performed

using IBM SPSS Statistics softwarepackage (version 20.0; IBM). Two-tailed values of P , 0.05 were con-sidered statistically significant.

Results

The study group consisted of 309consecutive TJR cases performed in309 patients (219 TKRs and 90THRs) from November 1, 2013, toNovember 30, 2014. The controlgroup consisted of 525 consecutiveTJRs performed in 525 patients (382TKRs and 143 THRs) who under-went surgery from November 1,2011, to November 30, 2013. Thestudy and control patients were com-pared with respect to demographicsand risk factors for infection (Table 1).There were no patients who were

Ashley J. Tisosky, MD, et al

October 2017, Vol 1, No 7

lost to follow-up at the final follow-up interval of 12 months in eitherthe control group or the study group.The incidence of any infection was

significantly lower in the silverdressing group in comparison withthe control group (Table 2). Therewere no deep PJIs in the grouptreated with the silver nylon dress-ing. There were 12 deep PJIs (2.3%)in the control group, including 11TKRs and 1 THR. The intervalfrom the date of surgery until theonset of infection was a median of58 days (range, 11–147 days). Mostdeep infections were acute infec-tions that presented within 3 weeksof the onset of symptoms4 (Figure

1). All cases of deep PJI were treatedsurgically and received parenteralantibiotic treatment: eight requiredirrigation and débridement withliner exchange and four requiredtwo-stage revision. The organismsisolated included the following:methicillin-susceptible S aureus,Corynebacterium, Streptococcusviridans,Mycobacterium tuberculosis,Aggregatibacter, Pseudomonasaeruginosa, Peptostreptococcus, andmixed flora (Proteus mirabilis andEnterobacter cloacae). No organ-isms were identified in four cases.There were a total of 46 superficial

infections, including 12 in the studygroup (3.9%: 10 TKRs and 2 THRs)

and 32 in the control group (6.1%:28 TKRs and 4 THRs). The types ofsuperficial infection are displayed inFigure 2. In all cases of superficialinfection, the clinical signs of infec-tion resolved following a course oforal antibiotic treatment. At finalfollow-up, none of the superficialinfections recurred or progressed todeep PJI.TheORs for the tested variables are

summarized in Table 3. The odds ofany infection were significantly lessin the silver dressing group com-pared with the control group (OR,0.382; 95% CI, 0.250–0.583; P ,0.0001), as were the odds of super-ficial infection (OR, 0.546; 95% CI,0.343–0.870; P = 0.011). The oddsof any infection, as well as superficialinfection, were lower in the silverdressing group in both immuno-competent and immunocompro-mised patients.The results of a GEE logistic

regression analysis are displayed inTable 4. The odds of any infectionwere significantly less with the useof a silver dressing, independent ofimmunocompromised status highASA score, and age at surgery (OR,0.396; 95% CI, 0.273–0.573;P , 0.001).

Table 1

Baseline Characteristics of the Patients

Population Characteristics No Silver: 62.9% (n = 525) With Silver: 37.1% (n = 309) P

Age at surgery, yrs, mean 6 SD 63.406 10.98 62.53 6 11.57 0.280Sex 0.136Female, % (n) 62.9 (330) 68 (210) —Male, % (n) 37.1 (195) 32 (99) —

ASA high score (3–5), % (n) 51 (268) 46.3 (143) 0.183Tobacco use, % (n) 13.9 (73) 21.7 (67) 0.004Body mass index, median (interquartile range) 31 (27–36) 31 (27–36.5) 0.814Immunocompromised, % (n) 41.9 (220) 31.7 (98) 0.003Steroids, % (n) 3.4 (18) 2.6 (8) 0.500Diabetes, % (n) 27 (142) 20.4 (63) 0.031HIV, % (n) 2.7 (14) 1 (3) 0.094Liver disease, % (n) 7.2 (38) 6.5 (20) 0.675

Table 2

Clinical Outcomes

OutcomeNo Silver: 62.9%

(n = 525)With Silver: 37.1%

(n = 309) P

Infection, % (n) 8.4 (44) 3.9 (12) 0.012Infection type 0.010Deep infection, % (n) 2.3 (12) 0 (0) —Superficial infection, % (n) 6.1 (32) 3.9 (12) —

Treatment 0.032Medical alone, % (n) 5.9 (31) 3.2 (10) —Surgery, % (n) 2.5 (13) 0.6 (2) —

Expired patients, % (n) 2.5 (13) 1 (3) 0.126

Use of a Silver Nylon Dressing

4 Journal of the American Academy of Orthopaedic Surgeons

Discussion

Infection following THR and TKRsurgeries is a devastating complica-tion with tremendous implicationspertaining to patient morbidity andmortality. This is further com-pounded by the astronomical healthcare costs associated with the treat-ment of PJIs. As the number of jointreplacements performed in theUnited States continues to rise, so dothe numbers of patients affected byinfection. The downstream effects ofthese infections on patients’ livesinclude poor functional outcome anddisability, patient dissatisfaction, lostwages because of work absence, andgeneral depreciation in health-related quality of life. To circum-vent the adverse effects of PJI,prevention is paramount.Silver nylon dressings have been

found to reduce SSI in a number ofsurgical procedures, including car-diothoracic surgery, colorectal sur-gery, cesarean sections, spinalsurgery, and recently in the fieldof arthroplasty.11–16 These priorreports have shown that silver iondressings are safe to be used in thesurgical setting and are furthermoreeffective in reducing SSI. In a priorretrospective study, the use of adressing that combines silver withhydrofiber layer that gels on contactwith the wound (Acquacel Hydro-fiber) was found to reduce the inci-dence of acute PJI following hip andknee arthroplasty.17 We sought todetermine whether a silver nylondressing can reduce SSI followingTHR and TKR. The Silverlondressing differs from the Acquaceldressing in silver content (546 mg/100 cm2 versus 8.3 mg/100/cm2,respectively) and absence of thehydrofiber gel layer. The silver ionis in direct contact with the skin inthe Silverlon dressing, unlike othercontemporary silver-containing dress-ings, and thus we sought to determine

whether this dressing with more than5-fold increased silver content, anddirect silver contact with the skin,could be used safely in this population.Furthermore, at our institution, theSilverlon dressing was less expensivethan Acquacel ($30 per dressing ver-sus $39 of the similar size).Our study found that the use of a

silver nylon dressing significantlyreduced the incidence of both super-ficial infection and deep PJI followingTHR and TKR in all subjects. Theodds of any infection were all signif-icantly lower in the silver nylondressing group. The strengths of thisstudy include its prospective nature,

use of a control group, and follow-upduration of 1 year. We did not iden-tify any adverse reaction to the silverdressing in our population. Thisarticle is the first to our knowledge toevaluate the effect of a silver nylondressing on SSI following TJR sur-gery of the hip and knee in anonindustry–supported study. Fur-thermore, this is the first study tospecifically investigate the use of theSilverlon dressing in this patientpopulation.Superficial wound infections have

been demonstrated to be a risk factorfor deep prosthetic infection.20,21 Itstands to reason that topical agents

Figure 2

Superficial infection type.

Figure 1

Deep infection type.

Ashley J. Tisosky, MD, et al

October 2017, Vol 1, No 7

aimed at reducing superficial infec-tions may in turn prove to success-fully decrease the incidence of PJI. Inthis study, the use of a silver nylondressing was associated with signif-icantly lower odds of developing asuperficial wound infection com-pared with a traditional dressing.The use of silver-impregnateddressings has been shown toimprove the length of the primarydressing wear time, decrease thenumber of dressing changes,decrease the length of hospitalstay, and avoid skin blistering.22

Although the beneficial effectsof silver dressing use aremultifactorial, minimizing the riskof superficial wound complications

is likely a major step toward less-ening the risk of subsequent PJI.As expected, immunocompromised

patients displayed higher oddsof developing any infection follo-wing THR and TKR comparedwith immunocompetent individuals.However, there were no deep infec-tions in any study patients, includingimmunocompromised patients.Although the control group con-tained significantly greater immuno-compromised patients (Table 1), amultivariate logistic regressionanalysis determined that the use of asilver nylon dressing was indepen-dently associated with significantlylower odds of developing post-operative infection following TJR,

regardless of the immunocompe-tence or ASA score (Table 4).The organisms isolated in the cases

of deep PJI were nearly all atypicalpathogens. Gram-positive cocci arethemost commonpathogens typicallyinvolved in hip and knee PJI. S aureusand coagulase-negative Staphylococcusaccount for 50% to 60% of PJIs,whereas streptococci and enterococcicontribute to approximately 10% ofcases.23 In our series, only one case ofdeep PJI had cultures that grewmethicillin-sensitive S aureus. Fur-thermore, most of the patients withdeep infections were immunocom-promised subjects. We attribute theatypical organisms isolated in the ca-ses of deep PJI to the greater numberof immunocompromised and com-plex patients cared for at our tertiarycare center. Four of 12 patients withdeep PJI had cultures that yielded nogrowth. This is consistent with pre-vious reports in the literature, inwhich 5% to 34% of PJIs were foundto be culture negative.24,25

The cost per unit of silver dressingexceeds that of a standard dressing($30 for Silverlon versus approxi-mately $2 for a standard gauzedressing). The cost of medical and

Table 3

Association Between the Use of Silver Dressing and Any Infection, Superficial Infection, Deep Infection: Results of aUnivariate Logistic Regression Stratified by the Immune Status

Subjects Type of InfectionNo Silver,

62.9%(n= 525)With Silver,

37.1% (n = 309) POdds Ratio for Infection(With Silver) and 95% CI

All subjects Any infection 8.4 (44) 3.9 (12) ,0.0001 0.382 (0.250–0.583)Deep infections 2.3 (12) 0 (0) 0.007 —Superficial infections 6.1 (32) 3.9 (12) 0.011 0.546 (0.343–0.870)

Immunocompetent:N = 59.1% (305)and N = 40.9% (211),without and with silver

Any infectionDeep infectionsSuperficial infectionsAny infection

7.5 (23)1.3 (4)6.2 (19)9.5 (21)

3.3 (7)0 (0)

3.3 (7)5.1 (5)

0.0160.1490.083

,0.001

0.327 (0.131–0.811)—

0.413 (0.152–1.123)0.543 (0.492–0.600)

Immunocompromised:N = 69.2% (220)and N = 30.8% (98),without and with silver

Deep infectionsSuperficial infections

3.6 (8)5.9 (13)

0 (0)5.1 (5)

0.1130.301

—0.935 (0.824–1.061)

The values in parentheses are the number of subjects in that category.CI = confidence interval.

Table 4

Association Between the Use of Silver Dressing and Any Infection: Resultsof a Multivariate Logistic Regression

Population Characteristics Odds Ratio 95% CI P

Age at surgery, yrs 0.989 0.975–1.002 0.102Immunocompromised 1.304 1.092–1.556 0.003ASA high score (3–5) 1.390 0.944–2.047 0.095Silver dressing 0.396 0.273–0.573 ,0.001

CI = confidence interval.

Use of a Silver Nylon Dressing

6 Journal of the American Academy of Orthopaedic Surgeons

surgical treatment for 1 patient withPJI is estimated between $30,000 and$70,000.26,27 Thus, the use of silver-impregnated dressings at the time ofprimary THR and TKRmay result inconsiderable cost savings. A cost-benefit analysis is beyond the scopeof the current study and could be thesubject of future investigation.Our study is limited because of the

short follow-up duration; however,our follow-up interval is sufficient todetect earlyPJI,whichwas theprimaryintent of the study. A longer follow-upduration would be needed to deter-mine the effect of silver dressings onlate PJI. Another limitation is the lackof randomization of patients to spe-cific treatment arms. The study groupswere generated based on the initiationof silver dressing use in our institution.Although the burden of illness associ-ated with PJI is substantial and theoverall national prevalence of PJIcontinues to rise, the overall infectionrates of PJI remain low and are typi-cally in the range of 0.5% to 2%.Thus, our sample size in a single insti-tution may not have been sufficientlylarge to detect significant differenceswith certain subgroups, such as theASA score. This study was conductedas a quality assurance assessment inour specific patient population, andthus results may not be generalizableto the population as a whole.In conclusion, we demonstrated a

significant decrease in superficial anddeepPJIs associatedwith theuse of theSilverlon silver nylon dressing in pri-mary total hip and knee arthroplasty.To our knowledge, this is the firstnon–industry-supported report onthe use of silver-impregnated dress-ings in joint replacement surgery andthe first to specifically investigate theuse of the Silverlon dressing.We hopethat this study will serve as a pilot tofacilitate a power calculation for asubsequent multicentered random-ized controlled trial to further evalu-ate the efficacy of silver dressings intotal hip and knee arthroplasty.

Acknowledgments

The authors acknowledge Dr. RobertG. Davis for the involvement of hispatients in the study.

References

1. Lamagni T: Epidemiology and burden ofprosthetic joint infections. J AntimicrobChemother 2014;69:5-10.

2. Bozic KJ, Kurtz SM, Lau E, Ong K, Chiu V,Vail TP: The epidemiology of revision totalknee arthroplasty in the United States. ClinOrthop Relat Res 2010;468:45-51.

3. Bozic KJ, Kurtz SM, Lau E, Ong K, Vail TP,Berry DJ: The epidemiology of revisiontotal hip arthroplasty in the United States. JBone Joint Surg Am 2009;91:128-133.

4. Tsukayama DT, Estrada R, Gustilo RB:Infection after Total Hip Arthroplasty: Astudy of the treatment of one hundred andsix infections. J Bone Joint Surg Am 1996;78:512-523.

5. Zimmerli W, Trampuz A, Ochsner PE:Prosthetic-joint infections. N Engl J Med2004;351:1645-1654.

6. Kurtz S, Ong K, Lau E, Mowat F, HalpernM: Projections of primary and revision hipand knee arthroplasty in the United Statesfrom 2005 to 2030. J Bone Joint Surg Am2007;89:780-785.

7. Edwards JR, Peterson KD, Mu Y, et al:National healthcare safety network(NHSN) report: Data summary or 2006through 2008, issued December 2009.Am JInfect Control 2009;37:783-805.

8. Sculco TP: The economic impact of infectedtotal joint arthroplasty. Orthopedics 1995;18:871.

9. Kurtz SM, Lau E, Watson H, Schmier JK,Parvizi J: Economic burden ofperiprosthetic joint infection in the UnitedStates. J Arthroplasty 2012;27:61-65.

10. Barillo DJ, Marx DE: Silver in medicine: Abrief history BC 335 to present. Burns2014;40:S3-S8.

11. Galli MM, Protzaman NM, Brigido SA:Utilization of silver hydrogel sheet dressingon postsurgical incisions: A pilot surgery infoot and ankle surgery. Foot Ankle Spec2013;6:422-433.

12. Krieger BR, Davis DM, Sanchez JE, et al:The use of silver nylon in preventingsurgical site infections following colon andrectal surgery. Dis Colon Rectum 2011;54:1014-1019.

13. Schwartz J, Goss S, Facchin F, et al: Aprospective two-armed trial assessing theefficacy and performance of a silverdressing used postoperatively on high-risk,

clean surgical wounds. Ostomy WoundManage 2014;60:30.

14. Siah CJ, Yatim J: Efficacy of a totalocclusive ionic silver-containing dressingcombination in decreasing risk of surgicalsite infection: An RCT. J Wound Care2011;20:561-568.

15. Connery SA, Doens KL, Young C: Aretrospective study evaluating silver-impregnated dressings on cesarean woundhealing. Adv Skin Wound Care 2012;25:414-419.

16. Abboud EC, Legare TB, Settle JC, et al: Dosilver-based wound dressings reduce pain?A prospective study and review of theliterature. Burns 2014;40:S40-S47.

17. Cai J, Karam JA, Parvizi J, Smith EB,Sharkey PF: Aquacel surgical dressingreduces the rate of acute PJI following totaljoint arthroplasty: A case–control study. JArthroplasty 2014;29:1098-1100.

18. Keats AS: The ASA classification ofphysical status—A recapitulation.Anesthesiology 1978;49:233-236.

19. Parvizi J, Zmistowski B, Berbari EF, et al:New definition for periprosthetic jointinfection: From the Workgroup of themusculoskeletal infection society. ClinOrthop Relat Res 2011;469:2992-2994.

20. Pulido L, Ghanem E, Joshi A, Purtill JJ,Parvizi J: Periprosthetic joint infection: Theincidence, timing and predisposing factors.ClinOrthop Relat Res 2008;466:1710-1715.

21. Galat DD, McGovern SC, Larson DR,Harrington JR, Hanssen AD, Clarke HD:Surgical treatment of early woundcomplications following primary total kneearthroplasty. J Bone Joint Surg Am 2009;91:48-54.

22. Hopper GP, Deakin AH, Crane EO, ClarkeJV: Enhancing patient recovery followinglower limb arthroplasty with a modernwound dressing: A prospective, comparativeaudit. J Wound Care 2012;21:200-203.

23. Tande AJ, Patel R: Prosthetic joint infection.Clin Microbiol Rev 2014;27:302-345.

24. Biring GS, Kostamo T, Garbuz DS, MasriBA, Duncan CP: Two-stage revisionarthroplasty of the hip for infection usingan interim articulated Prostalac hip spacer.Bone Joint J 2009;91:1431-1437.

25. Mahmud T, Lyons MC, Naudie DD,MacDonald SJ, McCalden RW: Assessingthe gold standard: A review of 253 two-stage revisions for infected TKA. ClinOrthop Relat Res 2012;470:2730-2736.

26. Backe H, Wolff D, Windsor R: Total kneereplacement infection after 2-stage re-implantation. Clin Orthop 1996;331:125-131.

27. Darouiche RO: Treatment of infectionsassociated with surgical implants. N Engl JMed 2004;350:1422-1429.

Ashley J. Tisosky, MD, et al

October 2017, Vol 1, No 7