Younger Age Is Associated with a Higher Risk of Early Periprosthetic Joint Infection and Aseptic Mechanical FailureAfterTotalKneeArthroplasty

Transcription

1 529 COPYRIGHT Ó 2014 BY THE JOURNAL OF BONE AND JOINT SURGERY, INCORPORATED Younger Age Is Associate with a Higher Risk of Early Periprosthetic Joint Infection an Aseptic Mechanical FailureAfterTotalKneeArthroplasty John P. Meehan, MD, Beate Danielsen, PhD, Sunny H. Kim, PhD, Amir A. Jamali, MD, an Richar H. White, MD Investigation performe at the University of California Davis Health System, Sacramento, California Backgroun: Although early aseptic mechanical failure after total knee arthroplasty has been reporte in younger patients, it is unknown whether early revision ue to periprosthetic joint infection is more or less frequent in this patient subgroup. The purpose of this stuy was to etermine whether the incience of early periprosthetic joint infection requiring revision knee surgery is significantly ifferent in patients younger than fifty years of age compare with oler patients following primary unilateral total knee arthroplasty. Methos: A large population-base stuy was conucte with use of the California Patient Discharge Database, which allows serial linkage of all ischarge ata from nonfeeral hospitals in the state over time. Patients unergoing primary unilateral total knee arthroplasty uring 2005 to 2009 were ientifie. Principal outcomes were partial or complete revision arthroplasty ue to periprosthetic joint infection or ue to aseptic mechanical failure within one year. Multivariate analysis inclue risk ajustment for important emographic an clinical variables. The effect of hospital total knee arthroplasty volume on the outcomes of infection an mechanical failure was analyze with use of hierarchical moeling. Results: At one year, 983 (0.82%) of 120,538 primary total knee arthroplasties ha unergone revision ue to periprosthetic joint infection an 1385 (1.15%) ha unergone revision ue to aseptic mechanical failure. The cumulative incience in patients younger than fifty years of age was 1.36% for revision ue to periprosthetic joint infection an 3.49% for revision ue to aseptic mechanical failure. In risk-ajuste moels, the risk of periprosthetic joint infection was 1.8 times higher in patients younger than fifty years of age (os ratio = 1.81, 95% confience interval = 1.33 to 2.47) compare with patients sixty-five years of age or oler, an the risk of aseptic mechanical failure was 4.7 times higher (os ratio = 4.66, 95% confience interval = 3.77 to 5.76). The rate of revision ue to infection at hospitals in which a mean of more than 200 total knee arthroplasties were performe per year was lower than the expecte (mean) value (p = 0.04). Conclusions: Patients younger than fifty years of age ha a significantly higher risk of unergoing revision ue to periprosthetic joint infection or to aseptic mechanical failure at one year after primary total knee arthroplasty. Level of Evience: Prognostic Level III. See Instructions for Authors for a complete escription of levels of evience. Peer Review: This article was reviewe by the Eitor-in-Chief an one Deputy Eitor, an it unerwent bline review by two or more outsie experts. It was also reviewe byan expert in methoologyan statistics. The Deputy Eitor reviewe each revision of the article, an it unerwent a final reviewbythe Eitor-in-Chief prior to publication. Final corrections an clarifications occurreuring one or more exchanges between the author(s) an copyeitors. Disclosure: None of the authors receive payments or services, either irectly or inirectly (i.e., via his or her institution), from a thir party in support of any aspect of this work. One or more of the authors, or his or her institution, has ha a financial relationship, in the thirty-six months prior to submission of this work, with an entity in the biomeical arena that coul be perceive to influence or have the potential to influence what is written in this work. No author has ha any other relationships, or has engage in any other activities, that coul be perceive to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitte by authors are always provie with the online version of the article. A commentary by Kelly G. Vince, MD, is linke to the online version of this article at jbjs.org. J Bone Joint Surg Am. 2014;96:

2 530 The annual number of total knee arthroplasties continues to rise. Extrapolating the trens in the number of arthroplasties performeuring the past ecae, Kurtz et al. projecte that by 2030, patients younger than sixty-five years of age will make up the majority of patients unergoing primary total knee arthroplasty an that up to one million total knee arthroplasties may be performe annually in patients younger than fifty-five years of age 1. The incience of early failure of prosthetic knees in the younger patient population has not been extensively stuie to our knowlege. Intermeiate an long-term failure rates in younger patients unergoing total knee arthroplasty are higher than those in oler patients, but less is known about the ifference in early or short-term failure rates with respect to age 2,3.Publishe stuies of total knee arthroplasty outcomes have shown that early aseptic mechanical (prosthesis) failure an early periprosthetic joint infection each lea to revision surgery in approximately 1% to 2% of all patients 4-7. However, most of those stuies involve patients in the age range of sixty to eighty years. An analysis of all age groups by the Australian National Joint Replacement Registry inicate a higher cumulative revision rate at early follow-up times after primary knee replacement in younger compare with oler iniviuals, but the analysis i not istinguish between infectious an aseptic causes of revision accoring to age 8. The primary purpose of the present stuy was to etermine whether the incience of periprosthetic joint infection requiring revision surgery after primary unilateral total knee arthroplasty is ifferent in patients younger than fifty years of age. We hypothesize that the one-year incience of periprosthetic joint infection requiring revision arthroplasty woul be similar in this patient group an oler patients. Materials an Methos This was a retrospective observational stuy esigne to analyze the outcomes of patients who ha unergone a primary total knee arthroplasty (ICD-9-CM [International Classification of Diseases, 9th Revision, Clinical Moification] coe 81.54) in a nonfeeral hospital in California uring a fiveyear perio from January 1, 2005, to December 31, This stuy was approve by the California Committee for the Protection of Human Subjects an by the University of California Davis institutional review boar. Database California hospitals must report information about all hospitalize patients after they are ischarge. This Patient Discharge Database (PDD) can be use to ientify temporally linke serial hospitalizations with use of an encrypte recor linkage number (RLN) that is base on the patient s Social Security Number. Recor linkage allows late complications to be ientifie even if the patient is hospitalize in a ifferent hospital in the state. All PDD recors inclue emographic information, a principal meical iagnosis, up to twentyfour aitional seconary iagnoses, an a principal proceure an up to twenty seconary proceures, all coe with use of the ICD-9-CM. Assembly of the Total Knee Arthroplasty Cohort All patients coe as having only one total knee arthroplasty performeuring the stuy perio were ientifie. We then minimize the number of patients who may have ha a primary total knee arthroplasty prior to the stuy perio by excluing all patients coe as having ha a primary total knee arthroplasty (ICD-9-CM 81.54), a revision total knee arthroplasty (81.55 or to 00.84), or any history of a prior total knee arthroplasty (V43.65) at any ischarge from 1991 to We also exclue all patients with a secon primary total knee arthroplasty (81.54) within one year after the inex total knee arthroplasty an those coe as having cancer (190.x to 209.x) or a rheumatic isease (714.xx, 710.xx, 696.x). (For all coes, x inicates that any integer was acceptable.) Outcomes Outcomes were prespecifie with use of specific ICD-9-CM coes. These outcomes inclue specific major orthopaeic complications that require hospitalization couple with specific proceures occurring within 365 ays after the inex total knee arthroplasty. The principal outcome of interest was a major periprosthetic joint infection leaing to total joint revision surgery, tibial liner removal, arthrotomy, ebriement, synovectomy, or other excision, arthrotomy for removal of the prosthesis, or above-the-knee amputation from thirty to 365 ays after the inex operation. We also etermine the incience of aseptic mechanical failure. For patients coe as having only mechanical failure, we inclue all cases of revision knee arthroplasty or revision of the femoral component, patellar component, tibial component, or tibial liner within one year after the inex arthroplasty. For each knee arthroplasty failure, we istinguishe between infectious an mechanical causes with use of the meical iagnosis coes that accompanie the revision knee proceure coe. To be categorize as an infectionrelate revision, we require (1) a coe for infection ue to an internal joint prosthesis (996.66); infection ue to a evice, implant, or graft (996.67); other postoperative infection (998.59); postoperative seroma (998.51); or periostitis without mention of osteomyelitis (730.36), couple with either (2) a complete or partial knee arthroplasty revision (81.55, through 00.84, 80.06, or 84.17) or (3) arthrotomy (80.16), excisional ebriement (86.22), synovectomy (80.76), or other local excision or estruction (80.86) performe more than thirty ays after the inex total knee arthroplasty. A minor periprosthetic joint infection was efine as (1) one of the coes for infection, couple with (2) liner removal (00.84), arthrotomy (80.16), ebriement (86.22), synovectomy (80.76), or other excision (80.86) within thirty ays after the total knee arthroplasty, without subsequent revision knee arthroplasty. Mechanical complications leaing to revision consiste of (1) a mechanical complication of an orthopaeic evice (996.4x), ankylosis of the joint (718.56), recurrent islocation (718.36), complications cause by a joint prosthesis (996.77), or a complication ue to an internal orthopaeic evice (996.78), couple with (2) revision arthroplasty (81.55), removal of an implante evice (78.67), internal femoral fixation without a fracture (78.55), other partial ostectomy (77.86), close reuction without internal fixation (79.06), or ivision of the joint capsule, ligament, or cartilage (80.46). As a means of assessing the external valiity of the ata, we also inclue a thir outcome measure, major cariovascular complications within thirty ays, which was a composite of coronary outcomes (acute myocarial infarction, coronary angioplasty/stenting, or coronary artery bypass surgery), postoperative shock, or eath uring this time perio. Statistics Age was categorize as younger than fifty years, fifty to sixty-four years, or sixtyfive years of age or oler. Race/ethnicity was categorize as Hispanic, black, Asian, white, an other, which inclue mixe race an Native Americans. Logistic regression was use to ajust for emographic variables an the number of chronic comorbiities. The effect of annual hospital total knee arthroplasty volume was evaluate with use of hierarchical moeling, with the hospital as a ranom effect. Specifically, the number of observe periprosthetic joint infections in hospitals that performe <50, 50 to 100, 101 to 200, or >200 such proceures per year uring the stuy perio was compare with the expecte (mean) number preicte with use of the risk-ajuste moel. Categorical ata were analyze with use of chi-square testing. Significance testing for trens in categorical ata was performe with use of the Cochran- Armitage test for tren with a two-sie p value.

3 531 TABLE I Clinical Characteristics of Patients with Revision Due to Infection or Aseptic Mechanical Failure within One Year After Total Knee Arthroplasty (TKA) Major Knee Infection Aseptic Mechanical Failure Risk Factor No. of Patients No. Revise Percentage P Value for Tren* No. Revise Percentage P Value for Tren* Total 120, Patient age in yr <0.001 <0.001 < , , Sex F 74, M 46, Race/ethnicity White 87, Hispanic 17, Black Asian Other Annual hospital TKA volume < < to , to , >200 57, Calenar year , , , , , No. of comorbiities < , , , Comorbiity Anemia 18, # Heart failure # Pulmonary obst. is. 18, # Depression 12, # # Diabetes 24, # Obesity 26, # # Peripheral vascular is # Psychosis # # Pulmonary circ. is Renal failure Valvular is *Cochran-Armitage test for tren. P < (chi-square, within risk-group ifference). P < 0.05 (chi-square). P > 0.05 (chi-square). #P < (chi-square).

4 532 TABLE II Risk-Ajuste Results for Major Periprosthetic Infection Risk Factor OR 95% CI P Value Age in yr < Ref. Sex F Ref. M Race/ethnicity White Ref Hispanic Black Asian Other No. of comorbiities* 0 Ref < < <0.001 *Elixhauser comorbiity count. Source of Funing No external funing was receive for this stuy. Results The clinical characteristics of the patient cohort an patients who unerwent revision as a result of infection or mechanical malfunction within one year are shown in Table I. We ientifie 120,538 patients who unerwent a primary unilateral total knee arthroplasty uring the five-year time perio ani not appear to have unergone either a prior total knee arthroplasty performe subsequent to 1990 or a secon primary total knee arthroplasty performeuring the first year after the inex total knee arthroplasty. One or more total knee arthroplasties were performe at approximately 300 hospitals in California. Overall, 0.82% (983) of the patients require revision knee surgery within one year because of a major periprosthetic joint infection. The incience was highest (1.36%, seventy-two patients) among those younger than fifty years of age. The rate in this age group was approximately twice that in patients sixtyfive years of age or oler (0.73%, 545 patients). The one-year incience of major periprosthetic joint infection ecrease significantly (p < 0.001) with increasing age group. Overall, aseptic mechanical failure requiring revision surgery within one year occurre in 1.15% (1385) of the patients. The highest incience of 3.49% (185 patients) was among patients younger than fifty years of age, an the incience ecrease significantly (p < 0.001) with increasing age group, reaching 0.75% (560 patients) among those sixty-five years of age or oler. In each of the age groups, the incience of infection was significantly higher (p < 0.001) in men (1.04%, 477 patients) than in women (0.68%, 506 patients), but there was no ifference in the incience of aseptic mechanical failure, which was 1.17% (541 patients) in men compare with 1.13% (844 patients) in women. There was a significant ecrease in the incience of major periprosthetic knee infection from 2005 (0.96%) to 2009 (0.76%, p = 0.02) but no significant change in the incience of aseptic mechanical failure. There was a steay increase in the one-year incience of periprosthetic joint infection with an increase in the number of chronic comorbi conitions present (from 0.64% for patients with no comorbiity to 1.81% for those with five or more comorbiities, p < 0.001). The results of the multivariate logistic regression moeling of risk factors for periprosthetic joint infection are shown in Table II. Significant preictors of knee revision surgery ue to infection inclue an age younger than fifty years (os ratio [OR] = 1.81, 95% confience interval [CI] = 1.33 to 2.47, relative to an age of sixty-five years or over), male sex (OR = 1.60, 95% CI = 1.31 to 1.97), an black race (OR = 1.38 relative to non-hispanic whites, 95% CI = 1.06 to 1.80). Risk (relative to zero comorbiities) increase steaily with the number of chronic comorbiities, from OR = 1.21 (95% CI = 1.02 to 1.44) for one comorbiity to OR = 2.92 (95% CI = 2.23 to 3.84) for four or more. The comorbiities associate with the highest risk of periprosthetic joint infection were psychosis (OR = 1.65, 95% CI = 1.21 to 2.25), congestive heart failure (OR = 1.64, 95% CI = 1.26 to 2.13), obesity (OR = 1.33, 95% CI = 1.15 to 1.54), aniabetes (OR = 1.31, 95% CI = 1.13 to 1.52). Moeling the hospital as a ranom effect, the ratio of the number of observe cases of periprosthetic knee infection requiring revision to the expecte (mean) number (the O/E ratio) was significantly lower (0.80, 95% CI = 0.64 to 0.99) for hospitals that performe >200 total knee arthroplasties per year (Table III). In the multivariate moel preicting the risk of revision knee arthroplasty ue to aseptic mechanical failure (Table IV), the risk was significantly higher for patients younger than fifty years of age compare with patients sixty-five years of age or oler (OR = 4.66, 95% CI = 3.77 to 5.76, p < 0.01) an for blacks compare with non-hispanic whites (OR = 1.44, 95% CI = 1.17 to 1.79, p = 0.002). The presence of four or more TABLE III Effect of Hospital Volume on Major Periprosthetic Infection* Annual Total Knee Arthroplasty Volume O/E Ratio 95% CI P Value < > *The hospital was treate as a ranom effect.

5 533 TABLE IV Risk-Ajuste Results for Aseptic Mechanical Failure Risk Factor OR 95% CI P Value Age in yr < < < Ref. Sex F Ref. M Race/ethnicity White Ref. Hispanic Black <0.01 Asian Other No. of comorbiities* 0 Ref *Elixhauser comorbiity count. TABLE V Effect of Hospital Volume on Aseptic Mechanical Failure* Annual Total Knee Arthroplasty Volume O/E Ratio 95% CI P Value < > *The hospital was treate as a ranom effect. comorbiities was associate with a significantly higher risk of revision compare with no comorbiities (OR = 1.39, 95% CI = 1.05 to 1.84, p = 0.03). Sex was not a risk factor for aseptic mechanical failure. Greater hospital volume was also not a preictor of a lower incience of aseptic mechanical failure; the number of such failures ecrease as hospital volume increase, but in each category the observe failure rate i not iffer significantly from that expecte on the basis of all hospital volumes combine (Table V). The incience of the composite outcome of eath or a major cariovascular complication within thirty ays after total knee arthroplasty was low, 0.58% (95% CI = 0.54% to 0.62%). In the multivariate moel, the risk increase significantly (p < 0.001) with increasing age; was significantly higher in men (OR = 2.50, 95% CI = 1.96 to 3.20); an was strongly associate with the number of comorbiities, increasing from OR = 1.65 (95% CI = 1.29 to 2.10) for one comorbiity (relative to none) to OR = 7.91 (95% CI = 5.97 to 10.48) for four or more. Higher hospital total knee arthroplasty volume was associate with a lower incience, but the ratio between the number of observe cases for hospitals performing >200 such proceures per year an the expecte (mean) number i not quite reach significance (O/E ratio = 0.84, 95% CI = 0.70 to 1.00, p = 0.052). In orer to allow for a more irect comparison of these results with registry ata, we performe an aitional analysis of patients in California who unerwent revision knee surgery ue to periprosthetic joint infection or ue to aseptic mechanical failure within two years (see Appenix). The results incience of periprosthetic joint infection an incience of aseptic mechanical failure within two years were likewise greater in patients younger than fifty years of age. Discussion Much of the information publishe regaring the causes of total knee arthroplasty failure an inications for revision in the Unite States has come from case series from a single center, cohort stuies from large acaemic institutions, or aministrative atabases such as the Nationwie Inpatient Sample (NIS) an the Meicare 5% national sample aministrative ata set Many of these stuies involve primarily patients who are sixty-five years or oler, an most escribe only intermeiate to long-term failure rates an mechanisms. Elevate intermeiate an long-term rates of total knee arthroplasty revision ue to aseptic mechanical failure have been escribe in younger patients Typically, the elevate rate is attribute to the greater activity level in these younger iniviuals an to the cumulative stresses on the prosthesis, bone, an soft-tissue interfaces over many years. Less is known about the relationship between younger age an the short-term outcomes of total knee arthroplasty. Julin et al. use the Finnish Arthroplasty Register, which recors ata prospectively an captures 96% of primary knee replacements performe in Finlan, to ientify younger age (especially less than fifty-five years of age) as a risk factor for revision of total knee arthroplasty within five years ue to aseptic mechanical failure 19. Their primary outcome measure was the prosthesis survival rate, which they escribe as the proportion of prostheses surviving without revision uring the follow-up perio. However, they faile to ientify any association between patient age an failure specifically ue to periprosthetic joint infection. Bohm et al. reporte finings from the Canaian Institute for Health Information Discharge Abstract Database regaring >65,000 hip an knee arthroplasties performe from 2005 to They conclue that the rehospitalization rate ue to infection within the first year after knee replacement was significantly higher (p < 0.001) in men (1.6%) than in women (1.2%). They also state that infection rates i not vary significantly accoring to age. With the American Joint Replacement Registry still in its early stages, the ability to ientify a complication within a large

6 534 population of patients who unergo a specific proceure has typically involve use of the NIS atabase or the Meicare 5% national sample aministrative ata set. For joint replacement proceures, however, these ata sets have limitations because they o not inclue information on all patients who unerwent the proceure. Instea, they use sampling to extrapolate finings to the larger population. Also, prior to 2005, ICD-9-CM coes were not use to ientify the specific type of prosthetic joint failure. We use the California PDD to ientify an capture all patients (not limite to Meicare beneficiaries) who unerwent primary total knee arthroplasty from January 1, 2005, to December 31, 2009, an subsequent revision ue to either periprosthetic infection or aseptic mechanical failure within one year. This atabase provies the unique ability to ientify an serially follow all patients who are hospitalize at any nonfeeral hospital in California, regarless of whether the hospital was the site of the primary or the revision proceure. With the resulting ata set, we ientifie younger age as an inepenent risk factor for both periprosthetic joint infection an aseptic mechanical failure within one year. To our knowlege, this is the first stuy to ocument a higher incience of early revision ue to periprosthetic joint infection after knee arthroplasty in patients younger than fifty years of age. Patients who were younger than fifty years of age when they unerwent the primary knee replacement were almost twice as likely to require revision surgery within one year ue to periprosthetic joint infection compare with patients who were sixty-five years of age or oler. Also, patients who were younger than fifty years of age were almost five times as likely to unergo revision ue to aseptic mechanical failure within one year compare with patients sixty-five years of age or oler. One potential explanation for the higher incience of revision arthroplasty ue to periprosthetic joint infection in younger patients is the higher prevalence of seconary (e.g., posttraumatic) osteoarthritis in younger patients compare with primary osteoarthritis in oler patients. Patients with posttraumatic osteoarthritis frequently have unergone an arthrotomy with open reuction an internal fixation of the traumatic injury. A previous arthrotomy is a recognize risk factor for periprosthetic knee infection 21. However, a clear relationship between previous arthrotomy an aseptic mechanical failure has not been establishe. Prosthesis fixation issues, such as the use of fixation without cement, have also been implicate as a potential cause of aseptic mechanical failure in the young patient 2. Inthepresentanalysis,wewerealsoabletoevaluatetheeffect of hospital volume an the specific outcomes of periprosthetic joint infection an aseptic mechanical failure. In agreement with previous stuies, higher-volume hospitals (>200 total knee arthroplasties per year) ha a lower incience of early periprosthetic joint infections compare with lower-volume hospitals 5,22.Wei not emonstrate a similar relationship between higher hospital volume an a lower rate of early aseptic mechanical failure, although we i observe a tren. Manley et al. also i not fin a relationship between hospital total knee arthroplasty volume an early aseptic mechanical failure in the Meicare population, although they i fin that patients treate at the lowest-volume hospitals (one to twenty-five such proceures per year) ha a higher risk of revision at five an eight years compare with those treate at the highest-volume hospitals (>200 proceures) 14. The strengths of the present stuy inclue the inclusion of all patients who unerwent primary total knee arthroplasty in California uring a five-year time perio an the ability to ientify all patients who unerwent revision surgery in the state, regarless of where the primary arthroplasty ha been performe. All complications were efine explicitly with use of ICD-9-CM coing 23,24, the major orthopaeic complication coes use have high preictive value 25, an these coes were couple with major proceure coes, making case ientification even more reliable. The limitations of the stuy inclue the retrospective esign, reliance on aministrative ata, an inability to account for the effect of iniviual surgeons. We also coul not ajust for potential confouners such as the severity of the joint isease, fixation with an without cement, an the type or uration of antibacterial prophylaxis use. In conclusion, the results of this stuy reveale that younger patients who unerwent a primary total knee arthroplasty ha higher one-year an two-year inciences of both periprosthetic joint infection an aseptic mechanical failure requiring revision arthroplasty. Thus, performing this proceure in iniviuals younger than fifty years of age shoul be approache with caution. The high incience of early infectious an mechanical failures warrants further investigation to etermine which specific patient characteristics in this younger age group contribute to their higher risk. The fining of a lower incience of periprosthetic joint infection at high-volume hospitals is consistent with previous stuies an supports the concept of utilizing specialty hospitals an hospitals with specialize surgical services as a means to reuce complications. Appenix A table showing the clinical characteristics of patients unergoing revision ue to infection or aseptic mechanical failure within two years after total knee arthroplasty is available with the online version of this article as a ata supplement at jbjs.org. n John P. Meehan, MD Department of Orthopaeic Surgery, University of California Davis, 4860 Y Street, Suite 3800, Sacramento, CA aress: John.Meehan@ucmc.ucavis.eu Beate Danielsen, PhD Health Information Solutions, 2425 Clubhouse Drive, Rocklin, CA Sunny H. Kim, PhD School of Meicine, University of California Davis, 2921 Stockton Boulevar, Suite 1400, Sacramento, CA 95817

7 535 Amir A. Jamali, MD Joint Preservation Institute, 2825 J Street, Suite 440, Sacramento, CA Richar H. White, MD Department of Meicine, University of California Davis, Suite 2400, PSSB, 4150 V Street, Sacramento, CA References 1. Kurtz SM, Lau E, Ong K, Zhao K, Kelly M, Bozic KJ. Future young patient eman for primary an revision joint replacement: national projections from 2010 to Clin Orthop Relat Res Oct;467(10): Epub 2009 Apr Paxton EW, Namba RS, Maletis GB, Khato M, Yue EJ, Davies M, Low RB Jr, Wyatt RW, Inacio MC, Funahashi TT. A prospective stuy of 80,000 total joint an 5000 anterior cruciate ligament reconstruction proceures in a community-base registry in the Unite States. J Bone Joint Surg Am Dec;92(Suppl 2): Harrysson OL, Robertsson O, Nayfeh JF. Higher cumulative revision rate of knee arthroplasties in younger patients with osteoarthritis. Clin Orthop Relat Res Apr;421: Kurtz SM, Ong KL, Lau E, Bozic KJ, Berry D, Parvizi J. Prosthetic joint infection risk after TKA in the Meicare population. Clin Orthop Relat Res Jan;468(1):52-6. Epub 2009 Aug Katz JN, Barrett J, Mahome NN, Baron JA, Wright RJ, Losina E. Association between hospital an surgeon proceure volume an the outcomes of total knee replacement. J Bone Joint Surg Am Sep;86(9): Losina E, Katz JN. Total knee arthroplasty on the rise in younger patients: are we sure that past performance will guarantee future success? Arthritis Rheum Feb;64(2): Weinstein AM, Rome BN, Reichmann WM, Collins JE, Burbine SA, Thornhill TS, Wright J, Katz JN, Losina E. Estimating the buren of total knee replacement in the Unite States. J Bone Joint Surg Am Mar 6;95(5): Australian Orthopaeic Association National Joint Replacement Registry. Hip an knee arthroplasty annual report. Accesse 2013 Sep Bozic KJ, Kurtz SM, Lau E, Ong K, Chiu V, Vail TP, Rubash HE, Berry DJ. The epiemiology of revision total knee arthroplasty in the Unite States. Clin Orthop Relat Res Jan;468(1): Epub 2009 Jun Bozic KJ, Lau E, Kurtz S, Ong K, Berry DJ. Patient-relate risk factors for postoperative mortality an periprosthetic joint infection in meicare patients unergoing TKA. Clin Orthop Relat Res Jan;470(1): Fehring TK, Oum S, Griffin WL, Mason JB, Naau M. Early failures in total knee arthroplasty. Clin Orthop Relat Res Nov;392: Callahan CM, Drake BG, Heck DA, Dittus RS. Patient outcomes following tricompartmental total knee replacement. A meta-analysis. JAMA May 4; 271(17): Galat DD, McGovern SC, Larson DR, Harrington JR, Hanssen AD, Clarke HD. Surgical treatment of early woun complications following primary total knee arthroplasty. J Bone Joint Surg Am Jan;91(1): Manley M, Ong K, Lau E, Kurtz SM. Total knee arthroplasty survivorship in the Unite States Meicare population: effect of hospital an surgeon proceure volume. J Arthroplasty Oct;24(7): Epub 2008 Nov Robertsson O, Knutson K, Lewol S, Ligren L. The Sweish Knee Arthroplasty Register : an upate with special emphasis on 41,223 knees operate on in Acta Orthop Scan Oct;72(5): Gioe TJ, Novak C, Sinner P, Ma W, Mehle S. Knee arthroplasty in the young patient: survival in a community registry. Clin Orthop Relat Res Nov;464: Gioe TJ, Killeen KK, Grimm K, Mehle S, Scheltema K. Why are total knee replacements revise?: analysis of early revision in a community knee implant registry. Clin Orthop Relat Res Nov;428: Australian Orthopaeic Association National Joint Replacement Registry. Hip an knee arthroplasty. Annual report ocuments/10180/42662/annual%20report%202008?version=1. 1&t= Accesse 2009 Apr Julin J, Jämsen E, Puolakka T, Konttinen YT, Moilanen T. Younger age increases the risk of early prosthesis failure following primary total knee replacement for osteoarthritis. A follow-up stuy of 32,019 total knee replacements in the Finnish Arthroplasty Register. Acta Orthop Aug;81(4): Bohm ER, Dunbar MJ, Froo JJ, Johnson TM, Morris KA; Rehospitalizations, early revisions, infections, an hospital resource use in the first year after hip an knee arthroplasties. J Arthroplasty Feb;27(2): Epub 2011 Jul Ayers DC, Dennis DA, Johanson NA, Pellegrini JVD. Instructional course lectures, the American Acaemy of Orthopaeic Surgeons - common complications of total knee arthroplasty. J Bone Joint Surg Am. 1997;79(2): Hervey SL, Purves HR, Guller U, Toth AP, Vail TP, Pietrobon R. Provier volume of total knee arthroplasties an patient outcomes in the HCUP-Nationwie Inpatient Sample. J Bone Joint Surg Am Sep;85(9): Birman-Deych E, Waterman AD, Yan Y, Nilasena DS, Rafor MJ, Gage BF. Accuracy of ICD-9-CM coes for ientifying cariovascular an stroke risk factors. Me Care May;43(5): White RH, Garcia M, Saeghi B, Tancrei DJ, Zrelak P, Cuny J, Sama P, Gammon H, Schmaltz S, Romano PS. Evaluation of the preictive value of ICD-9-CM coe aministrative ata for venous thromboembolism in the Unite States. Thromb Res Jul;126(1):61-7. Epub 2010 Apr Bozic KJ, Chiu VW, Takemoto SK, Greenbaum JN, Smith TM, Jerabek SA, Berry DJ. The valiity of using aministrative claims ata in total joint arthroplasty outcomes research. J Arthroplasty Sep;25(6)(Suppl):58-61.Epub 2010 Jun 08.