Literature Decoded

Living reprintoriginally published in JAMA Network Open 2023CC BY 4.0Read the version of record Opt out

Literature DecodedLiving reprint
Living reprint · journal article Journal version available

Incidence, Microbiological Studies, and Factors Associated With Prosthetic Joint Infection After Total Knee Arthroplasty

Study first published in JAMA Network Open (2023), reprinted in full under its CC BY 4.0 licence.

Erica J. Weinstein1*, Alisa J. Stephens-Shields2, Craig W. Newcomb2, Randi Silibovsky3, Charles L. Nelson4, Judith A. O’Donnell3, Laurel J. Glaser5, Evelyn Hsieh6, Jennifer S. Hanberg7, Janet P. Tate8, Kathleen M. Akgün9, Joseph T. King10, Vincent Lo Re1

1Division of Infectious Diseases, Department of Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia; Center for Real-World Effectiveness and Safety of Therapeutics, Center for Clinical Epidemiology and Biostatistics, Department of Biostatistics, Epidemiology, and Informatics, Perelman School of Medicine, University of Pennsylvania, Philadelphia2Center for Real-World Effectiveness and Safety of Therapeutics, Center for Clinical Epidemiology and Biostatistics, Department of Biostatistics, Epidemiology, and Informatics, Perelman School of Medicine, University of Pennsylvania, Philadelphia3Division of Infectious Diseases, Department of Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia4Department of Orthopedic Surgery, Perelman School of Medicine, University of Pennsylvania, Philadelphia5Department of Pathology, Perelman School of Medicine, University of Pennsylvania, Philadelphia6Veterans Affairs (VA) Connecticut Health System, West Haven; Section of Rheumatology, Allergy and Immunology, Department of Medicine, Yale University School of Medicine, New Haven, Connecticut7Veterans Affairs (VA) Connecticut Health System, West Haven; Department of Medicine, Massachusetts General Hospital, Boston8Veterans Affairs (VA) Connecticut Health System, West Haven; Department of Medicine, Yale University School of Medicine, New Haven, Connecticut9Section of Pulmonary, Critical Care, and Sleep Medicine, VA Connecticut Health System, West Haven; Section of Pulmonary, Critical Care, and Sleep Medicine, Yale University School of Medicine, New Haven, Connecticut10Veterans Affairs (VA) Connecticut Health System, West Haven; Department of Neurosurgery, Yale University School of Medicine, New Haven, Connecticut*Corresponding author

Reprinted 2026-10-01 21 min read Living reprint · journal article Version of record: JAMA Network Open 2023Licence: CC BY 4.0

Reading mode
In plain languageStudy first published in JAMA Network Open (2023), reprinted in full under its CC BY 4.0 licence.

Study first published in JAMA Network Open (2023), reprinted in full under its CC BY 4.0 licence.

No plain-language summary has been written for this reprint yet. The authors' abstract and full text follow, unchanged apart from layout.

Educational summary of research findings; not medical advice. Discuss care decisions with a qualified clinician.

Bottom line

What the evidence supports

How to read it

No plain-language summary has been written for this reprint yet. The authors' abstract and full text follow, unchanged apart from layout.

Limitations

The findings apply to the included study populations and may not generalise to every person or setting.

Disclaimer

Educational summary of research findings; not medical advice. Discuss care decisions with a qualified clinician.

The paper

Full manuscript

Abstract

Importance

Despite the frequency of total knee arthroplasty (TKA) and clinical implications of prosthetic joint infections (PJIs), knowledge gaps remain concerning the incidence, microbiological study results, and factors associated with these infections.

Objectives

To identify the incidence rates, organisms isolated from microbiological studies, and patient and surgical factors of PJI occurring early, delayed, and late after primary TKA.

Design, Setting, and Participants

This cohort study obtained data from the US Department of Veterans Affairs (VA) Corporate Data Warehouse on patients who underwent elective primary TKA in the VA system between October 1, 1999, and September 30, 2019, and had at least 1 year of care in the VA prior to TKA. Patients who met these criteria were included in the overall cohort, and patients with linked Veterans Affairs Surgical Quality Improvement Program (VASQIP) data composed the VASQIP cohort. Data were analyzed between December 9, 2021, and September 18, 2023.

Exposures

Primary TKA as well as demographic, clinical, and perioperative factors.

Main Outcomes and Measures

Incident hospitalization with early, delayed, or late PJI. Incidence rate (events per 10 000 person-months) was measured in 3 postoperative periods: early (≤3 months), delayed (between >3 and ≤12 months), and late (>12 months). Unadjusted Poisson regression was used to estimate incidence rate ratios (IRRs) with 95% CIs of early and delayed PJI compared with late PJI. The frequency of organisms isolated from synovial or operative tissue culture results of PJIs during each postoperative period was identified. A piecewise exponential parametric survival model was used to estimate IRRs with 95% CIs associated with demographic and clinical factors in each postoperative period.

Results

The 79 367 patients (median (IQR) age of 65 (60-71) years) in the overall cohort who underwent primary TKA included 75 274 males (94.8%). A total of 1599 PJIs (2.0%) were identified. The incidence rate of PJI was higher in the early (26.8 [95% CI, 24.8-29.0] events per 10 000 person-months; IRR, 20.7 [95% CI, 18.5-23.1]) and delayed periods (5.4 [95% CI, 4.9-6.0] events per 10 000 person-months; IRR, 4.2 [95% CI, 3.7-4.8]) vs the late postoperative period (1.3 events per 10 000 person-months). Staphylococcus aureus was the most common organism isolated overall (489 [33.2%]); however, gram-negative infections were isolated in 15.4% (86) of early PJIs. In multivariable analyses, hepatitis C virus infection, peripheral artery disease, and autoimmune inflammatory arthritis were associated with PJI across all postoperative periods. Diabetes, chronic kidney disease, and obesity (body mass index of ≥30) were not associated factors. Other period-specific factors were identified.

Conclusions and Relevance

This cohort study found that incidence rates of PJIs were higher in the early and delayed vs late post-TKA period; there were differences in microbiological cultures and factors associated with each postoperative period. These findings have implications for postoperative antibiotic use, stratification of PJI risk according to postoperative time, and PJI risk factor modification.

Introduction

Total knee arthroplasty (TKA) is one of the most common elective surgeries in the US due to increasing rates of obesity and an aging population.1,2 Prosthetic joint infection (PJI), one of the most serious complications of TKA, can result in substantial morbidity, mortality, and health care costs.3 Despite the clinical implications of PJIs, there are major knowledge gaps regarding the epidemiological features, microbiological studies, and factors of these infections in the US.

Previous studies of PJI after primary (ie, initial) TKA conducted predominantly with claims data were limited by unvalidated or surrogate end points,4,5,6,7,8 had short follow-up duration,9,10 focused on adults 65 years or older,7,11,12,13 involved single centers,14,15,16,17 and evaluated few risk factors in mainly unadjusted analyses.18,19,20,21 To our knowledge, no US study has compared the incidence, microbiological studies, and factors of PJI occurring in key periods after TKA: early PJI (≤3 months), delayed PJI (between >3 and ≤12 months), and late PJI (>12 months). Prior literature has suggested that the pathogenesis of PJI differs by time points after TKA and that these specified periods may classify the different mechanisms and organisms associated with this infection.22,23 These data could identify subgroups at highest risk for PJI after primary TKA and could inform empirical antibiotic therapy for suspected PJI after this surgery.

To address these knowledge gaps, we evaluated a national cohort of veterans who underwent primary TKA within the US Department of Veterans Affairs (VA) Healthcare System. We aimed to identify the incidence rates, organisms isolated from microbiological cultures, and patient and surgical factors of PJI occurring early, delayed, and late after primary TKA.

Methods

Study Design and Data Source

We conducted a retrospective cohort study of patients who underwent primary TKA in the national VA system. The Human Investigations Committee at the VA Connecticut Health System, Yale University, and University of Pennsylvania approved the study and waived the informed consent requirement because this research could not practicably be conducted without this waiver. We followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.24

We collected electronic health record (EHR) data from the VA Corporate Data Warehouse. The data set included demographic characteristics, hospital and ambulatory diagnoses (recorded using International Classification of Diseases, Ninth Revision [ICD-9] and International Statistical Classification of Diseases and Related Health Problems, Tenth Revision [ICD-10] diagnosis codes), procedures (recorded using Current Procedural Terminology codes), laboratory results, microbiological culture results, and dispensed medications. Death date was ascertained from the VA Vital Status File.

To obtain operative variables, we linked patients’ records to the Veterans Affairs Surgical Quality Improvement Program (VASQIP), which included surgical data entered by nurses who reviewed records using standardized definitions.25 In general, surgical data at each VA center are identified on an 8-day cycle to ensure that data collection periods begin on different days of the week and are collected for up to 36 surgeries per cycle; thus, not all TKAs are entered into VASQIP. However, exclusion is random and is based on case volume at each center.26

Study Patients

Patients were eligible for inclusion in the study if they (1) underwent elective primary TKA (eTable 1 in Supplement 1) between October 1, 1999, and September 30, 2019, and (2) had at least 1 year of care in the VA prior to TKA. Primary TKAs can be accurately identified from the VA data set with a positive predictive value (PPV) of 95.2% (95% CI, 92%-99%).27 Patients were excluded if they underwent partial or revision TKA, underwent TKA for nonelective reasons (malignant neoplasm or fracture), or had a history of prior PJI or native septic joint arthritis of the knee (eTable 1 in Supplement 1).

We defined the index date as the date of hospital admission for primary TKA. The 12 months prior to admission represented the baseline period. Follow-up continued until the occurrence of one of the following: hospitalization with PJI, death, aseptic revision arthroplasty of the knee or contralateral TKA (since these procedures could affect PJI risk), or last VA visit before October 1, 2020 (to ensure patients had the opportunity for at least 1 year of follow-up after TKA).

Main Study Outcome

The primary end point was incident hospitalization with PJI. We classified patients as being hospitalized with PJI after primary TKA if they had (1) principal or contributory PJI ICD-9 or ICD-10 diagnosis codes at hospital discharge; (2) knee radiography within 90 days of PJI diagnosis; and (3) microbiological culture, arthrocentesis, or arthrotomy of the knee within 90 days of PJI diagnosis (eFigure 1 in Supplement 1). This algorithm accurately identified PJI with a PPV of 75.0% (95% CI, 64.1%-84.0%) in the ICD-9 era and 85.0% (95% CI, 75.3%-92.0%) in the ICD-10 era.28

The PJI date was defined as the date of hospital discharge with PJI diagnosis. We defined early PJI as occurring 3 months or less after primary TKA, delayed PJI as occurring between more than 3 months and 12 months or less after TKA, and late PJI as occurring more than 12 months after TKA. Patients stopped contributing to follow-up at their initial PJI.29

Data Collection

At the time of TKA, we collected data on age, sex, self-reported race and ethnicity, year of surgery, urban vs rural VA center,30 body mass index (BMI; calculated as weight in kilograms divided by height in meters squared), and tobacco use (current, former, or never smokers).31 Race and ethnicity data were collected as variables to contribute data on the representativeness of the study sample.

Validated algorithms based on 1 or more inpatient or 2 or more outpatient ICD-9 or ICD-10 diagnosis codes were used to identify baseline comorbidities, including alcohol use disorder (AUD),32 diabetes,33 heart failure,34,35 hypertension,36 hepatitis B virus infection, hepatitis C virus infection,37 HIV infection,38 and peripheral artery disease (PAD)39,40 (eTable 2 in Supplement 1). Autoimmune inflammatory arthritis was defined by 1 inpatient or 2 outpatient ICD-9 or ICD-10 diagnosis codes and required at least 1 rheumatologic clinic visit.41 Baseline hemoglobin and serum creatinine levels were collected from dates closest to, but within 365 days prior to TKA. Surgical factors extracted from VASQIP were recorded at the time of surgery and included anesthesia technique (general vs regional), American Society of Anesthesiology (ASA) Physical Status Classification System score for preanesthesia comorbidity (range: 1 [indicating a healthy patient] to 4 [indicating a patient with incapacitating severe disease that is a constant threat to life]),42 corticosteroid use within 30 days prior to TKA, intraoperative transfusion of packed red blood cells, and operative time (hours).

Results of microbiological cultures performed in the VA were documented in patients’ medical records as free text. All synovial fluid and tissue culture results from the knee within 90 days prior to PJI diagnosis date were obtained. Results were classified as follows: (1) culture-positive, defined as growth of 1 or more bacteria or fungal species from a culture; (2) culture-negative, defined as no growth of organisms; or (3) not collected or missing fluid or tissue culture. Culture-positive specimens were classified according to a hierarchy (eFigure 2 in Supplement 1).

Statistical Analysis

Among the overall cohort, we estimated incidence rates (events per 10 000 person-months) with 95% CIs of early, delayed, and late PJI. We used unadjusted Poisson regression to estimate the incidence rate ratio (IRR) with 95% CI of early and delayed PJI compared with late PJI. The standard Kaplan-Meier method with noninformative censoring was used to demonstrate the cumulative incidence of PJI at 3, 12, and 24 months. We measured the frequency of gram-positive, gram-negative, fungal, polymicrobial, and culture-negative PJI within each postoperative period and compared the frequencies of organisms using χ2 tests.

For the subgroup of patients with linked VASQIP data (referred to as the VASQIP cohort), we explored demographic, baseline clinical, and perioperative factors of early, delayed, and late PJI. The factors we evaluated included age, sex, urban vs rural center, BMI, AUD, hepatitis B virus infection, hepatitis C virus infection, heart failure, HIV infection, hypertension, PAD, autoimmune inflammatory arthritis, tobacco use, anemia (defined as hemoglobin <12.0 g/dL; to convert to milligrams per deciliter, multiply by 10.0), diabetes, chronic kidney disease (CKD; defined by estimated glomerular filtration rate <60 mL/min/1.73 m2, calculated using the Modified Diet in Renal Disease equation43), anesthesia technique, ASA score, corticosteroid use within 30 days prior to surgery, intraoperative transfusion of packed red blood cells, and prolonged operative time (≥2 hours). The degree of missing data for variables in the VASQIP cohort was low, with the highest being 2.9% for BMI data. As a result, we implemented a complete case analysis to identify factors for PJI in the VASQIP cohort.

We used a piecewise exponential parametric survival model fit through Poisson regression to estimate adjusted IRRs (with 95% CIs) of early, delayed, and late PJI associated with risk factors. Variables were retained in multivariable models if they were associated with PJI (statistical significance: P < .10) in univariable analysis and had statistical significance of P < .05 in multivariable analysis. We retained BMI in all models given its clinical importance44 and conflicting association with PJI in prior studies,12 regardless of statistical significance. We used the svy module in Stata to account for clustering by VA center and also adjusted for year of TKA (categorized in 5-year periods from 1999 to 2019).45 Data were analyzed between December 9, 2021, and September 18, 2023, using Stata, version 16.1 (StataCorp LLC).

Results

A total of 83 973 patients underwent elective primary TKA during the study period, of whom 4606 were excluded, leaving 79 367 patients in the overall cohort (Figure 1). This cohort comprised 4093 females (5.2%) and 75 274 males (94.8%) with a median (IQR) age of 65 (60-71) years, and most patients identified as non-Hispanic White individuals (73.5%) (Table 1). The median (IQR) duration of follow-up was 58.0 (22.4-108.2) months. Hypertension (86.0%), obesity (BMI of 30.0-39.9 and ≥40; 61.4%), current tobacco use (27.0%), and CKD (18.4%) were common. Among the overall cohort, 82.1% had linked VASQIP data; thus, the VASQIP cohort consisted of 65 188 patients. The characteristics of the VASQIP cohort were similar to those of the overall cohort (eTable 3 in Supplement 1).

Figure 1.. Patients Selected for Inclusion
Figure 1.. Patients Selected for Inclusion Analyses of incidence rates of early, delayed, and late prosthetic joint infection and microbiological studies were conducted in the overall cohort. Analyses of factors associated with prosthetic joint infection were evaluated in the Veterans Affairs Surgical Quality Improvement Program (VASQIP) complete case cohort. TKA indicates total knee arthroplasty; VA, US Department of Veterans Affairs.
Table 1.. Demographic, Baseline Clinical, and Perioperative Characteristics of the Overall Cohort (n = 79 367)
CharacteristicPatients, No. (%)Person-months of follow-up
Follow-up, median (IQR), mo58.0 (22.4-108.2)5 647 008
Age, y
Median (IQR)65 (60-71)NA
<6019 682 (24.8)1 606 579
60-6934 632 (43.6)2 394 754
≥7025 053 (31.6)1 645 674
Sex
Female4093 (5.2)274 224
Male75 274 (94.8)5 372 783
Race and ethnicitya
Hispanic3484 (4.4)243 900
Non-Hispanic Black10 642 (13.4)734 945
Non-Hispanic White58 307 (73.5)4 160 585
Otherb4181 (5.3)328 922
Unknown2753 (3.5)178 655
Year of TKA
1999-200412 599 (15.9)1 503 698
2005-200918 767 (23.6)1 742 641
2010-201423 475 (29.6)1 598 900
2015-201924 526 (30.9)801 767
Urban vs rural center
Rural2551 (3.2)205 065
Urban76 808 (96.8)5 441 347
Missing data8 (<0.1)NA
BMI
<25.05595 (7.0)396 666
25.0-29.922 114 (27.9)1 591 410
30.0-39.943 315 (54.6)2 984 255
≥405387 (6.8)412 417
Missing data2956 (3.7)NA
Tobacco use
Never or former56 851 (71.6)4 066 693
Current21 434 (27.0)1 516 583
Missing data1082 (1.4)NA
Baseline comorbiditiesc
AUD6436 (8.1)409 380
Heart failure3839 (4.9)220 875
Hypertension68 054 (86.0)4 767 230
PAD3815 (4.8)232 108
HIV infection133 (0.2)9246
HCV infection2509 (3.2)188 743
HBV infection120 (0.2)9655
Autoimmune inflammatory arthritis2226 (2.8)149 622
Anemia4009 (5.2)247 181
Missing data1546 (1.9)NA
Diabetes5602 (7.1)433 175
CKD14 403 (18.4)1 018 159
Missing data1186 (1.5)NA
Perioperative characteristics obtained from VASQIP cohortd
Corticosteroids use in prior 30 d1114 (1.7)72 682
Missing data1 (<0.1)NA
ASA score
1100 (0.2)10 098
215 314 (23.5)1 244 416
348 000 (73.6)3 206 147
41773 (2.7)105 895
Missing data1 (<0.1)NA
Anesthesia type
General41 383 (63.5)2 980 363
Regionale23 805 (36.5)1 586 289
Operative time, h
Median (IQR)2.0 (1.7-2.5)
<231 030 (47.6)2 149 297
≥234 156 (52.4)2 417 239
Missing data2 (<0.1)NA
Intraoperative transfusion of packed RBCs359 (0.6)32 226

Abbreviations: ASA, American Society of Anesthesiologists Physical Status Classification System (score range: 1 [indicating a healthy patient] to 4 [indicating a patient with incapacitating severe disease that is a constant threat to life]); AUD, alcohol use disorder; BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; HBV, hepatitis B virus; HCV, hepatitis C virus; ICD-9, International Classification of Diseases, Ninth Revision; ICD-10, International Statistical Classification of Diseases and Related Health Problems, Tenth Revision; NA, not applicable; PAD, peripheral artery disease; RBC, red blood cell; TKA, total knee arthroplasty; VASQIP, Veterans Affairs Surgical Quality Improvement Project.

a Race and ethnicity were self-reported.

b Included Asian, American Indian or Alaska native, Native Hawaiian or Other Pacific Islander, and multiracial.

c Comorbidities of AUD, heart failure, hypertension, diabetes, PAD, HIV infection, HBV infection, and HCV infection were defined by 1 inpatient or 2 outpatient ICD-9 or ICD-10 diagnosis codes. Autoimmune inflammatory arthritis was defined as 1 inpatient or 2 outpatient ICD-9 or ICD-10 diagnosis codes among those with at least 1 outpatient rheumatologic clinic visit during the baseline period. The ICD-9 or ICD-10 diagnosis codes used to identify comorbidities are provided in eTable 2 in Supplement 1. Anemia was defined as hemoglobin level less than 12 g/dL (to convert to grams per liter, multiply by 10.0), and CKD was defined as an eGFR less than 60 mL/min/1.73 m2; eGFR was calculated using the Modified Diet in Renal Disease equation: 175 × (serum creatinine)–1.154 × (age)–0.203 × (0.742, if female) × (1.212, if Black patient).

d Perioperative characteristics were available for only a subset of patients in the VASQIP cohort (n = 65 188). Percentages reflect this subset of patients.

e Included epidural, spinal, or monitored anesthesia.

Incidence Rates and Microbiological Studies of Early, Delayed, and Late PJIs

In the overall cohort, we identified 1599 incident PJIs (2.0%) across all 3 postoperative periods. We observed 627 early PJIs (39.2%) during 234 153 person-months of follow-up, yielding an incidence rate of 26.8 (95% CI, 24.8-29.0) events per 10 000 person-months. We identified 356 delayed PJIs (22.2%) during 654 498 person-months of follow-up, yielding an incidence rate of 5.4 (95% CI, 4.9-6.0) events per 10 000 person-months. There were 616 late PJIs (38.5%) during 4 758 356 person-months of follow-up, yielding an incidence rate of 1.3 (95% CI, 1.2-1.4) events per 10 000 person-months. The rate of PJI after primary TKA was significantly higher in the early (IRR, 20.7; 95% CI, 18.5-23.1) and delayed periods (IRR, 4.2; 95% CI, 3.7-4.8) vs the late period. The cumulative incidence of PJI was 0.80% (95% CI, 0.74%-0.86%) at 3 months, 1.28% (95% CI, 1.21%-1.36%) at 12 months, and 1.60% (95% CI, 1.51%-1.70%) at 24 months (Figure 2).

Figure 2.. Kaplan-Meier Failure Curves for Prosthetic Joint Infection (PJI) for the Overall Study Period and the First 24 Months After Primary Total Knee Arthroplasty
Figure 2.. Kaplan-Meier Failure Curves for Prosthetic Joint Infection (PJI) for the Overall Study Period and the First 24 Months After Primary Total Knee Arthroplasty

Among the 1599 incident PJIs, 126 (7.9%) did not have a microbiological culture specimen collected or the culture result was missing. Among 1473 patients with culture results, 459 (31.2%) had culture-negative results (Table 2). A culture-negative result was more common in late PJI compared with the combined early and delayed PJIs (200 [35.1%] vs 259 [28.7%]; P = .009). A total of 102 PJIs (6.9%) were polymicrobial, which was more frequently observed in the early postoperative period compared with the combined delayed and late postoperative periods (64 [11.4%] vs 38 [4.2%]; P < .001).

Table 2.. Organisms Isolated From Synovial Fluid or Operative Tissue Cultures of Prosthetic Joint Infections (PJIs) After Primary Total Knee Arthroplasty
OrganismNo. (%)
Overall (n = 1473)aEarly PJI (n = 560)Delayed PJI (n = 344)Late PJI (n = 569)
Culture-negative459 (31.2)171 (30.5)88 (25.6)200 (35.1)
Polymicrobialb102 (6.9)64 (11.4)18 (5.2)20 (3.5)
≥2 Gram-positive33 (2.4)20 (3.6)6 (1.7)7 (1.2)
≥2 Gram-negative24 (1.6)13 (2.3)9 (2.6)2 (0.4)
Gram-positive and gram-negative70 (4.8)44 (7.9)10 (2.9)16 (2.8)
Gram-positive infections710 (48.2)286 (51.1)177 (51.5)247 (43.4)
Staphylococcus species536 (36.4)245 (43.8)123 (35.8)168 (29.5)
Staphylococcus aureus489 (33.2)223 (39.8)108 (31.4)158 (27.8)
Coagulase-negative staphylococci286 (19.4)99 (17.7)73 (21.2)114 (20.0)
Streptococcus species122 (8.3)19 (3.4)40 (11.6)63 (11.1)
Enterococcus species65 (4.4)35 (6.3)16 (4.7)14 (2.5)
Anaerobes10 (0.7)4 (0.7)4 (1.2)2 (0.4)
Cutibacterium acnes6 (0.4)1 (0.2)3 (0.9)2 (0.4)
Otherc12 (0.8)4 (0.7)1 (0.3)7 (1.2)
Gram-negative infections164 (11.1)86 (15.4)31 (9.0)47 (8.3)
Pseudomonas aeruginosa40 (2.7)19 (3.4)8 (2.3)13 (2.3)
Escherichia coli34 (2.3)19 (3.4)8 (2.3)7 (1.2)
Klebsiella species32 (2.2)18 (3.2)4 (1.2)10 (1.8)
Enterobacter species36 (2.4)23 (4.1)10 (2.9)3 (0.5)
Serratia species12 (0.8)7 (1.3)2 (0.6)3 (0.5)
Anaerobes3 (0.2)1 (0.2)1 (0.3)1 (0.2)
Otherd38 (2.6)16 (2.9)8 (2.3)14 (2.5)
Fungal infectionse3 (0.2)0 (0.0)1 (0.3)2 (0.4)

a Among the 1599 PJIs in the overall cohort, microbiological data for 126 patients were missing or not collected. Consequently, 1473 PJIs with available microbiological study data were included in these results. Numbers may not sum to group totals or percentages as polymicrobial infections were possible.

b Polymicrobial defined by multiple genera of organisms identified. Numbers may not sum to group totals or percentages as polymicrobial infections with more than 1 organism were possible.

c Included Micrococcus species and diphtheroids.

d Included Salmonella, Pasteurella, Citrobacter, Morganella, Acinetobacter, Pasteurella, Burkholderia species, and gram-negative organisms not otherwise specified.

e All fungal infections were identified as Candida species.

Gram-positive infections were the most common across all postoperative periods (710 of 1473 infections [48.2%]), with Staphylococcus aureus being the most frequently identified (489 [33.2%]). Staphylococcus (245 [43.8%]) and Enterococcus (35 [6.3%]) species were more common in early PJIs, whereas Streptococcus species were more common in delayed (40 [11.6%]) and late (63 [11.1%]) PJIs (Table 2). Gram-negative organisms were observed in 164 infections (11.1%) and more commonly occurred in early PJIs than in combined delayed and late PJIs (86 [15.4%] vs 78 [8.5%]; P < .001). Pseudomonas aeruginosa was the most common gram-negative organism overall (40 [2.7%]), but Enterobacter species were the most frequently isolated in early PJI (23 [4.1%]).

Risk Factors for Early, Delayed, and Late PJI

Among the 65 188 patients in the VASQIP cohort, 61 701 (94.6%) had complete data on risk factors and so were included in the complete case analysis. Hepatitis C virus infection, PAD, and autoimmune inflammatory arthritis were associated with PJI across all postoperative periods (Table 3). Early PJI was also associated with current smoking, heart failure, hypertension, urban location, and prolonged operative time (≥2 hours). Autoimmune inflammatory arthritis was a robust factor of early PJI (adjusted IRR, 2.3; 95% CI, 1.5-3.5). Delayed PJI was also associated with BMI of 40 or higher, AUD, general anesthesia, anemia, and prolonged operative time (≥2 hours). Body mass index of 40 or higher was a robust factor of delayed PJI (adjusted IRR, 2.7; 95% CI, 1.5-4.9). Late PJI was also associated with AUD, heart failure, anemia, and younger age (<70 years). Anemia was a factor of late PJI (adjusted IRR, 2.2; 95% CI, 1.5-3.1). Factors asociated with PJI at any time after TKA are reported in eTable 4 in Supplement 1. Diabetes and chronic kidney disease were not associated factors.

Table 3.. Adjusted Incidence Rate Ratios (IRRs) of Prosthetic Joint Infections (PJIs) After Primary Total Knee Arthroplasty (TKA) in the Veterans Affairs Surgical Quality Improvement Project Complete Case Cohort (n = 61 701)
CharacteristicAdjusted IRR (95% CI)a
Early PJI: ≤3 moDelayed PJI: >3 and ≤12 moLate PJI: >12 mo
No. of PJIs513286500
Age, y
<601 [Reference]1 [Reference]1 [Reference]
60-690.9 (0.7-1.1)0.8 (0.6-1.1)0.9 (0.7-1.2)
≥700.8 (0.6-1.0)0.8 (0.6-1.2)0.6 (0.5-0.9)
Urban vs rural center
Rural1 [Reference]1 [Reference]1 [Reference]
Urban2.3 (1.1-4.9)1.3 (0.7-2.6)1.6 (0.8-3.3)
BMI
<25.01 [Reference]1 [Reference]1 [Reference]
25.0-29.90.8 (0.6-1.2)1.7 (0.9-2.9)0.8 (0.6-1.2)
30.0-39.90.8 (0.6-1.2)1.7 (1.0-2.7)0.9 (0.7-1.3)
≥400.9 (0.6-1.4)2.7 (1.5-4.9)1.3 (0.8-1.9)
AUD
No1 [Reference]1 [Reference]1 [Reference]
Yes1.1 (0.8-1.6)2.0 (1.4-2.9)1.5 (1.1-2.0)
Tobacco use
Former or never1 [Reference]1 [Reference]1 [Reference]
Current1.3 (1.0-1.6)0.9 (0.7-1.2)1.0 (0.8-1.3)
Heart failure
No1 [Reference]1 [Reference]1 [Reference]
Yes1.8 (1.3-2.4)0.8 (0.5-1.3)1.9 (1.3-2.8)
Hypertension
No1 [Reference]1 [Reference]1 [Reference]
Yes1.5 (1.2-2.0)1.2 (0.8-1.7)1.1 (0.8-1.4)
PAD
No1 [Reference]1 [Reference]1 [Reference]
Yes1.6 (1.1-2.2)1.6 (1.0-2.4)1.9 (1.3-2.7)
HCV infection
No1 [Reference]1 [Reference]1 [Reference]
Yes1.9 (1.3-2.8)1.7 (1.0-2.9)2.1 (1.5-3.0)
Autoimmune inflammatory arthritis
No1 [Reference]1 [Reference]1 [Reference]
Yes2.3 (1.5-3.5)1.7 (1.0-2.9)1.9 (1.2-3.0)
Anemia
No1 [Reference]1 [Reference]1 [Reference]
Yes1.4 (0.9-1.9)1.8 (1.3-2.7)2.2 (1.5-3.1)
Anesthesia type
Regional1 [Reference]1 [Reference]1 [Reference]
General1.1 (0.9-1.4)1.3 (1.0-1.6)1.1 (0.8-1.3)
Operative time, h
<21 [Reference]1 [Reference]1 [Reference]
≥21.3 (1.0-1.6)1.3 (1.0-1.7)1.1 (0.9-1.4)

Abbreviations: AUD, alcohol use disorder; BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); HCV, hepatitis C virus; PAD, peripheral artery disease; VA, US Department of Veterans Affairs.

a The IRR for each variable was adjusted for all other variables as well as by VA center and year of TKA (categorized in 5-year periods from 1999 to 2019).

Discussion

Using EHR data from the national VA system, we found that PJI was uncommon after TKA (2.0%), and the incidence rate was highest in the early and delayed periods (20.7 and 4.2 times higher, respectively) compared with the late period. Staphylococcus and Enterococcus species were more commonly isolated from early PJIs, whereas Streptococcus species were more commonly isolated from delayed or late PJIs. Gram-negative infections were more common in the early postoperative period. Hepatitis C infection, PAD, and autoimmune inflammatory arthritis were associated with PJI across all postoperative periods.

This study demonstrated that the rate of PJI was highest within the initial 3 months after primary TKA, and the cumulative incidence at 24 months was 1.60%. These results are consistent with the findings of a cohort study of 69 663 US Medicare beneficiaries in which the cumulative incidence of PJI diagnoses was 1.55% within the first 24 months after elective TKA.7 Since the incidence of PJI was highest soon after primary TKA, development and testing of interventions to modify risk factors for early PJI should be prioritized.

Few studies have examined the microbiological characteristics of PJI after TKA. In a cross-sectional study of 1651 patients who had a primary or revision TKA or total hip arthroplasty at Mayo Clinic (2010-2019) and who were observed for PJI, coagulase-negative staphylococci were the most common organisms isolated from cultures.46 In a separate cohort study of 231 patients with hip or knee PJI from a single hospital in China (2006-2015), coagulase-negative staphylococci were also the most common organisms isolated.47 The finding of the present study that Staphylococcus aureus was the most frequently identified causative organism may be due to inclusion of only primary TKAs, whereas previous studies included PJIs of the knee and hip as well as primary and revision arthroplasties.

Gram-negative organisms were isolated from 11.1% of all PJIs and 15.4% of early PJIs. The gram-negative organisms identified from the PJIs in this study, including the early infections, are typically resistant to first-generation cephalosporins. Given these findings, empirical gram-negative antibiotic therapy, in addition to the usual gram-positive coverage, should be considered during suspected early PJI. Cefazolin, the recommended antibiotic for perioperative TKA prophylaxis, would not provide adequate protection against most of the gram-negative organisms identified in this study.48 To date, the only intervention associated with a decreased rate of PJI is perioperative antibiotic prophylaxis.49,50 Future studies should examine the antimicrobial susceptibility profiles of gram-negative infections that were isolated during early PJI to determine the extent to which broader perioperative gram-negative antibiotic prophylaxis is warranted.

We explored factors of early, delayed, and late PJI. Factors associated with inflammation and tissue hypoxia were associated with PJI in each postoperative period. Hepatitis C virus infection and autoimmune inflammatory arthritis were associated with PJI across all postoperative periods and contributed to systemic inflammation that compromised wound healing.51,52,53 Peripheral artery disease was also a significant factor in tissue hypoxia and compromised skin or soft tissue integrity54 across the 3 postoperative periods. Early PJI was associated with factors that promoted tissue hypoxia (smoking, heart failure, prolonged operative time, and hypertension).20,55,56,57,58 Delayed PJI was associated with factors contributing to tissue hypoxia (prolonged operative time, general anesthesia, and anemia) and impaired wound healing and falls (morbid obesity and AUD).58,59,60,61,62,63 Additional factors of late PJI were associated with impaired wound healing and falls (AUD, heart failure, and anemia).56,59,60,61,62 Prognostic models should be developed to classify patients according to risk of PJI, which could help guide informed consent, surgical planning, and postoperative monitoring.

We observed a reduced rate of late infections among patients 70 years or older compared with those younger than 60 years, which is consistent with findings of large European and Canadian studies and may reflect a tendency to avoid invasive diagnostic testing and workup for painful joints in patients 70 years or older.17,64 Chronic kidney disease and diabetes were not associated with PJI during any period after TKA. The lack of association between CKD and PJI has been reported in a meta-analysis.65 Diabetes has been a factor of PJI in most prior studies.15,16,17,66 The lack of a similar association in the present study might be due to the definition of diabetes used. No pattern was observed between BMI and PJI risk, making the clinical significance of the association between BMI of 40 or higher and risk of PJI in the delayed postoperative period unclear. The role that obesity plays in postoperative infection risk has been inconsistent, with some studies reporting no association2,14,21,67 and other studies observing an increased risk.12,17,68 Results of the present study suggest that patients with obesity or CKD should not be restricted from undergoing TKA because of concerns about increased PJI risk.

Strengths and Limitations

Strengths of this study included the large sample of veterans undergoing primary TKAs, access to national EHR data that included microbiological culture results, and linkage to surgical-specific data. This study also had several potential limitations. First, misclassification of outcomes was possible, but use of a validated PJI end point minimized this risk. It is possible that some PJI outcomes were missed in patients who presented to a non-VA hospital. Second, some culture results could have included organisms that were contaminants. Third, residual confounding by unmeasured factors, such as perioperative antibiotic prophylaxis and chronic skin conditions, could have affected risk factor analyses. Moreover, we evaluated comorbidities as binary (present or absent), without accounting for illness severity, and accounted only for baseline comorbidities or surgical characteristics without evaluating time-varying factors. Fourth, the sample consisted of mostly males with a high prevalence of comorbid conditions. Results may have less generalizability to females and the overall population receiving TKAs.

Conclusions

In this cohort study of patients who underwent primary TKA in the VA system, incidence rates of PJI occurring in the early and delayed post-TKA periods were higher than incidence of PJI occurring later. Staphylococcus and Enterococcus species were more commonly isolated from PJIs occurring in the early postoperative period, whereas Streptococcus species were more commonly isolated from PJIs occurring in the delayed or late postoperative periods. Gram-negative infections were more commonly observed in the early postoperative period, which suggests that empirical gram-negative antibiotic therapy should be considered along with the usual gram-positive coverage for early PJIs. Additionally, differences in demographic, clinical, and perioperative factors associated with early, delayed, and late PJIs were identified. These findings have implications for postoperative antibiotic use, stratification of PJI risk according to post-TKA periods, and PJI risk factor modification.

Declarations

Supplement 1.: eTable 1. Diagnosis and Procedures Codes Used to Define Inclusion and Exclusion Criteria, Prosthetic Joint Infection Outcomes, and Censoring Events — available with the version of record.

Supplement 2.: Data Sharing Statement — available with the version of record.

Sources

References

68 references, in the article's own order. Citation numbers in the text are this list's numbers (68 records are cited).

  1. Total knee arthroplasty volume, utilization, and outcomes among Medicare beneficiaries, 1991-2010

    Cram P, Lu X, Kates SL, Singh JA, Li Y, Wolf BR

    2012JAMA 308(12):1227–1236 · PMID 23011713doi:10.1001/2012.jama.11153

  2. Rates of total joint replacement in the United States: future projections to 2020-2040 using the National Inpatient Sample

    Singh JA, Yu S, Chen L, Cleveland JD

    2019J Rheumatol 46(9):1134–1140 · PMID 30988126doi:10.3899/jrheum.170990

  3. Prosthetic joint infection update

    Beam E, Osmon D

    2018Infect Dis Clin North Am 32(4):843–859 · PMID 30241717doi:10.1016/j.idc.2018.06.005

  4. Patient and surgical factors affecting procedure duration and revision risk due to deep infection in primary total knee arthroplasty

    Badawy M, Espehaug B, Fenstad AM, et al.

    2017BMC Musculoskelet Disord 18(1):544 · PMID 29268748doi:10.1186/s12891-017-1915-4

  5. Trends in periprosthetic knee infection and associated costs: a population-based study using national data

    Brochin R, Poeran J, Vig KS, et al.

    2021J Knee Surg 34(10):1110–1119 · PMID 32131096doi:10.1055/s-0040-1701516

  6. Are we winning or losing the battle with periprosthetic joint infection: trends in periprosthetic joint infection and mortality risk for the Medicare population

    Kurtz SM, Lau EC, Son MS, Chang ET, Zimmerli W, Parvizi J

    2018J Arthroplasty 33(10):3238–3245 · PMID 29914821doi:10.1016/j.arth.2018.05.042

  7. Prosthetic joint infection risk after TKA in the Medicare population

    Kurtz SM, Ong KL, Lau E, Bozic KJ, Berry D, Parvizi J

    2010Clin Orthop Relat Res 468(1):52–56 · PMID 19669386doi:10.1007/s11999-009-1013-5

  8. Quantifying the burden of revision total joint arthroplasty for periprosthetic infection

    Kamath AF, Ong KL, Lau E, et al.

    2015J Arthroplasty 30(9):1492–1497 · PMID 25865815doi:10.1016/j.arth.2015.03.035

  9. Incidence and risk factors of surgical site infection after total knee arthroplasty: results of a retrospective cohort study

    Baier C, Adelmund S, Schwab F, et al.

    2019Am J Infect Control 47(10):1270–1272 · PMID 31153712doi:10.1016/j.ajic.2019.04.010

  10. Surgical site infection after primary hip and knee arthroplasty: a cohort study using a hospital database

    Grammatico-Guillon L, Baron S, Rosset P, et al.

    2015Infect Control Hosp Epidemiol 36(10):1198–1207 · PMID 26154882doi:10.1017/ice.2015.148

  11. Infection burden for hip and knee arthroplasty in the United States

    Kurtz SM, Lau E, Schmier J, Ong KL, Zhao K, Parvizi J

    2008J Arthroplasty 23(7):984–991 · PMID 18534466doi:10.1016/j.arth.2007.10.017

  12. Risk factors associated with deep surgical site infections after primary total knee arthroplasty: an analysis of 56,216 knees

    Namba RS, Inacio MC, Paxton EW

    2013J Bone Joint Surg Am 95(9):775–782 · PMID 23636183doi:10.2106/JBJS.L.00211

  13. Risk factors for surgical site infection following total joint arthroplasty

    Rasouli MR, Restrepo C, Maltenfort MG, Purtill JJ, Parvizi J

    2014J Bone Joint Surg Am 96(18):e158 · PMID 25232088doi:10.2106/JBJS.M.01363

  14. Risk factors for prosthetic joint infection: case-control study

    Berbari EF, Hanssen AD, Duffy MC, et al.

    1998Clin Infect Dis 27(5):1247–1254 · PMID 9827278doi:10.1086/514991

  15. Risk factors for periprosthetic joint infection after total knee arthroplasty

    Blanco JF, Díaz A, Melchor FR, da Casa C, Pescador D

    2020Arch Orthop Trauma Surg 140(2):239–245 · PMID 31707484doi:10.1007/s00402-019-03304-6

  16. Patient-related medical risk factors for periprosthetic joint infection of the hip and knee

    Eka A, Chen AF

    2015Ann Transl Med 3(16):233 · PMID 26539450doi:10.3978/j.issn.2305-5839.2015.09.26

  17. Risk factors associated with revision for prosthetic joint infection following knee replacement: an observational cohort study from England and Wales

    Lenguerrand E, Whitehouse MR, Beswick AD, et al., National Joint Registry for England, Wales, Northern Ireland and the Isle of Man

    2019Lancet Infect Dis 19(6):589–600 · PMID 31005559doi:10.1016/S1473-3099(18)30755-2

  18. Sarcopenia as a risk factor for prosthetic infection after total hip or knee arthroplasty

    Babu JM, Kalagara S, Durand W, Antoci V, Deren ME, Cohen E

    2019J Arthroplasty 34(1):116–122 · PMID 30337254doi:10.1016/j.arth.2018.09.037

  19. Is hemoglobin A1c or perioperative hyperglycemia predictive of periprosthetic joint infection or death following primary total joint arthroplasty?

    Chrastil J, Anderson MB, Stevens V, Anand R, Peters CL, Pelt CE

    2015J Arthroplasty 30(7):1197–1202 · PMID 25697889doi:10.1016/j.arth.2015.01.040

  20. Is there an association between smoking status and prosthetic joint infection after primary total joint arthroplasty?

    Gonzalez AI, Luime JJ, Uçkay I, Hannouche D, Hoffmeyer P, Lübbeke A

    2018J Arthroplasty 33(7):2218–2224 · PMID 29573917doi:10.1016/j.arth.2018.02.069

  21. Low albumin levels, more than morbid obesity, are associated with complications after TKA

    Nelson CL, Elkassabany NM, Kamath AF, Liu J

    2015Clin Orthop Relat Res 473(10):3163–3172 · PMID 25995174doi:10.1007/s11999-015-4333-7

  22. Accuracy of different diagnostic tests for early, delayed and late prosthetic joint infection

    Fernández-Sampedro M, Fariñas-Alvarez C, Garces-Zarzalejo C, et al.

    2017BMC Infect Dis 17(1):592 · PMID 28841913doi:10.1186/s12879-017-2693-1

  23. Microbiologic epidemiology depending on time to occurrence of prosthetic joint infection: a prospective cohort study

    Triffault-Fillit C, Ferry T, Laurent F, et al., Lyon BJI Study Group

    2019Clin Microbiol Infect 25(3):353–358 · PMID 29803842doi:10.1016/j.cmi.2018.04.035

  24. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies

    von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP, STROBE Initiative

    2007Ann Intern Med 147(8):573–577 · PMID 17938396doi:10.7326/0003-4819-147-8-200710160-00010

  25. The Department of Veterans Affairs’ NSQIP: the first national, validated, outcome-based, risk-adjusted, and peer-controlled program for the measurement and enhancement of the quality of surgical care

    Khuri SF, Daley J, Henderson W, et al., National VA Surgical Quality Improvement Program

    1998Ann Surg 228(4):491–507 · PMID 9790339doi:10.1097/00000658-199810000-00006

  26. Practical guide to surgical data sets: Veterans Affairs Surgical Quality Improvement Program (VASQIP)

    Massarweh NN, Kaji AH, Itani KMF

    2018JAMA Surg 153(8):768–769 · PMID 29617543doi:10.1001/jamasurg.2018.0504

  27. Accuracy of VA databases for diagnoses of knee replacement and hip replacement

    Singh JA, Ayub S

    2010Osteoarthritis Cartilage 18(12):1639–1642 · PMID 20950694doi:10.1016/j.joca.2010.10.003

  28. Development and validation of case-finding algorithms to identify prosthetic joint infections after total knee arthroplasty in Veterans Health Administration data

    Weinstein EJ, Stephens-Shields A, Loabile B, et al.

    2021Pharmacoepidemiol Drug Saf 30(9):1184–1191 · PMID 34170057doi:10.1002/pds.5316

  29. Prosthetic joint infection

    Tande AJ, Patel R

    2014Clin Microbiol Rev 27(2):302–345 · PMID 24696437doi:10.1128/CMR.00111-13

  30. US Department of Veterans Affairs. Rural veterans. Accessed May 18, 2022. https://www.ruralhealth.va.gov/aboutus/ruralvets.asp#def

    US Department of Veterans Affairs

  31. Validating smoking data from the Veteran’s Affairs Health Factors dataset, an electronic data source

    McGinnis KA, Brandt CA, Skanderson M, et al.

    2011Nicotine Tob Res 13(12):1233–1239 · PMID 21911825doi:10.1093/ntr/ntr206

  32. Psychiatric and neurocognitive disorders among HIV-positive and negative veterans in care: Veterans Aging Cohort Five-Site Study

    Justice AC, McGinnis KA, Atkinson JH, et al., Veterans Aging Cohort 5-Site Study Project Team

    2004AIDS 18(suppl 1):S49–S59 · PMID 15075498doi:10.1097/00002030-200401001-00008

  33. HIV infection and the risk of diabetes mellitus

    Butt AA, McGinnis K, Rodriguez-Barradas MC, et al., Veterans Aging Cohort Study

    2009AIDS 23(10):1227–1234 · PMID 19444074doi:10.1097/QAD.0b013e32832bd7af

  34. A systematic review of validated methods for identifying heart failure using administrative data

    Saczynski JS, Andrade SE, Harrold LR, et al.

    2012Pharmacoepidemiol Drug Saf 21(suppl 1):129–140 · PMID 22262599doi:10.1002/pds.2313

  35. Association between HIV infection and the risk of heart failure with reduced ejection fraction and preserved ejection fraction in the antiretroviral therapy era: results from the Veterans Aging Cohort Study

    Freiberg MS, Chang CH, Skanderson M, et al.

    2017JAMA Cardiol 2(5):536–546 · PMID 28384660doi:10.1001/jamacardio.2017.0264

  36. The use of claims data algorithms to recruit eligible participants into clinical trials

    Tamariz L, Palacio A, Denizard J, Schulman Y, Contreras G

    2015Am J Manag Care 21(2):e114–e118 · PMID 25880486

  37. The validity of viral hepatitis and chronic liver disease diagnoses in Veterans Affairs administrative databases

    Kramer JR, Davila JA, Miller ED, Richardson P, Giordano TP, El-Serag HB

    2008Aliment Pharmacol Ther 27(3):274–282 · PMID 17996017doi:10.1111/j.1365-2036.2007.03572.x

  38. Development and verification of a “virtual” cohort using the National VA Health Information System

    Fultz SL, Skanderson M, Mole LA, et al.

    2006Med Care 44(8):S25–S30 · PMID 16849965doi:10.1097/01.mlr.0000223670.00890.74

  39. Association of human immunodeficiency virus infection and risk of peripheral artery disease

    Beckman JA, Duncan MS, Alcorn CW, et al.

    2018Circulation 138(3):255–265 · PMID 29535090doi:10.1161/CIRCULATIONAHA.117.032647

  40. Microvascular disease, peripheral artery disease, and amputation

    Beckman JA, Duncan MS, Damrauer SM, et al.

    2019Circulation 140(6):449–458 · PMID 31280589doi:10.1161/CIRCULATIONAHA.119.040672

  41. Identification of rheumatoid arthritis patients using an administrative database: a Veterans Affairs study

    Ng B, Aslam F, Petersen NJ, Yu HJ, Suarez-Almazor ME

    2012Arthritis Care Res (Hoboken) 64(10):1490–1496 · PMID 22623324doi:10.1002/acr.21736

  42. A review of ASA physical status–historical perspectives and modern developments

    Mayhew D, Mendonca V, Murthy BVS

    2019Anaesthesia 74(3):373–379 · PMID 30648259doi:10.1111/anae.14569

  43. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation

    Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D, Modification of Diet in Renal Disease Study Group

    1999Ann Intern Med 130(6):461–470 · PMID 10075613doi:10.7326/0003-4819-130-6-199903160-00002

  44. What is the impact of body mass index cutoffs on total knee arthroplasty complications?

    DeMik DE, Muffly SA, Carender CN, Glass NA, Brown TS, Bedard NA

    2022J Arthroplasty 37(4):683–687.e1 · PMID 34954020doi:10.1016/j.arth.2021.12.024

  45. Stata Survey Data Reference Manual. Release 17

    StataCorp LLC

    2021

  46. Microbiology of hip and knee periprosthetic joint infections: a database study

    Tai DBG, Patel R, Abdel MP, Berbari EF, Tande AJ

    2022Clin Microbiol Infect 28(2):255–259 · PMID 34129907doi:10.1016/j.cmi.2021.06.006

  47. Trends in microbiological profiles and antibiotic resistance in periprosthetic joint infections

    Hu L, Fu J, Zhou Y, et al.

    2021J Int Med Res 49(3):3000605211002784 · PMID 33787371doi:10.1177/03000605211002784

  48. Clinical practice guidelines for antimicrobial prophylaxis in surgery

    Bratzler DW, Dellinger EP, Olsen KM, et al., American Society of Health-System Pharmacists (ASHP), Infectious Diseases Society of America (IDSA), Surgical Infection Society (SIS), Society for Healthcare Epidemiology of America (SHEA)

    2013Surg Infect (Larchmt) 14(1):73–156 · PMID 23461695doi:10.1089/sur.2013.9999

  49. Prevention and management of infection after total joint replacement

    Gillespie WJ

    1997Clin Infect Dis 25(6):1310–1317 · PMID 9431369doi:10.1086/516134

  50. Antibiotic prophylaxis in orthopedic prosthetic surgery

    Marculescu CE, Osmon DR

    2005Infect Dis Clin North Am 19(4):931–946 · PMID 16297740doi:10.1016/j.idc.2005.07.002

  51. The preoperative management of Hepatitis C may improve the outcome after total knee arthroplasty

    Schwarzkopf R, Novikov D, Anoushiravani AA, et al.

    2019Bone Joint J 101-B(6):667–674 · PMID 31154838doi:10.1302/0301-620X.101B6.BJJ-2018-0723.R3

  52. Rheumatoid arthritis

    Smolen JS, Aletaha D, McInnes IB

    2016Lancet 388(10055):2023–2038 · PMID 27156434doi:10.1016/S0140-6736(16)30173-8

  53. Rheumatoid arthritis, disease modifying agents, and periprosthetic joint infection: what does a joint surgeon need to know?

    Yeganeh MH, Kheir MM, Shahi A, Parvizi J

    2018J Arthroplasty 33(4):1258–1264 · PMID 29325724doi:10.1016/j.arth.2017.11.031

  54. Managing the patient with peripheral vascular disease before total knee arthroplasty surgery

    Fraval A, Hozack WJ

    2023Orthop Clin North Am 54(3):259–267 · PMID 37271554doi:10.1016/j.ocl.2023.02.011

  55. Hypertensive patients show delayed wound healing following total hip arthroplasty

    Ahmed AA, Mooar PA, Kleiner M, Torg JS, Miyamoto CT

    2011PLoS One 6(8):e23224 · PMID 21853091doi:10.1371/journal.pone.0023224

  56. Perioperative outcomes and complications in patients with heart failure following total knee arthroplasty

    Curtis GL, Newman JM, George J, Klika AK, Barsoum WK, Higuera CA

    2018J Arthroplasty 33(1):36–40 · PMID 28844769doi:10.1016/j.arth.2017.07.043

  57. The musculoskeletal effects of smoking

    Porter SE, Hanley EN

    2001J Am Acad Orthop Surg 9(1):9–17 · PMID 11174159doi:10.5435/00124635-200101000-00002

  58. Longer operative time results in a higher rate of subsequent periprosthetic joint infection in patients undergoing primary joint arthroplasty

    Wang Q, Goswami K, Shohat N, Aalirezaie A, Manrique J, Parvizi J

    2019J Arthroplasty 34(5):947–953 · PMID 30765229doi:10.1016/j.arth.2019.01.027

  59. Preoperative anemia in total joint arthroplasty: is it associated with periprosthetic joint infection?

    Greenky M, Gandhi K, Pulido L, Restrepo C, Parvizi J

    2012Clin Orthop Relat Res 470(10):2695–2701 · PMID 22773393doi:10.1007/s11999-012-2435-z

  60. Alcohol abuse: a risk factor for surgical wound infections?

    Rantala A, Lehtonen OP, Niinikoski J

    1997Am J Infect Control 25(5):381–386 · PMID 9343620doi:10.1016/S0196-6553(97)90082-1

  61. Iron deficiency anemia is associated with increased early postoperative surgical and medical complications following total hip arthroplasty

    Sequeira SB, Quinlan ND, Althoff AD, Werner BC

    2021J Arthroplasty 36(3):1023–1028 · PMID 33067093doi:10.1016/j.arth.2020.09.043

  62. Alcohol consumption increases post-operative infection but not mortality: a systematic review and meta-analysis

    Shabanzadeh DM, Sørensen LT

    2015Surg Infect (Larchmt) 16(6):657–668 · PMID 26244748doi:10.1089/sur.2015.009

  63. The effects of epidural and general anesthesia on tissue oxygenation

    Treschan TA, Taguchi A, Ali SZ, et al.

    2003Anesth Analg 96(6):1553–1557 · PMID 12760974doi:10.1213/01.ANE.0000063824.43113.DB

  64. Incidence and predictors of prosthetic joint infection following primary total knee arthroplasty: a 15-year population-based cohort study

    McMaster Arthroplasty Collaborative (MAC)

    2022J Arthroplasty 37(2):367–372.e1 · PMID 34678445doi:10.1016/j.arth.2021.10.006

  65. Effect of chronic kidney disease on outcomes of total joint arthroplasty: a meta-analysis

    Kim CW, Kim HJ, Lee CR, Wang L, Rhee SJ

    2020Knee Surg Relat Res 32(1):12 · PMID 32660587doi:10.1186/s43019-020-0029-8

  66. Patient-related risk factors for periprosthetic joint infection after total joint arthroplasty: a systematic review and meta-analysis

    Kunutsor SK, Whitehouse MR, Blom AW, Beswick AD, INFORM Team

    2016PLoS One 11(3):e0150866 · PMID 26938768doi:10.1371/journal.pone.0150866

  67. Complication rates and resource utilization after total hip and knee arthroplasty stratified by body mass index

    Turcotte J, Kelly M, Aja J, King P, MacDonald J

    2021J Orthop 24:111–120 · PMID 33679036doi:10.1016/j.jor.2021.02.024

  68. Greater risk of all-cause revisions and complications for obese patients in 3 106 381 total knee arthroplasties: a meta-analysis and systematic review

    Onggo JR, Ang JJM, Onggo JD, de Steiger R, Hau R

    2021ANZ J Surg 91(11):2308–2321 · PMID 34405518doi:10.1111/ans.17138

Article record

The record

Status

Living reprint · journal article Journal version available

Status
Version of record: JAMA Network Open 2023
Journal version
JAMA Network Open (2023) · doi:10.1001/jamanetworkopen.2023.40457
Confirmed
2023-10-31
Licence
CC BY 4.0

This reprint has no DOI of its own: cite the journal article.

The journal's version is the version of record. This page re-typesets it under the article's licence; where they differ, the journal's version is authoritative.

Version history

  1. Published in JAMA Network Open2023-10-31
  2. Living reprint on Literature Decoded2026-10-01
  3. Reprinted here2026-10-01
  4. This HTML2026-10-01

Cite this article

Citation
Weinstein EJ, Stephens-Shields AJ, Newcomb CW, Silibovsky R, Nelson CL, O’Donnell JA, et al. Incidence, Microbiological Studies, and Factors Associated With Prosthetic Joint Infection After Total Knee Arthroplasty. JAMA Netw Open. 2023;6(10):e2340457. doi:10.1001/jamanetworkopen.2023.40457
BibTeX
@article{Weinstein2023Incidence,
  title   = {Incidence, Microbiological Studies, and Factors Associated With Prosthetic Joint Infection After Total Knee Arthroplasty},
  author  = {Erica J. Weinstein and Alisa J. Stephens-Shields and Craig W. Newcomb and Randi Silibovsky and Charles L. Nelson and Judith A. O’Donnell and Laurel J. Glaser and Evelyn Hsieh and Jennifer S. Hanberg and Janet P. Tate and Kathleen M. Akgün and Joseph T. King and Vincent Lo Re},
  journal = {JAMA Network Open},
  year    = {2023},
  volume  = {6},
  number  = {10},
  pages   = {e2340457},
  doi     = {10.1001/jamanetworkopen.2023.40457},
  pmid    = {37906194}
}
RIS
TY  - JOUR
TI  - Incidence, Microbiological Studies, and Factors Associated With Prosthetic Joint Infection After Total Knee Arthroplasty
AU  - Erica J. Weinstein
AU  - Alisa J. Stephens-Shields
AU  - Craig W. Newcomb
AU  - Randi Silibovsky
AU  - Charles L. Nelson
AU  - Judith A. O’Donnell
AU  - Laurel J. Glaser
AU  - Evelyn Hsieh
AU  - Jennifer S. Hanberg
AU  - Janet P. Tate
AU  - Kathleen M. Akgün
AU  - Joseph T. King
AU  - Vincent Lo Re
JO  - JAMA Network Open
PY  - 2023
VL  - 6
IS  - 10
SP  - e2340457
DO  - 10.1001/jamanetworkopen.2023.40457
SN  - 2574-3805
UR  - https://doi.org/10.1001/jamanetworkopen.2023.40457
ER  -

Cite the original article (the version of record), not this reprint.

Take it with you

One self-contained HTML file: the text, data and figures, readable offline.

Disclaimer and limitations

Educational summary of research findings; not medical advice. Discuss care decisions with a qualified clinician.

The findings apply to the included study populations and may not generalise to every person or setting.

Licence and attribution

This page reproduces Incidence, Microbiological Studies, and Factors Associated With Prosthetic Joint Infection After Total Knee Arthroplasty by Erica J. Weinstein, Alisa J. Stephens-Shields, Craig W. Newcomb, Randi Silibovsky, Charles L. Nelson, Judith A. O’Donnell, Laurel J. Glaser, Evelyn Hsieh, Jennifer S. Hanberg, Janet P. Tate, Kathleen M. Akgün, Joseph T. King, Vincent Lo Re, first published in JAMA Network Open 2023;6(10):e2340457, doi:10.1001/jamanetworkopen.2023.40457, PMID 37906194, PMC10618849. Copyright 2023 Weinstein EJ et al. JAMA Network Open. It is used under the CC BY 4.0 licence.

Changes made:

The words, data, figures and conclusions are the authors’. Literature Decoded is not the publisher of record and does not claim endorsement by the authors or the journal. The version of record is the journal’s; cite it.

Licence statement in the article: “This is an open access article distributed under the terms of the CC-BY License.”

Are you an author of this article? Claim it or ask us to take it down.

Cite this article

Living reprint · journal article