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

Abstract

ImportanceDespite 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.ObjectivesTo 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 ParticipantsThis 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.ExposuresPrimary TKA as well as demographic, clinical, and perioperative factors.Main Outcomes and MeasuresIncident 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.ResultsThe 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 RelevanceThis 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.

Risk equations for prosthetic joint infections (PJIs) in UK: a retrospective study using the Clinical Practice Research Datalink (CPRD) AURUM and GOLD databases

“…The observed PJI incidence in this study (~0.7% after 5 years) is lower than that sometimes reported. One possible explanation for this was the use of a specific ICD-10 code for the identification of PJI (T84.5), while in our studies that reported higher PJI rates (about 2.5% over 5 years), additional ICD-10 codes were employed to identify PJI, for example, T81.4 (infection following a procedure) and T84.7 (infection and inflammatory reaction due to other internal orthopaedic prosthetic devices, implants and grafts) 33 ; however, such codes appear less specific as, for example, T81.4 may lead to considering a superficial infection of the surgical wound as a PJI, while M00 corresponds to a diagnosis of pyogenic arthritis. 5 When using only the presence of ICD-10 code T84.5, the PJI incidence at 5 years observed in this study is similar to the 0.89% previously reported.…”

Infecção de sítio cirúrgico em pacientes submetidos a artroplastias de quadril e joelho

“…entre os pacientes considerados de alto risco, como os diabéticos e hipertensos, para o desenvolvimento de ISC 16,20 . Adicionalmente, cita-se a relevância de mensurar, respectivamente, indicadores relacionados ao tempo dos processos operatórios e de internação, como a duração cirúrgica e de permanência hospitalar, pois estudos mostram que tempo operatório prolongado (≥2 horas) 21 e de internação hospitalar acima de cinco dias 22 são fatores relacionados à ocorrência de ISC, corroborando com os achados da presente pesquisa.…”

Upcoming evidence in clinical practice of two-stage revision arthroplasty for prosthetic joint infection

“…Author details 1 Service of Infectious Diseases, Cardarelli Hospital, Via A. Cardarelli 9, 80131 Naples, Italy. 2 Department of Public Health, Orthopedic Unit, "Federico II" University, Naples, Italy. 3 Unit of Infectious Diseases, Department of Medicine, Surgery and Dentistry, "Scuola Medica Salernitana", University of Salerno, Baronissi, Italy.…”

“…Prosthetic joint infection (PJI) is a serious complication of total joint arthroplasty. Data obtained from the US Department of Veterans Affairs on a sample of around 80,000 patients undergoing primary total knee arthroplasty identified a percentage of patients with PJI approaching 2%, with the highest number of PJI cases diagnosed within 24 months from primary knee arthroplasty [ 2 ]. Further data obtained by a meta-analysis including articles investigating the incidence of PJI after primary hip arthroplasty highlight an incidence approaching 1%, with the highest number of cases among people aged ≥ 70 years and a great difference in terms of incidence among different countries [ 3 ].…”