Incidence and Microbial Profiles of Periprosthetic Joint Infection After Total Knee Arthroplasty in Obese vs Non-Obese Patients

Overview

Importance: After a primary total knee arthroplasty (TKA), we anticipate that obese individuals are more likely to experience a of periprosthetic joint infection (PJI). Knowing the microbial profile in obese patients can potentially help in choosing the proper prophylactic measures for this group as well as tailoring the empirical antibiotics in relation to PJI in obese patients. Objective: To compare incidence of PJI revisions within 2 years after TKA in obese vs non-obese OA patients. Secondarily: To compare microbial profiles of infections in these groups in two time windows: early (≤90 days) vs late (91-730 days) and incidence of revisions due to all causes within 2 years.Design and Setting: We will include patients having primary knee arthroplasty using Danish national registers. Patients will be stratified into anthropometric groups, referred to as exposed and unexposed to obesity, based on their baseline Body Mass Index (BMI). Participants included in the analysis population will be followed up for 2 years, until first revision, death or migration whichever comes first. Participants: We will include adult patients with available weight and height data with primary/idiopathic or secondary (due to meniscus or cruciate ligament lesion) OA who received primary TKA in the period from 2011-01-01 and 2021-02-28. Patients will be identified from the Danish knee arthroplasty register. Exposure and Comparator: The cohort will be divided into obese (exposed), defined as body mass index (BMI) ≥ 30 kg/m2; and non-obese (unexposed), defined as BMI \< 30 kg/m2. Main Outcomes and Measures: The primary outcome will be revision due to prosthetic joint infection (PJI) within 730 days following TKA. Secondary endpoints will then be to examine revision due to all causes within 730 days following TKA and type of microbial infection between obese and non-obese in the first 90 days and the period from 91 days to 730 days following TKA. Planned Statistical Analyses: We will use descriptive statistics to summarize the baseline characteristics of the two groups. Hazard ratios with corresponding two-sided 95% confidence intervals (95%CIs) for experiencing the outcome will be estimated using a Cox proportional hazards regression model. We will fit both unadjusted (crude) model and a propensity score adjusted model calculated based on age, sex, highest completed education, household income, comedications, and Elixhauser Comorbidity Measure (ECM), all collected up to the day of surgery. Stratified analyses will categorize PJI cases by bacterial infection and compare crude proportions between groups based on the absolute risk difference with 95% confidence intervals. Ethical Considerations and Registration: The study and its statistical analysis plan will be registered in clinicaltrials.org prior to conducting the study.