Septic Arthritis Risk Calculator: External Validation of a Prediction Model for Native Adult Knees Among Veterans

Area of Science:

• Orthopedics
• Infectious Diseases
• Biostatistics

Background:

• Septic arthritis, particularly in acutely inflamed knees (AIK), demands rapid diagnosis to avert joint destruction and systemic spread.
• Current diagnostic criteria for septic arthritis are insufficient, complicating clinical decision-making.
• A prior multivariable prediction model for septic arthritis in nonveterans showed high accuracy and was adapted into the Septic Arthritis Risk Calculator (SARC).

Purpose of the Study:

• To externally validate the accuracy and calibration of the Septic Arthritis Risk Calculator (SARC) in a veteran population.
• To assess SARC's performance in estimating the risk of native knee septic arthritis.

Main Methods:

• A retrospective cohort study of veterans diagnosed with native AIK between 2007 and 2022 at a Veterans Affairs Medical Center.
• Data collected included demographics, comorbidities, clinical findings, and diagnostic results.
• SARC was used to calculate septic arthritis risk; discrimination was assessed using receiver operating characteristic analysis, and calibration was evaluated with Hosmer-Lemeshow goodness-of-fit and Cox regression.

Main Results:

• The study included 241 veterans, with 50 cases (20.7%) of septic arthritis.
• SARC demonstrated high discrimination, with an area under the curve of 0.89, correctly classifying 90% of cases.
• The model showed excellent calibration, with Hosmer-Lemeshow P = 0.43 and a Cox regression slope of 1.0 (95% CI: 0.74 to 1.3).

Conclusions:

• The Septic Arthritis Risk Calculator (SARC) is accurate and well-calibrated for use in veterans with acutely inflamed knees.
• SARC's validation in this cohort supports its clinical utility for risk assessment in AIK cases.
• Future implementation of SARC as an online decision support tool is planned for point-of-care use.