Advertisement for orthosearch.org.uk
Orthopaedic Proceedings Logo

Receive monthly Table of Contents alerts from Orthopaedic Proceedings

Comprehensive article alerts can be set up and managed through your account settings

View my account settings

Visit Orthopaedic Proceedings at:

Loading...

Loading...

Full Access

General Orthopaedics

A Nomogram to Predict Success of Two-Stage Revisions for Treating Knee Prosthetic Joint Infections

International Society for Technology in Arthroplasty (ISTA) 2012 Annual Congress



Abstract

Background

Two-stage revision is considered the gold standard for treatment of knee prosthetic joint infections. Current guidelines for selecting the most appropriate procedure to eradicate knee prosthetic joint infections are based upon the duration of symptoms, the condition of the implant and soft tissue evaluated during surgery and the infecting organism. A more robust tool to identify candidates for two-stage revision and who are at high risk for treatment failure might improve preoperative risk assessment and increase a surgeon's index of suspicion, resulting in closer monitoring, optimization of risk factors for failure and more aggressive management of those patients who are predicted to fail.

Methods

Charts from 3,809 revision total joint arthroplasties were reviewed. Demographic data, clinical data and disease follow-up on 314 patients with infected total knee arthroplasty treated with two-stage revision were collected. Univariate analyses were performed to determine which variables were independently associated with failure of the procedure to eradicate the prosthetic joint infections. Cox regression was used to construct a model predicting the probability of treatment failure and the results were used to generate a nomogram which was internally validated using bootstrapping.

Results

209 (66.6%) cases experienced reinfection at an average of 429 days (range, 9 to 3,886) following the two-stage revision. Univariate analysis identified multiple variables independently associated with reinfection including: a longer duration of symptoms (p<0.001), a longer time from the index total knee arthroplasty (p=0.003), a higher number of previous surgeries in the same joint (p<0.001), an elevated C-reactive protein (p=0.005), an elevated erythrocyte sedimentation rate (p=0.006), a low hemoglobin (p=0.001), a previous infection in the same joint (p<0.001), diabetes (p<0.001), and heart disease (p=0.006). Among 1,000 bootstrap samples, the bias corrected receiver operating characteristic for the nomogram was 0.77.

Conclusions

Preoperative knowledge of the probability that a treatment strategy will eradicate a patient's prosthetic joint infection may improve risk assessment and allow adequate time for consideration of alternative therapies.