Abstract
The purpose of this study was to investigate the usefulness of PET-CT in the diagnosis and treatment of long bone infections following trauma.
All patients referred to the limb reconstruction service for management of non-unions were treated by the same protocol. PET-CT with FDG was performed in all patients to assess if the non-union site was infected and if so, the extent of the infection. Those requiring operative management were treated in a 2-stage manner. Initially with debridement based on PET-CT; sampling for microbiology and histology; and then Teicoplanin and Ciprofloxacin. If samples were positive then the patients were treated for a total of 6 weeks with antibiotics based on microbiology advice before undergoing definitive fixation. The sensitivity, specificity, PPV, and NPV were then calculated for PET-CTs ability to predict presence of infection using extended cultures and histology as the gold standard.
38 consecutive patients underwent surgery, 24 male and 14 female. 24 were deemed infected on extended culture or histology. PET-CT was anecdotally found to be extremely useful at determining the extent of infection to plan debridement.
PPV 0.83
NPV 0.89
Sensitivity 0.96
Specificity 0.61
As well as providing unique ability to demarcate areas of bony infection in the presence of metalwork, the ability to detect or exclude infection was exceptional.
This is a test, however, which is operator dependent and requires a skilled Nuclear Radiology Consultant to accurately interpret images. In our relatively small pilot study the accuracy improved noticeably over one year.
PET-CT has potential to be a powerful tool in the diagnosis and treatment of long bone infection following trauma and certainly warrants further investigation.
