Diagnostic performance of 18F-FDG-PET/CT in vascular graft infections

The aim of this study was to evaluate the diagnostic accuracy of positron emission tomography/computed tomography with (18)F-fludeoxyglucose (FDG-PET/CT) in a population with suspected graft infection and to validate a new diagnostic imaging score for FDG-PET/CT.

This was a prospective cohort study. FDG-PET/CT was performed prospectively in 34 patients with suspected graft infection, in 12 of them before the start of antimicrobial treatment. Diagnostic accuracy was assessed using a new five point visual grading score and by using a binary score. Maximum standardized uptake values (SUVmax) were calculated for quantitative measurements of metabolic activity, and cut off points were calculated using the receiver operator curve (ROC). The standard of reference was a microbiological culture, obtained after open biopsy or graft explantation.

Using the new scale, FDG-PET/CT correctly recognized 27 patients with graft infection, one patient was diagnosed as false positive, six patients were correctly classified as true negative, and no patients were rated false negative. Hence, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy of FDG-PET/CT for the diagnosis of graft infections were 100%, 86%, 96%, 100%, and 97%, respectively. Using a previously established binary score, sensitivity, specificity, PPV, NPV, and accuracy were 96%, 86%, 96%, 86%, and 94% respectively. ROC analysis suggested an SUVmax cut off value of ≥3.8 to differentiate between infected and non-infected grafts (p < .001). Additionally, FDG-PET/CT provided a conclusive clinical diagnosis in six of seven patients without graft infection (i.e., other sites of infections).

The diagnostic accuracy of FDG-PET/CT in the detection of aortic graft infection is high. A newly introduced five point visual grading score and early imaging prior to antimicrobial treatment may further improve the diagnostic accuracy.