Abstract
1692
Introduction: Differentiating infection and sterile inflammation is the main clinical concern of clinicians as they are closely related to each other. Although 18F FDG PET/CT imaging is widely used, its main disadvantage is its lack of specificity to discriminate aseptic from septic inflammation. WBC PET/CT scan is a promising tool for the accurate diagnosis of infection owing to its high specificity. The aim of the study was to determine the utility of WBC PET/CT in the diagnosis of occult infections and to assess its incremental value over routine 18F FDG PET/CT scan.
Methods: This prospective observational diagnostic accuracy study was conducted between October 2018 and October 2020 at the Fortis Memorial Research Institute, Gurugram, in the department of Nuclear Medicine. 33 patients with fever of unknown origin or suspected periprosthetic infection and raised CRP and TLC were included in the study. Patients documented as having favorable clinical response to antibiotics were not included in the study. All the patients underwent both WBC PET/CT scan and 18F FDG PET/CT scan using a standard protocol on two different days. Images of both the scans were evaluated by both visual analysis based on uptake intensity and quantitative grading based on lesion to background SUVmax values. Final diagnosis was made by histopathological, microbiological analysis or clinical-radiological work-up. ROC curve analysis was done to calculate the optimal SUVmax threshold cutoff for both the imaging modalities.
Results: 33 patients were included in the study. 29 foci of suspected infection were found in 25 patients by either 18F FDG PET/CT or 18F FDG labelled WBC PET/CT scan. No abnormal uptake of either 18F FDG or 18F FDG WBC was seen in 8 patients. There was a concordance of 18F FDG and 18F FDG WBC in 27 and 28 sites using grade 1b of visual and quantitative analysis respectively. Of the 29 suspicious infected foci, 18 were proven positive for infection (14/18 sites by the histopathological / microbiological culture and the rest 4/18 sites by clinical / radiological work-up). Culture of aspirates from 9 suspicious sites did not grow an active organism and were considered non infective and 2 sites were considered non infective on clinical follow up. Mean clinical follow up was 8 months (1 - 15 months). Overall significantly higher diagnostic accuracy was demonstrated with WBC PET/CT in comparison to 18F FDG PET/CT for the detection of infection (p < 0.05).The highest diagnostic accuracy of WBC PET/CT scan was reported with both grade 1b of visual (for truncal lesions uptake equivalent to liver or lumbar vertebrae uptake; for extremity lesions uptake significantly higher than neighboring soft tissue uptake or higher than neighboring bone marrow uptake) as well as of quantitative analysis (Lesion to background SUVmax: 2.5 - 3.5) and grade 2 for both visual and quantitative analysis for 18F FDG PET CT. On the basis of ROC analysis, optimal threshold cut off for the differential diagnosis of the infection and aseptic inflammation was 3.5 for WBC PET/CT and 5.8 for 18F FDG PET/CT scan. Using this threshold for SUVmax, the sensitivity and specificity of WBC PET/CT was 94.4 % and 84.21 %, respectively, which were both higher than the sensitivity (83.3%) and specificity (78.95 %) of 18F FDG PET/CT scan.
Conclusions: WBC PET/CT has a higher diagnostic accuracy along with an incremental value over 18F FDG PET/CT for the diagnosis of occult infection.
