LAUSR

Purpose The aim of this study was to retrospectively analyze 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography (PET)/ computed tomography (CT) images of non-infected hip arthroplasty patients and summarize findings that may be useful for clinical practice. Methods 18F-FDG PET/CT images of non-infected hip arthroplasty patients were collected from September 2009 to August 2021. The region of interest was independently delineated by two physicians and maximum standardized uptake values (SUVmax) were recorded and compared. Serologic data were also collected and the correlation between SUVmax and serologic parameters was analyzed, while the images were classified based on the 18F-FDG uptake pattern in the images using the diagnostic criteria proposed by Reinartz et al. (9). The interval between hip replacement and PET/CT was classified by year and the characteristics of the two groups were compared. The images of patients who underwent PET/CT multiple times were analyzed dynamically. Results A total of 121 examinations were included; six patients underwent PET/CT twice and two patients had three scans. There were no significant correlations between SUVmax and serologic results. The interobserver agreement between the two physicians in the classification according to the criteria of Reinartz et al. (9) was 0.957 (P < 0.005). Although there was non-specific uptake in cases with an arthroplasty-to-PET/CT interval this was non-significant. Additionally, 18F-FDG showed potential utility for dynamic observation of the condition of the hip. Conclusion SUVmax provided information independent of serologic results, meanwhile 18F-FDG showed potential applicability to the dynamic monitoring of hip arthroplasty-related diseases. However, the presence of blood vessels and muscles affected image interpretation and the specificity of 18F-FDG was not optimal. A more specific radionuclide is needed to maximize the benefits of using PET/CT for the assessment of periprosthetic joint infection (PJI).