Background and purpose - 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) can be used in the diagnostic work-up of a patient with suspected periprosthetic joint infection (PJI) but, due to a lack of accurate interpretation criteria, this technique is not routinely applied. Since the physiological uptake pattern of FDG around a joint prosthesis is not fully elucidated, we determined the physiological FDG uptake in non-infected total hip prostheses. Patients and methods - Patients treated with primary total hip arthroplasty (1995-2016) who underwent a FDG-PET/CT for an indication other than a suspected PJI were retrospectively evaluated. Scans were both visually and quantitatively analyzed. Semi-quantitative analysis was performed by calculating maximum and peak standardized uptake values (SUVmax and SUVpeak) by volume of interests (VOIs) at 8 different locations around the prosthesis. Results - 58 scans from 30 patients were analyzed. In most hips, a diffuse heterogeneous uptake pattern around the prosthesis was observed (in 32/38 of the cemented prostheses, and in 16/20 of the uncemented prostheses) and most uptake was located around the neck of the prosthesis. The median SUVmax in the cemented group was 2.66 (95% CI 2.51-3.10) and in the uncemented group 2.87 (CI 2.65-4.63) (Median difference = -0.36 [CI -1.2 to 0.34]). In uncemented prostheses, there was a positive correlation in time between the age of the prosthesis and the FDG uptake (rs = 0.63 [CI 0.26-0.84]). Interpretation - Our study provides key data to develop accurate interpretation criteria to differentiate between physiological uptake and infection in patients with a prosthetic joint.