Ten per cent of patients with hip replacement will eventually complain of significant pain after surgery, often requiring a revision arthroplasty. The majority of these patients experience aseptic loosening rather than infection. Despite significant advances made in diagnostic imaging, distinguishing infection from aseptic loosening remains a significant challenge. Imaging using fluorodeoxyglucose (FDG) positron emission tomography (PET) has been reported to have excellent sensitivity in detecting infections associated with hip prostheses. However, in some studies, a high rate of false positive results has been reported, especially when increased tracer uptake adjacent to the prosthesis (which is not surrounded by bone) is used as the sole criterion for diagnosing infection. The objective of this investigation was to determine the optimal criteria for diagnosing periprosthetic infection, thereby avoiding false positive results in this setting. A total of 41 total hip arthroplasties from 32 patients and for whom complete clinical follow-up was available were included in this analysis. The location and intensity of FDG uptake were determined for each scan. Final diagnosis was made by microbiology, histopathology, surgical findings and clinical follow-up. Patients who did not undergo surgery were followed up to at least 9 months. Twelve patients were proven eventually to have periprosthetic infection. Images from 11 of these patients displayed increased tracer uptake along the interface between bone and prosthesis. The intensity of the increased tracer uptake varied from mild to moderate, with standardized uptake values less than 2. In contrast, images from uninfected, loose hip prostheses revealed very intense uptake around the head or neck of the prosthesis with standardized uptake values as high as 7. It is concluded that the intensity of increased FDG uptake is less important than the location of increased FDG uptake when FDG PET is used to diagnose periprosthetic infection in patients with hip arthroplasty. Using increased uptake as the sole criterion for diagnosing infection will result in false positive results in this setting.