PURPOSE:To evaluate the sensitivity and specificity of PET/CT findings in PMR and generate a diagnostic algorithm utilizing a minimum number of musculoskeletal sites. METHODS:Steroid-naïve patients with newly diagnosed PMR (2012 EULAR/ACR classification criteria) were prospectively recruited to undergo whole-body 18F-FDG PET/CT. Each PMR case was age- and sex-matched to four PET/CT controls. Control scan indication, diagnosis and medical history were extracted from the clinical record. Qualitative and semi-quantitative scoring (maximum standardized uptake value [SUVmax]) of abnormal 18F-FDG uptake at 21 musculoskeletal sites was undertaken for cases and controls. Results informed the development of a novel PET/CT diagnostic algorithm using a classification and regression trees (CART) method. RESULTS:Thirty-three cases met the inclusion criteria and were matched to 132 controls. Mean age was 68.6 ± 7.4 years for cases compared with 68.2 ± 7.3 for controls, and 54.5% were male. Median CRP was 49 mg/L (32-65) and ESR 41.5 mm/h (24.6-64.4) in the PMR group. The predominant control indication for PET/CT was malignancy (63.6%). Individual musculoskeletal sites proved insufficient for diagnostic purposes. A novel algorithm comprising 18F-FDG uptake ≥ 2 adjacent to the ischial tuberosities in combination with either abnormalities at the peri-articular shoulder or interspinous bursa achieved a sensitivity of 90.9% and specificity of 92.4% for diagnosing PMR. CONCLUSIONS:The presence of abnormal 18F-FDG uptake adjacent to the ischial tuberosities together with findings at the peri-articular shoulder or interspinous bursa on whole-body PET/CT is highly sensitive and specific for a diagnosis of PMR. TRIAL REGISTRATION:Clinical Trial Registration: Australian New Zealand Clinical Trials Registry, http://www.anzctr.org.au , ACTRN1261400696695.