[18F]Fluorodeoxyglucose positron emission tomography (FDG-PET) is increasingly used to demonstrate inflammation in specific sites typical for polymyalgia rheumatica (PMR). Scoring systems based on FDG uptake have been proposed to increase diagnostic accuracy. Methods: Retrospective inclusion of 198 consecutive patients ≥40 years of age referred for FDG-PET from the Department of Rheumatology. We assessed the degree of FDG uptake in predilection sites visually, as well as semiquantitatively, and through logistic regression analyses, we evaluated the performance of existing scoring systems as well as a new, simplified scoring system, against the final clinical diagnosis at 6 months after the FDG-PET scan. Results: We found high diagnostic accuracy for the diagnosis of PMR (range 0.74–0.91) using most of the existing scoring systems in glucocorticoid-naïve patients. A simplified scoring system including only periarticular FDG uptake in the shoulders and the ischiogluteal bursae retained high sensitivity and specificity (0.92 and 0.86, respectively). We found a detrimental effect on diagnostic accuracy in all scoring systems in patients treated with glucocorticoids within 4 weeks prior to FDG-PET. Conclusion: Most FDG-PET scoring systems perform well for the diagnosis of PMR, and there is no loss of either sensitivity or specificity in the simplest scoring systems evaluating FDG uptake in only a few selected anatomical regions. However, systemic glucocorticoid treatment up to 4 weeks prior to FDG-PET has a markedly detrimental effect on the diagnostic accuracy of all scoring systems.