Retrospective study on clinical value and optimal use of [F] FDG PET/CT for inflammation of unknown origin in Japanese patients.

Despite advancements in medical technology, the challenges of diagnosing fever of unknown origin (FUO) and inflammation of unknown origin (IUO) persist. Positron emission tomography/computed tomography (PET/CT) has been used to assess these conditions; however, it is unclear which patients most benefit from this approach. This study aimed to assess the clinical value and optimal use of fluorine-18-labelled-fluorodeoxyglucose (F-FDG)-PET/CT in patients with FUO/IUO. We reviewed FDG-PET/CT scans conducted at our department between January 2014 and December 2020 to further assess FUO/IUO. FUO was defined as a fever lasting over three weeks without a diagnosis based on conventional assessment. IUO was defined as prolonged inflammation with a body temperature below 38.3 ℃ and without a diagnosis after conventional diagnostic procedures. The efficacy of FDG-PET/CT was determined based on the diagnostic outcomes achieved by integrating imaging findings with histopathological or microbiological evidence, clinical criteria, or follow-up assessments. We conducted crude and age-sex adjusted logistic regression for potential risk factors to determine the degree of burden on demographic and clinical characteristics. Forty-five patients with FUO/IUO underwent FDG-PET/CT, and final diagnoses were made in 32 patients (71.1%). The percentages of diagnostic categories were 53.3%, 8.9%, and 6.7% for non-infectious inflammatory diseases, malignancy, and infection, respectively. In multivariable logistic regression analysis, IUO was the only independent predictor of the efficacy of FDG-PET/CT in the final diagnosis (odds ratio, 67.02; 95% confidence interval, 4.02-1119). Our results indicate that IUO was associated with a higher diagnostic yield compared to those with FUO.