Objectives: The goal of this multicenter retrospective study was to evaluate the contribution of 18F-FDG PET/CT in the diagnosis of patients with inflammation of unknown origin (IUO). In addition, C-reactive protein (CRP) level and erythrocyte sedimentation rate were assessed as possible predictors for the outcome of 18F-FDG PET/CT.
Methods: Inflammation of unknown origin was defined as prolonged and perplexing inflammation, with repeated CRP levels more than 20 mg/L or erythrocyte sedimentation rate more than 20 mm/h, body temperature of less than 38.3°C, and without a diagnosis after a variety of conventional diagnostic procedures.
A total of 140 patients with IUO (67 men, 73 women; mean age, 64.2 years; age range, 18–87 years) underwent 18F-FDG PET/CT. 18F-FDG PET/CT was considered helpful when the imaging findings led to a diagnosis, either confirmed by histopathology, microbiological assays, clinical and imaging follow-up, or response to treatment.
Results: In 104 patients (73%), a final diagnosis could be established as follows: infection in 35 patients, malignancy in 18 patients, noninfectious inflammatory disease in 44 patients, and a variety of uncommon conditions in 7 patients. 18F-FDG PET/CT was true positive in 95 patients, true negative in 30 patients (ie, self-limiting conditions), false positive in 6 patients, and false negative in 9 patients (predominantly systemic diseases). In this population, the positive predictive value, negative predictive value, and diagnostic accuracy of 18F-FDG PET/CT were 94%, 77%, and 89%, respectively. In a multivariate analysis, CRP was the only independent predictor for the outcome of 18F-FDG PET/CT.
Conclusions: 18F-FDG PET/CT correctly identified or excluded a causal explanation in approximately 90% of patients with IUO. However, a negative 18F-FDG PET/CT is indicative for a self-limiting condition only after systemic diseases are excluded by other diagnostic tests.
From the *Department of Nuclear Medicine, Medical Center Leeuwarden, Leeuwarden; †Department of Nuclear Medicine, Academic Medical Center, University of Amsterdam, Amsterdam; and ‡Department of Epidemiology, University Medical Center Groningen, Groningen, the Netherlands.
Received for publication August 7, 2013; revision accepted January 28, 2014.
Conflicts of interest and sources of funding: none declared.
Reprints: Hans Balink, MD, Department of Nuclear Medicine, Medical Center Leeuwarden, PO Box 850, 8901 BR Leeuwarden, the Netherlands. E-mail: firstname.lastname@example.org.