Background and purpose: The diagnosis of patients with fever of unknown origin (FUO) remains a challenging medical problem for internal medicine. A reliable estimate of the diagnostic performance of FDG-PET and FDG-PET/CT in the assessment of FUO unidentified by conventional workup has never been systematically assessed, and present systematic review was aimed at this issue.
Methods: A systematic search for relevant studies was performed of the PubMed, Embase, and Cochrane databases. Methodological quality of each study was assessed. Sensitivity, specificity and area under the curve (AUC) were meta-analyzed. Subgroup analyses were performed if results of individual studies were heterogeneous.
Results: The inclusion criteria were met by nine studies. Overall, the studies had good methodological quality. Pooled sensitivity and specificity of FDG-PET for the detection of FUO were 0.826 (95% CI; 0.729-0.899) and 0.578 (95% CI; 0.488-0.665), respectively, and the AUC was 0.810. Heterogeneity among the results of FDG PET studies was present (QSE=12.40, I2=67.7%; QSp=35.98, I2=88.9%). Pooled sensitivity and specificity of FDG-PET/CT were 0.982 (95% CI; 0.936-0.998) and 0.859 (95% CI; 0.750-0.934), respectively, and the AUC was 0.947. We did not find any statistical differences in the AUC and Q* index between FDG-PET and FDG-PET/CT (Z=0.566, p>0.05).
Conclusions: Although the FDG-PET studies that we examined were heterogeneous, FDG-PET appears to be a sensitive and promising diagnostic tool for the detection of the causes of FUO. FDG-PET/CT should be considered among the first diagnostic tools for patients with FUO in whom conventional diagnostics have been unsuccessful.
Copyright © 2010 Elsevier Ireland Ltd. All rights reserved.