Review Article

A Rationale for the Use of F18-FDG PET/CT in Fever and Inflammation of Unknown Origin

Table 2

Helpful contribution of hybrid F18-FDG PET/CT in FUO.

AuthorStudy design/technique P/RPatients numberPPV/NPVHelpful/contribution number/(%)Final Dx number/(%)

Federici et al. 2010 [59]R. Full-ring PET/CT14 * 7 (50%)10 (70%)
Keidar et al. 2008 [60]P. Full-ring PET/CT4881%/100%22 (46%)28 (60%)
Ferda et al. 2010 [61]R. Full-ring PET/CECT4898%/75%37 (77%)44 (92%)
Balink et al. 2009 [62]R. Full-ring PET/CECT6893%/100%38 (56%)47 (69%)
Sheng et al. 2011 [63]R. Full-ring PET/CECT4880%/50%#32 (67%)36 (75%)
Pelosi et al. 2011 [64]R. Full-ring PET/CT2485%/91%11 (46%)17 (71%)
Crouzet et al. 2012 [65]R. Full-ring PET/CT7995%/100%45 (57%)61 (77%)

PET/CTTotal number patients226Overall helpfulness of PET/CT 57%
(mean)
Overall percentage
final diagnosis
73% (mean)

Legends: P: prospective; R: retrospective; NPV: negative predictive value; PPV: positive predictive value; CECT: contrast-enhanced CT; NA: not applicable.
*Data could not be retrieved from the original publication.
We question interpretation of results and definition of false negatives. These are based on the later clinical course. However a time window has not been defined for the clinical course. It is therefore possible that the false negatives are explained by another disease process than that was present at the time of PET-CT. This can explain in part the discrepancy between this limited NPV and other publications.”
#This low NPV is probably explained by the relatively high prevalence of disease within the study population.