Introduction
Fever of unknown origin (FUO) is a frequently encountered condition on
internal medicine wards. It was first defined in 1961 as a fever greater
than 38.3°C (101°F) for 3 weeks or more, whose origin is not identified
despite a thorough work-up [1]. The recommended work-up has been
modified over the years. While an abundance of studies may aid in the
diagnosis, FUO often poses a great challenge to clinicians, and nearly
half the patients ultimately do not receive a definitive diagnosis
[2-4].
Positron emission tomography with computed tomography (PET/CT) combines
regular CT with nuclear imaging based on the enhanced uptake of
fluorodeoxyglucose (18F-FDG) by highly metabolic
inflammatory and malignant cells. In the last decade, the PET/CT
modality has gained widespread popularity. Several prospective and
retrospective studies have suggested an important role of PET/CT in the
work-up of FUO [5-8]. Nonetheless, much heterogeneity remains
regarding the place of PET/CT within the algorithm of recommended
studies. Some consider PET/CT a mandatory part of the work-up, while
others regard it as a complementary test that may be performed following
the use of other imaging modalities such as CT scans. Differences in
study methodologies have produced variable results regarding the value
of PET.
Given the lack of widespread availability of PET/CT, and its high costs,
the population of FUO patients that is expected to gain benefit from
this modality needs to be further characterized. In the current study,
we assessed the utility of PET/CT in revealing the correct cause of FUO
and evaluated for parameters that may increase the likelihood of
diagnostic yield in patients with FUO.