Fever of unknown origin (FUO): which are the factors influencing the final diagnosis? A 2005–2015 systematic review

Abstract

Background The differential diagnosis of Fever of Unknown Origin (FUO) is very extensive, and includes infectious diseases (ID), neoplasms and noninfectious inflammatory diseases (NIID). Many FUO remain undiagnosed. Factors influencing the final diagnosis of FUO are unclear. Methods To identify factors associated with FUO diagnostic categories, we performed a systematic review of classical FUO case-series published in 2005–2015 and including patients from 2000. Moreover, to explore changing over time, we compared these case-series with those published in 1995–2004. Results Eighteen case-series, including 3164 patients, were included. ID were diagnosed in 37.8% of patients, NIID in 20.9%, and neoplasm in 11.6%, FUO were undiagnosed in 23.2%. NIIDs significantly increased over time. An association exists between study country income level and ID (increasing when the income decreases) and undiagnosed FUO (increasing when the income increases); even if not significant, the use of a pre-defined Minimal Diagnostic Work-up to qualify a fever as FUO seems to correlate with a lower prevalence of infections and a higher prevalence of undiagnosed FUO. The multivariate regression analysis shows significant association between geographic area, with ID being more frequent in Asia and Europe having the higher prevalence of undiagnosed FUO. Significant associations were found with model of study and FUO defining criteria, also. Conclusions Despite advances in diagnostics, FUO still remains a challenge, with ID still representing the first cause. The main factors influencing the diagnostic categories are the income and the geographic position of the study country.

Monogenic autoinflammatory diseases in adults – a challenge to rheumatologic practice at the onset of the Polish national programme of interleukin 1 inhibitor treatment

“…In conclusion, the Polish national programme of anti-IL-1 treatment opens new possibilities for the treatment of AIDs. However, they remain underdiagnosed and more awareness is needed [62].…”

Fever of Unknown Origin (FUO): towards uniform definitions and classification system

“…According to Vanderschueren et al, 43.9% of cases were in the category undiagnosed if the definition from 1991 was used, and 53.0% were undiagnosed according to 1961 definition; however, no significant differences in diagnostic categories among diagnosed cases were noted (8). In a systematic review, the frequency of FUO from neoplasms was lower in patients selected with the definition from 1991, and the frequency of FUO from non-infectious inflammatory disorders (NIIDs) was lower in patients with the third FUO definition (31). Infections were more prominent, although not significantly, when the definition from 1991 was used compared to when the definition from 1961 was used (34).…”

“…In addition to the improvement in diagnostic methods, it became obvious that the type of diagnostic panel is more important than the duration of investigations and that it would be better if in the definition of FUO, a quantitative time defined criterion was replaced with a qualitative baseline set of obligatory investigations, which would include biochemistry, blood and urine cultures, basic imaging procedures, and a set of infectious disease screening tests determined from local epidemiological data (third FUO definition) (5,26,(29)(30)(31). In short, this attempt at a reformed definition proposed to change the quantitative criterion of the time period during which no diagnosis is made to a qualitative one via an initial appropriate intelligent standard diagnostic inpatient or outpatient workup.…”

“…Defining the necessary initial investigations to try to reach a diagnosis remains a matter of debate, but it is generally agreed that the standard initial diagnostic investigation protocol should at least include a comprehensive history and repeated physical examination, complete blood count with differential cell count, electrolytes, renal and liver function tests, protein electrophoresis, enzymes (alkaline phosphatase, aminotransferase, lactate dehydrogenase, creatine phosphokinase), CRP, ESR, microscopic urinalysis, three blood cultures (different sites, several hours apart), urine culture, chest X-ray, abdominal ultrasonography, a tuberculin skin test or interferon gamma release assay, which is quite often accompanied by testing for antinuclear antibodies, rheumatoid factor, and anti-HIV test (26,(29)(30)(31)(42)(43)(44). Further evaluations indicated by potentially diagnostic clues, which are defined as all localizing signs, symptoms, laboratory tests, and other abnormalities potentially pointing towards a diagnosis, can additionally be included (5,21,28,30,42).…”

Association of Schistosoma haematobium Infection Morbidity and Severity on Co-infections status in Pre-school Age Children Living in a Rural Endemic Area in Zimbabwe

“…There was a 10-fold chance of nding FUO and schistosomiasis as a coinfection, children with schistosomiasis had 138% risk of FUO. Despite advances in medicine, the proportion of patients discharged with undiagnosed FUO after systematic examination has not improved (47). The cause of febrile illness is not identi ed in approximately 9 -51% of patients, this is even higher in resource limited areas endemic for childhood illnesses (48).…”