Fever of Unknown Origin
AIMGP Seminar Series
Dr. Katina Tzanetos
February 2007
References

Mourad, O., et al. A Comprehensive EvidencedBased Approach to Fever of Unknown Origin.
Arch Inter Med 163: March 10, 2003.
 Roth, A. and Basello, G. Approach to the Adult
Patient with Fever of Unknown Origin. American
Family Physician 68 (11), 2223.
 Up To Date.
– Approach to the adult with fever of unknown origin
– Etiologies of fever of unknown origin in adults
* Much of this talk based on very helpful
article by Mourad et al. – Highly recommended
Case Discussion – Based on Real Patient






28-year old female, born in Canada, parents from
Hong Kong
2.5 week history of fever 40.0C or higher
Only other symptom is possible rash on lower legs
– intermittent, tender, red nodules
Works in bank
Non-smoker, non-drinker
Only medication is OCP
Take a minute to discuss…

Does she fit the criteria for Fever of
Unknown Origin

Why or why not?
Fever of Unknown Origin - Definition

Classic definition
– Temperature higher than 38.3C
– Several occasions
– Cause obscure after 1-week of in-patient
evaluation

Current definition
– recognizes acceptability of out-patient in place
of in-patient investigations
Case Discussion

Based on short duration and absence of
investigations patient does not fit diagnostic
criteria
 If fever persists, should pursue diagnosis
 Her fever persists
– What aspects of the history and physical
examination do you focus on during this initial
visit?
Four Proposed Categories of FUO

Based on potential etiology of FUO
 All require temperature > 38.3C
 Categorization be especially helpful in organizing
an “approach” to patient evaluation
– Classic
– Nosocomial
– Immune-deficient (neutropenic)
– HIV-related
Classic Category of FUO

Definition:
– Duration > 3 weeks, evaluation of at least 3
outpatient visits or 3 days in-hospital

Common etiologies:
– Infection, malignancy, CVD
This category will be the focus of this talk
Nosocomial Category of FUO

Definition:
– Hospitalization of at least 24 hrs with no fever
on admission, evaluation of at least 3 days

Common etiologies:
– C.Difficile, drugs, PE, septic thrombophlebitis,
sinusitis (intubated patients)
Immune-deficient (neutropenic)
Category of FUO

Definition:
– Neutrophil count < 500/mm3, evaluation of at
least 3 days

Etiologies:
– Opportunistic bacterial infections, aspergillosis,
candidiasis, herpes virus
HIV-Associated Category of FUO

Definition:
– Duration of at least 4 weeks for outpatients and
3 days for inpatients, HIV confirmed

Etiologies:
– Cytomegalovirus, MAI, Pneumocystis, drugs,
Kaposi’s, lymphoma
Etiology and Epidemiology
of Classic FUO

Infections: Most common cause accounting for 1/3 of
cases
– TB; Most common infection in non-elderly adults
– PPD positive in less than 50% of pts with TB and FUO, Sputum
samples positive in only ¼ of patients
– Abscesses
 Usually in abdomen or pelvis with some pre-disposing cause (e.g.
recent surgery, diabetes, biliary tract disease, recent UTI)
– Other infections: Osteomyelitis, endocarditis (esp. in pts with recent
antibiotic use or HACEK organisms)

Malignancy: Second most common cause
– Lymphoma (esp. non-Hodgkin’s), Leukemia, Renal cell, HCC, other
metastasis to liver

CVD: Third most common cause
– Adult Still’s disease in younger patients and giant cell arteritis in older
patients
Diagnostic Approach - History

History
–
–
–
–
–

Travel
Exposures to toxins, sick persons, animals
Immunosuppression
Localizing symptoms
Look for subtle findings: eg. Jaw claudication, nocturia
with prostatitis
Degree of fever, nature of fever curve, apparent
toxicity, and response to antipyretics not specific
enough to guide management
Diagnostic Approach –
Physical Examination

Repeated examination may be needed
 Careful attention to skin, mucous
membranes, lymph and abdominal system
 Ask pts to record and measure temperature
daily
 Yield from history and physical
examination unknown
Back to the case…

Thorough history and physical noncontributory except for intermittent skin
lesions

Given what you know thus far, what
investigations would you order?
Diagnostic Approach –
Laboratory Investigations


Suggested minimal diagnostic work-up to qualify as FUO
has varied over the years
Recent article by Mourad et al suggests following as
minimal:
–
–
–
–
–
–
–
–
–
–
–
History and physical examination
CBC and differential
Blood film reviewed by hematopathologist
Routine chemistry including LDH, bilirubin, liver enzymes
Urinalysis and microscopy
ANA, RH factor
HIV
CMV IgM; heterophil test if suspicious for Mononucleosis
Q-fever serology (if risk exists)
CXR
Hepatitis serology (if abnormal liver enzymes)
Diagnostic Approach –
Investigations and the Evidence

Abdominal CT
– Useful to look for abdominal lymphoma and
abscess
– Diagnostic yield in case series 19%
– Clinical follow-up showed that only 1/32
patients with normal scans had an intraabdominal cause for FUO
Diagnostic Approach –
Investigations and the Evidence

Nuclear Imaging:
– For localizing inflammatory or infectious focus
– Technetium scans likely have best test
characteristics overall and should be test of
choice



Technetium studies: specificity 93%, sensitivity 4075%; PLR 5.7-12.5
Indium-labeled WBC scans: specificity 69%-86%,
sensitivity 45%-82%
Gallium scans: (limited studies)
Diagnostic Approach –
Investigations and the Evidence

Duke criteria for endocarditis:
– Endocardities: 1-5% of all cases of FUO
– Sensitivity 82%, specificity 99%

Liver Biopsy:
– Diagnostic yield 14%-17% regardless of whether
abnormal physical exam or liver enzymes exist
– Complications in FUO from biopsy only 0.32% at most
– Recommended
Diagnostic Approach –
Investigations and the Evidence

Temporal artery biopsy
– Large studies comprised of elderly with FUO lacking
– Arteritis cause of FUO ~16% of pts (All comers)
– Safe, recommended in elderly with FUO

Leg dopplers
– DVT cause of FUO ~ 2-6% of pts
– Safe, easy to do, recommended
Diagnostic Approach –
Investigations and the Evidence

Bone Marrow Examination
– Diagnostic yield of culture 0-2%
– Not recommended in immunocompetent pts

Abdominal exploration
– Role of surgery in post-CT era uncertain

Empiric Therapy (antibiotics, anti-TB, steroids)
– Not studied
– Not recommended
Proposed Diagnostic Algorithm
Mourad, O. et al. Arch Intern Med 2003;163:545-551.
Back to the case…









CBC and differential, electrolytes, BUN, creatinine, Ca/Mg/Ph all
normal
Liver enzymes very slightly elevated then normalized (AST
68normal, ALT 78normal), bilirubin, ALP normal
Multiple blood cultures: no growth
ESR 39
Hepatitis, Lyme, PPD, Mononucleosis, Q-fever, HIV serology all
negative, ANA, RF negative
CT thorax and abdomen normal
2D Echo normal
Leg dopplers negative
Skin biopsy: unremarkable epidermis and dermis, no subcutaneous
material obtained; lesions resolved
Back to the case…





Fever of > 40C continued for more than 4 weeks
No diagnosis despite multiple out-pt visits and a
short in-hospital stay
Debated about going to bone marrow biopsy
versus liver biopsy
Decided on nuclear scan
However, pt was given short course of oral
antibiotics by family MD, symptoms resolved, pt
cancelled all further tests and follow-up
appointments with us and is doing fine
Conclusions from Case

Given our modern-day advances, prognosis in
patients who truly have no diagnosis after
extensive recommended work-up is very good
(most sinister diagnoses are discovered)
 In some cases, spontaneous resolution occurs, in
others, watchful waiting is necessary (but often
frustrating)
– 1930s: > 30% of FUO with no diagnosis died
– Today: 50-90% or more recover spontaneously