Fever of Unknown Origin: Focused Diagnostic Approach Based on Clinical Clues from the History, Physical Examination, and Laboratory Tests Burke A. Cunha, MD, MACP a,b, * a Infectious Disease Division, Winthrop-University Hospital, 259 First Street, Mineola, Long Island, NY 11501, USA b State University of New York School of Medicine, Stony Brook, NY, USA Few clinical problems are as challenging and difficult as that presented by the patient who has a febrile illness for more than 10 to 15 days, the origin of which remains obscure. No type of illness puts to a stronger test the phy- sician’s ability to approach a clinical problem effectively.. –Philip A. Tumulty, MD Fever of unknown origin (FUO) was first and correctly termed ‘‘pro- longed and perplexing fevers’’ by Kiefer and Leard [1]. Prolonged and per- plexing fevers are difficult-to-diagnose febrile disorders aptly termed FUOs. FUOs may be conveniently divided into four general categories based on the etiology of the FUO: (1) infectious, (2) rheumatic-inflammatory, (3) neo- plastic, or (4) miscellaneous disorders. Petersdorf and Beeson [2] in 1961 were the first to define FUO in terms of time-based diagnostic criteria. These have since been termed ‘‘classic’’ FUOs and may be defined as disorders with temperatures greater than or equal to 101 F that have persisted for at least 3 weeks that were not diagnosed after a week of intensive in-hospital testing. The classic and current causes of FUO have been modified, reflect- ing the changing spectrum of diseases and the availability of sophisticated diagnostic tests in the outpatient setting [3,4]. * Infectious Disease Division, Winthrop-University Hospital, 259 First Street, Mineola, Long Island, NY 11501. 0891-5520/07/$ - see front matter Ó 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.idc.2007.09.004 id.theclinics.com Infect Dis Clin N Am 21 (2007) 1137–1187
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Infect Dis Clin N Am 21 (2007) 1137–1187
Fever of Unknown Origin: FocusedDiagnostic Approach Based on Clinical
Clues from the History,Physical Examination,and Laboratory Tests
Burke A. Cunha, MD, MACPa,b,*aInfectious Disease Division, Winthrop-University Hospital, 259 First Street,
Mineola, Long Island, NY 11501, USAbState University of New York School of Medicine, Stony Brook, NY, USA
Few clinical problems are as challenging and difficult as that presented bythe patient who has a febrile illness for more than 10 to 15 days, the originof which remains obscure. No type of illness puts to a stronger test the phy-
sician’s ability to approach a clinical problem effectively..–Philip A. Tumulty, MD
Fever of unknown origin (FUO) was first and correctly termed ‘‘pro-longed and perplexing fevers’’ by Kiefer and Leard [1]. Prolonged and per-plexing fevers are difficult-to-diagnose febrile disorders aptly termed FUOs.FUOs may be conveniently divided into four general categories based on theetiology of the FUO: (1) infectious, (2) rheumatic-inflammatory, (3) neo-plastic, or (4) miscellaneous disorders. Petersdorf and Beeson [2] in 1961were the first to define FUO in terms of time-based diagnostic criteria. Thesehave since been termed ‘‘classic’’ FUOs and may be defined as disorderswith temperatures greater than or equal to 101�F that have persisted forat least 3 weeks that were not diagnosed after a week of intensive in-hospitaltesting. The classic and current causes of FUO have been modified, reflect-ing the changing spectrum of diseases and the availability of sophisticateddiagnostic tests in the outpatient setting [3,4].
* Infectious Disease Division, Winthrop-University Hospital, 259 First Street, Mineola,
Long Island, NY 11501.
0891-5520/07/$ - see front matter � 2007 Elsevier Inc. All rights reserved.
The main diagnostic difficulty with FUOs is an efficient and effective di-agnostic approach. Clinicians are advised that to diagnose FUOs effectively,they should be comprehensive. Unfortunately, often this has only resulted inexcessive diagnostic testing to rule out every disorder causing FUOs. A non-focused approach has the effect of incurring unnecessary expense, inconve-niencing patients, and delaying or obscuring the FUO diagnostic work-up.The undesirable effect of the ‘‘shotgun approach’’ to diagnostic testing isthat it underuses the FUO tests appropriate for the most likely diagnosticcategories, and it overtests for unlikely diagnoses [5].
The diagnostic approach to the FUO patient should be focused and rel-evant to the clinical syndromic presentation. Because all patients with FUOsby definition have fevers, the clinician should identify the predominant fea-tures of the clinical presentation to determine the general category of FUOof the patient. With infectious FUOs, fevers are often accompanied by chillsor night sweats. Weight loss without loss of appetite is another potential in-dicator of an infectious disease etiology. The clinical presentation of patientswith FUOs caused by rheumatic-inflammatory disorders is dominated byarthralgias, myalgias, or migratory chest or abdominal pain. The predomi-nant symptoms of patients with neoplastic FUOs are fatigue and weight losswith early or dramatic decrease in appetite. Night sweatsmay also be a featureof neoplastic disorders. Patients presenting with FUOs whose symptoms donot suggest an infectious, rheumatic-inflammatory, or neoplastic disordershould be considered as having an FUO of miscellaneous causes [6–8].
It makes little sense to get thyroiditis tests for every FUO patient if thereis not an antecedent history of thyroid or autoimmune disease or physicalfindings referable to subacute thyroiditis. Similarly, just because subacutebacterial endocarditis (SBE) is a common cause of FUO, transthoracicechocardiography (TTE)–transesophageal echocardiography (TEE) shouldnot be obtained on all FUO patients. In patients with FUOs, TTE-TEEshould be obtained only in those with heart murmurs. In FUO patientswith heart murmurs, vegetations seen on TTE-TEE can indicate SBE(culture positive and culture negative); systemic lupus erythematous (SLE;Libman-Sacks vegetations); or marantic (nonbacterial) endocarditis.
By using a focused approach the clinician can order tests that are morerelevant to the presenting clinical syndrome; such tests more efficientlyand effectively lead to a correct FUO diagnosis. The diagnostic approachto FUOs may be considered as consisting of three phases. The initial phaseconsists of the initial FUO history and physical examination and nonspecificlaboratory tests. This phase provides the clinician with a general sense ofwhether the FUO is likely to be caused by an infection or by a rheumatic-inflammatory or neoplastic disorder. Phase II involves re-evaluating thepatient using a focused FUO history and physical examination and addi-tional nonspecific and specific laboratory tests. The focused FUO evaluationhas the effect of narrowing diagnostic possibilities and eliminating possibil-ities from further diagnostic consideration [1,5].
1139FUO: DIAGNOSTIC APPROACH
Nonspecific laboratory tests included during the initial evaluation arehelpful to increase the diagnostic probability of some entities, whereasdecreasing or eliminating the diagnostic probability of others. Nonspecificlaboratory tests, as with other clinical findings, are more significant whenconsidered together rather than individually. For example, the combinationof the following nonspecific laboratory tests, which alone are unhelpfuldiagnostically, should suggest a particular diagnosis (ie, increased lactatedehydrogenase, atypical lymphocytes in the peripheral smear, and thrombo-cytopenia should suggest the possibility of malaria in patients with anappropriate epidemiologic exposure). The patient exposed in malariousareas is also exposed to typhoid fever. Both typhoid fever and malariahave some clinical features in common and neither has localizing signs,making these infections difficult to diagnose if the epidemiologic history isnot taken into account and the nonspecific laboratory clues are not appre-ciated for their diagnostic significance. The same three nonspecific labora-tory tests argue strongly against typhoid fever and should point theclinician to the possibility of malaria as the cause of the patient’s FUO [1,9].
Fever of unknown origin: classic and current causes
FUOs fall into four general categories. The relative frequency of thecauses of FUO in each category is the basis for a phased diagnostic ap-proach. Phase I of an FUO evaluation consists of a FUO relevant history,physical examination, and nonspecific laboratory tests. The phase I evalua-tion provides the basis for determining the course of the FUO work-up. Fea-tures in the history, physical findings, and laboratory abnormalities in theinitial FUO evaluation suggest which general category of disorder is respon-sible for the patient’s FUO. Although all FUOs, by definition, are associ-ated with fever, the predominant symptoms usually suggest a particularFUO category.
In general, infectious diseases may be associated with chills, night sweats,myalgias, or weight loss with an intact appetite. Arthralgias or myalgias arethe predominant complaints of the patient presenting with rheumatic-in-flammatory causes of FUO. These patients often have fatigue, but weightloss or night sweats are unusual findings. Even with some overlap in symp-toms, the clinician can usually determine from the dominant clinical featureswhether the patient is likely to have an infectious, rheumatic-immunologic,or neoplastic cause of their FUO. Typically, in addition to fever and fatigue,neoplastic disorders have night sweats and weight loss accompanied bya dramatic and profound loss of appetite. Patients that do not fit in anyof these categories have FUOs of a variety of miscellaneous causes. It issometimes difficult to differentiate between infectious and neoplastic or in-fectious and rheumatic disorders. In such situations, the next phase ofFUO investigation using a focused diagnostic approach provides additional
1140 CUNHA
information with a history and physical examination or additional labora-tory tests, which clearly differentiate one group from another, and areable to narrow differential diagnostic possibilities within a category. Classicand current FUO causes have been reviewed (Table 1) [2,3,6,10–13].
Fever of unknown origin: focused diagnostic approach
Overview
After the initial FUO-relevant evaluation most of the common causes ofFUOs in each category may be readily diagnosed. Combining the relevantFUO features on physical examination with selected nonspecific laboratorytest abnormalities limits diagnostic possibilities and eliminates other causesfrom further diagnostic consideration. The diagnostic significance of se-lected nonspecific tests cannot be overemphasized. The clinical significanceof nonspecific laboratory abnormalities is enhanced when they are consid-ered together. As with FUO-relevant historical facts or physical findings,nonspecific laboratory abnormalities taken together increase diagnosticspecificity and significance.
The function of the initial phase of FUO evaluation is to diagnose disor-ders, which are most easily diagnosed among the FUOs, and to limit differ-ential diagnostic possibilities that direct the second phase of focused FUOevaluation. The focused FUO history of physical examination and labora-tory tests has the purpose to refine further the differential diagnosis of dis-orders that have not been diagnosed during the initial evaluation [14,15].
Fever of unknown origin: initial evaluation
Although the initial direction of the diagnostic work-up is suggested byFUO-relevant aspects of the history and physical examination, the basicbattery of nonspecific laboratory tests helps to define further differential di-agnostic possibilities. Many disorders in all categories of FUO are ac-companied by some nonspecific laboratory abnormalities. The diagnosticsignificance of such findings alone and more critically taken together is oftenoverlooked as having no importance because the abnormalities are not ofsufficiently impressive magnitude or the abnormalities are associated withmany potential disorders. The basic nonspecific laboratory test battery in-cludes the complete blood count, erythrocyte sedimentation rate (ESR),C-reactive protein, and liver function tests. Imaging tests include a chest ra-diograph (if there are signs or symptoms referable to the chest) and CT andMRI scans of the abdomen and pelvis (as dictated by clinical clues suggest-ing an intra-abdominal or pelvic pathology. Blood cultures are also includedas part of the initial diagnostic evaluation [1,16–19].
Blood cultures pick up common causes of SBE (bacteremia from an in-tra-abdominal or pelvic or renal-perinephric source), and intra-abdominal
imaging provides important information for the focused phase of FUO eval-uation. If the patient has a very highly elevated ESR (R100 mm/h) it sug-gests possible FUO etiologies including abscesses, osteomyelitis, SBE, andadult Still’s disease. Among the rheumatic inflammatory causes of FUOthe ESR greater than or equal to 100 mm/h may point to adult Still’s dis-ease, polymyalgia rheumetica/temporal arteritis, late-onset rheumatoid ar-thritis, SLE, periarteritis nodosa, Takayasu’s arteritis, Kikuchi’s disease,or familial Mediterranean fever. Among neoplastic disorders an increasedESR rate may be present with any of them but an unelevated ESR has nodifferential diagnostic value and does not rule out neoplastic or other disor-ders. The high ESR rate may also point to drug fever and the miscellaneouscategory and elevated ESR rate may indicate drug fever, regional enteritis,subacute thyroiditis, deep vein thrombosis or small pulmonary emboli, andso forth.
Imaging tests (ie, CT and MRI scanning of the chest, abdomen, and pel-vis) may show otherwise unsuspected adenopathy, hepatomegaly or spleno-megaly, abscesses, or masses. The initial phase FUO evaluation provides theimportant diagnostic information that should guide the subsequent diagnos-tic process. The focused FUO diagnostic approach should be based on theinitial FUO evaluation. Findings of the initial FUO evaluation should bebased on this [5,10,16–20].
Fever of unknown origin: focused FUO evaluation
The focused FUO evaluation builds on the initial FUO diagnostic im-pression. During the second phase of FUO evaluation focused diagnosticapproach uses a more detailed history, physical examination, and additionalnonspecific laboratory tests not obtained during the initial evaluation. Thefocused FUO evaluation confirms or eliminates any differential diagnosticdifficulties encountered during the initial evaluation and is designed to iden-tify less common causes of FUO in each category. The laboratory tests in-cluded in the focused test battery include antinuclear antibodies, rheumatoidfactor, serum protein electrophoresis, serum ferritin, cold agglutinins, andso forth. Also included is serology for Epstein-Barr virus, cytomegalovirus,and Bartonella. If SLE is in the differential diagnosis, double-stranded DNAand anti–Smith antibodies are included. If malignancies are likely diagnosticpossibilities, then additional nonspecific tests, such as uric acid, lactate de-hydrogenase, and leukocyte alkaline phosphatase, are included. If diagnos-tic findings suggest the possibility of subacute thyroiditis, then tests forthyroid antibodies and thyroid function tests should be included. FUO pa-tients with a heart murmur should have a TTE-TEE as part of the work-upfor endocarditis. Patients with a heart murmur and a high-grade continuousbacteremia (with an organism associated with endocarditis) with or withoutperipheral manifestations are diagnosed with SBE. Patients with a heartmurmur and negative blood cultures without peripheral manifestations of
system; DVT, deep vein thrombosis; HL, Hodgkin’s lymphoma; N
1144 CUNHA
SBE have marantic endocarditis. Patients with atrial myxomas have a heartmurmur; vegetations on TTE-TEE with or without peripheral embolicphenomenon ‘‘culture-negative endocarditis’’ is a frequently misapplieddiagnosis indicating heart murmur with negative blood cultures. True cul-ture-negative endocarditis refers to patients with a heart murmur, vegeta-tions on TTE-TEE (no evidence of an atrial myxoma), with peripheralmanifestations of SBE [1,10,13].
Some tests obtained during the initial FUO evaluation may be of assis-tance with some disorders in this category (ie, regional enteritis [Crohn’s dis-ease]). Presenting as an FUO, Crohn’s disease may be a difficult diagnosiswhen unaccompanied by abdominal complaints. If there are findings onthe abdominal CT or MRI suggesting terminal ilial abnormalities then a gal-lium-indium scan may be obtained, which should also show increased up-take in the ileum. FUO patients with regional enteritis may present onlywith extraintestinal manifestations (eg, episcleritis). The diagnostic signifi-cance of episcleritis as an initial manifestation of Crohn’s disease is easilyoverlooked in an FUO patient. The patient with regional enteritis mayalso have an increased ESR and monocytosis, which together with otherfindings suggests that Crohn’s disease is indeed the cause of the patient’sFUO (see the article by Cunha elsewhere in this issue for further explorationof this topic) [1,15–28].
Diagnostic significance of fever patterns
Morning temperature spikesIn obscure causes of FUO, fever curves are useful diagnostically and of-
ten provide the only clue to the diagnosis. The first step in evaluating feverpatterns is to determine the time of the peak period during a 24-hour period.Most patients with fever have peak temperatures in the late afternoon orearly evening. This means that there are relatively few disorders associatedwith morning temperature elevations. If not altered by antipyretic medica-tions or devices, the periodicity of fever can be a useful diagnostic aid inobscure cases of FUO. The causes of FUO associated with morning temper-ature elevations are typhoid fever; tuberculosis; and among the noninfec-tious disorders, periarteritis nodosa [29,30].
Relative bradycardiaA pulse-temperature deficit is termed ‘‘relative bradycardia’’ (Faget’s
sign). For a pulse temperature to be termed relative bradycardia theremust be a significant pulse temperature deficit relative to the degree of fever.Relative bradycardia should not be applied to children or those withtemperatures of less than 102�F or adults with temperatures of less than102�F or those on b-blockers, diltiazem, verapamil, or who have pace-maker-induced rhythms or arrhythmias. The pulse rate for any given degreeof temperature elevation is physiologic and predictable. For every degree of
1145FUO: DIAGNOSTIC APPROACH
temperature elevation in degrees Fahrenheit there is a concomitant increasein pulse rate of 10 beats per minute. In the absence of the exclusion criteriamentioned, a temperature of 104�F should be accompanied by an appropri-ate pulse response of 130 beats per minute. This patient with relative brady-cardia would have a pulse less than or equal to 120 beats per minute.Appropriate pulse-temperature relationships are shown in Table 2. Appliedcorrectly in the appropriate clinical context in patients with FUO relativebradycardia is an important diagnostic sign. In FUO patients, relative bra-dycardia may occur in association with malaria, typhoid fever, any centralnervous system disorder, some lymphomas, and drug fever. Simultaneouspulses should be obtained in all patients with FUOs to determine if relativebradycardia is present. Relative tachycardia refers to an inappropriatelyrapid pulse for a given degree of temperature, and is only associated withpulmonary emboli among the causes of FUO [29–31].
Double quotidian feversDouble quotidian fevers refer to two temperature spikes occurring within
a 24-hour period. Although double quotidian fevers are not a common feverpattern, they are most helpful when present in febrile patients presentingwith a differential diagnosis. Infectious causes of FUO associated with dou-ble quotidian fevers include miliary tuberculosis, visceral leishmaniasis, andmixed malarial infections. In returning travelers from India, malaria andtyphoid fever are important differential diagnostic considerations. In sucha patient with a double quotidian fever, typhoid fever is immediately elim-inated from further diagnostic consideration. Although malaria caused byone Plasmodium species does not present with a double quotidian fever,a mixed malarial infection may be accompanied by a double quotidian feverpattern. For example, in a returning traveler from India, a double quotidianfever eliminates a mixed malarial infection and typhoid fever from diagnos-tic consideration, and the astute clinician should then consider the possibil-ity of visceral leishmaniasis.
Table 2
Physiologic pulse-temperature relationships
Appropriate temperature Pulse rate (beats/min) Pulse in relative bradycardiaa
106�F (41.1�C) 150 !140
105�F (40.6�C) 140 !130
104�F (40.7�C) 130 !120
103�F (39.4�C) 120 !110
102�F (38.9�C) 110 !100
a In adults with temperature O102�F and not on b-blockers, verapamil, diltiazem, or with
pacemaker pulses/second/third degree heart block.
Data from Cunha BA. Antibiotic essentials. 6th edition. Royal Oak (MI): Physicians Press;
2007.
1146 CUNHA
Among noninfectious causes of FUO, a double quotidian fever pattern isa key diagnostic finding in adult Still’s disease. Patients with adult Still’s dis-ease often present as an FUO without many multisystem symptoms or find-ings. If the clinical syndromic presentation includes adult Still’s disease thena double quotidian fever pattern is a key diagnostic finding because no otherrheumatic-inflammatory disorder is associated with a double quotidianfever.
In febrile patients, double quotidian fevers may be artificially induced byintermittent antipyretic medications; devices (eg, hypothermia blankets); orother body cooling mechanisms. Before using a double quotidian fever pat-tern as diagnostic sign the clinician must be sure that the patient has notbeen subjected to antifever medications or maneuvers [29,30].
Camelback (dromedary) feversA camelback or dromedary fever curve is one that has a few days with
fever, separated by a decrease in fever between the febrile episodes overthe period of a week. Graphed on temperature chart the two periods of tem-perature prominence are separated by a period of decreased temperatures,resembling a two-humped camel or dromedary silhouette. As with other un-usual fever curves, camelback fever patterns are of most use when the differ-ential diagnosis includes obscure otherwise difficult-to-diagnose infectionspresenting as FUOs. A camelback fever curve may occur in leptospirosis,brucellosis, and ehrlichiosis [29,30].
Relapsing feversRelapsing fevers refer to those that are recurring and separated by periods
with low-grade fever or no fever. Rat-bite fever, relapsing fever, Bartonella,tuberculosis, and relapsing fever patterns are important in FUOs because,by definition, the fever in patients with FUOs is of long duration (ie, R3weeks). Inherent in the definition of a relapsing fever is the notion that the un-derlying disorder responsible for ongoing fever continues to be clinically ac-tive in terms of its febrile expression. In contrast, recurrent fevers recurperiodically and are associated with fever flares, which is an expression ofthe flare of the underlying disorder (eg, SLE). A relapsing fever patternmay be difficult to appreciate in acute fevers where the duration of the fevermay not permit an appreciation of the relapsing nature of the fever.
Among the infectious causes of FUO, relapsing fever pattern is classicallyassociated with relapsing fever (Borrelia recurrentis) but has also been asso-ciated with typhoid fever, malaria, brucellosis, and rat-bite fever [29,30].
Nonrelapsing fevers may also be caused by a variety of noninfectious etiol-ogies. In theFUOpatient, noninfectious causes of relapsing fever include cyclicneutropenia, familial Mediterranean fever, SLE, vasculitis, hyperimmunoglo-bulinemia D syndrome, and Schnitzler’s syndrome. Relapsing fevers may bemimicked by antipyretic interventions, and by inappropriately or partiallytreated infectious diseases in FUO patients (Tables 3–21) [29,30,32].
1147FUO: DIAGNOSTIC APPROACH
Table 3
Sequence of diagnostic approach to FUO
Initial FUO assessment
Focused FUO assessment (after initial FUO
evaluation)
I. Initial FUO history
A. Initial FUO infectious disease history
PMH-FMH of infectious disease
Pet-animal contact
STD history
Travel
Heart murmur
Surgical-invasive procedures
B. Initial FUO rheumatic history
PMH-FMH of rheumatic disorder
SLE
RA
Gout
Sarcoidosis
HA, mental confusion
Eye symptoms
Neck or jaw pain
Sore throat
Mouth ulcers
Acalculous cholecystitis
Abdominal pain (intermittent, recurrent)
Heart murmur
Myalgias, arthralgias
Joint swelling, effusion
C. Initial FUO neoplastic history
PMH-FMH of malignancy
Night sweats
Decrease in appetite with weight loss
Fundi
D. Initial FUO miscellaneous history
Drug, medication, fume or exposure
Alcoholism
Thyroid, autoimmune disorders
IBD
II. Initial FUO physical examination
A. Infectious disease physical
examination
Fever pattern
Fundi
Nodes
Liver tenderness, hepatomegaly
Spleen tenderness, hepatomegaly
B. Rheumatic disease physical
examination
I. ID etiology suspected based on focused
ID history and physical examination
A. See Tables 3–10
II. RD etiology suspected based on focused
RD history and physical examination
A. See Tables 11–14
III. ND etiology suspected based on focused
ND history and physical examination
A. See Tables 15,16
IV. Miscellaneous disorders suspected based
on a negative focused ID, RD, ND
history and physical examination
A. See Table 17
V. Definitive FUO laboratory tests
A. ID suspected
TTE-TEE (if heart murmur)
Naprosyn test (if DDx between ND and ID)
Special blood culture-media incubation
Specific relevant serology
Tissue biopsy of appropriate nodes, liver,
bone marrow, and so forth
B. Rheumatic disorder suspected
TTE-TEE (if heart murmur)
Specific relevant serology
Tissue biopsy of appropriate nodes, liver,
bone marrow, and so forth
C. Neoplastic disorder suspected
TTE-TEE (if heart murmur)
Tissue biopsy of appropriate nodes, liver,
bone marrow, and so forth
Naprosyn test (if DDx between neoplastic
and infectious disorders)
D. Miscellaneous disorder suspected
History, physical examination, and
laboratory tests negative for infectious,
rheumatic, or neoplastic disorders
Individualized tests for obscure causes
(continued on next page)
1148 CUNHA
Fever of unknown origin: diagnostic usefulness of the Naprosyn test
The Naprosyn test was first developed by Chang [32], an oncologist. Us-ing Naprosyn (naproxen) over a 3-day period (375 mg orally twice daily) hewas able to differentiate neoplastic from infectious FUOs. The Naprosyntest is considered positive when there is a rapid or sustained defervescenceduring the 3 days of the test period. Fever in patients with neoplastic disor-ders recurs after cessation of the Naprosyn test. Those with infectious dis-eases undergo little or no drop in their temperatures during the test
þ, Usually present; �, may be present; –, usually not present.
* When presenting as an FUO.a Streptobacillus moniliformis.b If SBE, CNS coccidiomycosis may be only manifestation of disseminated coccidiomycosis.c In females.
ciency virus; KA, kala-azar (visceral leishmaniasis); SBE, subacute bacterial endoc
fever.
þ, Usually present; �, may be present; –, usually not present.
* When presenting as an FUO.a Pulse temperature deficit.b Vitreitis (‘‘headlight in the fog’’).c Phlyctenular keratoid conjunctivitis.d Subdiaphragmatic.e If SBE.f If liver abscess large.g If Addison’s disease.
þ, Usually present; �, may be present; –, usually not present.
* When presenting as an FUO.a Pulse temperature deficit.b If mixed malarial infection.c Conjunctival nodules.d Peripheral hypopigmented ‘‘histoplasmosis spots.’’e Retinal hemorrhages.f Cytoid bodies.g If SBE.h If myocarditis.i Streptobacillus moniliformis.j Streptobacillus minus.k If Addison’s disease.
Table 8
CMV EBV HIV CSD TOXO
– – – –
þ þ þ – –
þ þ þ – –
þ þ – – þ– – – – –
– – – – –
þ þ þ – –
– – – – –
– – – – –
– � � – �
– – þ – –
þ þ – – �– – – – –
D, cat-scratch disease (Bartonella); EBV, Ep-
ceral leishmaniasis); LFTs, liver function tests;
ase; SGPT, serum glutamic-pyruvic transami-
ic fever.
1155
FUO:DIA
GNOSTIC
APPROACH
Common focused infectious disease causes of FUO: clues from initial laboratory tests*
þ, Usually present; �, may be present; –, usually not present.
* When presenting as an FUO.a b12 is normal in HL, but may be Y in NHL; RF, rheumatoid factor.b May be the first sign of blast transformation in CML; AIHA, autoimmune hemolytic anemia.
mplement; DVTs, deep vein thrombosis; ESR, erythrocyte sed-
n, ¼normal; SGOT, serum glutamic-oxaloacetic transaminase;
tithyroid peroxidase.
1177
FUO:DIA
GNOSTIC
APPROACH
[ TPO titers
[ ATG titers – – – þ –
Urine
Temperature less than rectal
temperature
– – – – –
[ urine mevalonic acid,
neoptermin levels
– – – – –
Tissue biopsy
Thyroid – – – � –
Ileum – þ – – –
Liver � þ – – –
þ Doppler ultrasound (LE) – – – – –
Abbreviations: ATG, antithyroglobulins; CBC, complete blood count; C3/CH50, serum co
imentation rate; FSP, fibrin split products; LE, lower extremity; LFTs, liver function tests;
SGPT, serum glutamic-pyruvic transaminase; SPEP, serum protein electrophoresis; TPO, an
þ, Usually present; �, may be present; –, usually not present
* When presenting as an FUO.a Pulse-temperature deficit.b May be persistent during fever, attacks.c Presents with Crohn’s disease, but not ulcerative colitis.d Cervical adenopathy.e [ IgD.
Tab
Com
Infe
dise
Specific
[IgMtiters CSF
AFB
culture
DX
aspirate-biopsy
SBE þb – – –
Ab – – – þCN – þ þ –
Ren – – þ –
Mil – – þc þc
Typ þ – – þd
CM þ – – þEB þ – – þHIV þe – – –
CSD þ – – þTO þ – – þ
m; CSD, cat-scratch disease (B henselae); CSF, ce-
reb irus; KA, kala-azar (visceral leishmaniasis); MRI,
ma cic-transesophageal echocardiogram; TOXO, toxo-
þ, Usually present; �, may be present; –, usually not present.
* When presenting as an FUO.a If not already done as part of initial FUO diagnostic tests.b For culture-negative SBE pathogens (Q fever, brucellosis, and so forth).c AFB smear or culture of liver, lymph nodes, or bone marrow.d Blood, urine, stool, liver, or bone marrow.e HIV serology or viral load.f If SBE.
Table 20
Uncommon and rare infectious disease causes of FUO: further focused infectious disease diagnostic tests*
(S moniliformis, S minus); RF, relapsing fever (B recurrentis); TRICH, trichinosis; WD, Whipple’s disease.
þ, Usually present; �, may be present; –, usually not present.
* When presenting as an FUO.a If not already done as part of initial FUO diagnostic tests.b If heart murmur or signs of SBE.c [ phase I-II titers.d With HIV only.e Liver, spleen, node, muscle, or bone marrow.f With CNS abnormalities.g S moniliformis.h S minus.i Highly [ IgG C trachomatis L1-3 titers diagnostic.j Lymph node.k PCR or PAS þ stain of small intestine, nodes, or heart valve.
Table 2
Diagno
Obscur Diagnostic findings
Infected [ Uptake of wire, generator
Aortitis [ Uptake in aorta
Infected
infec
[ Uptake in graft, aneurysm
Graft, aneurysm periluminal thickening or
collection
Chroni Mucosal thickening, air fluid level
Relapsi Abnormal mastoid
Periapi Periapical collection
[ Periapical uptake
Subacu [ Vertebral uptake
Vertebral osteomyelitis
Chikun Small joint arthritis
[ Uptake posterior cervical nodes
Abb g; PET, positron emission tomography.
þ, U* W
1180
CUNHA
1
stic tests for selected uncommon infectious disease causes of FUO*
e causes of FUOs Tests that may provide clues
pacemaker wire or generator Indium scan or PET scan
Intermittently or persistently þ BCs
(infective) PET scan
aortic aneurysm or graft or
ted AV graft
Indium scan or PET scan
CT/MRI scan of aneurysm/graft
c sinusitis Head CT/MRI
ng mastoiditis Head CT/MRI
cal dental abscess Panorex radiograph of jaws
Gallium scan of jaws
te vertebral osteomyelitis Bone scan
CT/MRI of spine
gunya fever CT/MRI of hands and feet
Gallium/indium scan
reviations: AV, arteriovenous; CT, computer tomography; MRI, magnetic resonance imagin
sually present; �, may be present; –, usually not present.
hen presenting as an FUO.
1181FUO: DIAGNOSTIC APPROACH
period. The Naprosyn test may use other nonsteroidal anti-inflammatorydrugs, but experience with these drugs is limited. The Naprosyn test isvery useful diagnostically, but if the differential diagnosis of an FUO is be-tween a neoplastic and infectious disorder the Naprosyn test is unhelpful indifferentiating neoplastic from noninfectious disorders (ie, rheumatic, in-flammatory, or miscellaneous disorders) [32,33].
Fever of unknown origin: definitive evaluation
Definitive diagnostic testing is done in the third or final phase of diagnos-tic FUO evaluation. In patients with an appropriate epidemiologic history,serologic tests for visceral leishmaniasis should be obtained. Most infec-tious, rheumatic-inflammatory, neoplastic, and miscellaneous disordersshould be diagnosed after an initial and focused diagnostic FUO evaluation.The disorders not diagnosed to this point are uncommon causes of FUOand require special testing or tissue biopsy for diagnosis as guided by non-specific laboratory test abnormalities and pertinent features of the focusedFUO history and physical examination [1,14,15,34–38].
Invasive diagnostic tests
Liver biopsy
If there are signs and symptoms in a presenting FUO syndrome complexthat suggest liver involvement, then liver biopsymay be diagnostically helpful.Liver biopsy is most useful in granulomatous hepatitis where the differentialdiagnosis may be useful in differentiating granulomas caused by infections,rheumatic-inflammatory disorders, or neoplastic causes. A liver biopsy maybeuseful in diagnosing suspectedmiliary tuberculosis as a causeofFUO [4,39].
Lymph node biopsy
Lymph node biopsy is most useful to diagnose lymphomas. Anterior cer-vical, axillary, or inguinal nodes should not be biopsied if at all possible be-cause the pathology is invariably reported as ‘‘nonspecific/reactive cannotrule out lymphoma.’’ The preferred nodes to biopsy are the posterior cervical,epitrochlear, or supraclavicular nodes. Lymph node pathology is diagnosticwith lymphoma, lymphogranuloma venereum, toxoplasmosis, and Kikuchi’sarteritis. Granulomas in lymph node biopsies may represent a granulomatousdisorder (eg, tuberculosis, sarcoidosis) and lymphoma [40,41].
Bone marrow biopsy
Bone marrow biopsy, as with liver biopsy, may be helpful diagnosticallywith disorders that are associated with bone marrow abnormalities. Bonemarrow biopsy is of importance in diagnosing various neoplastic disorders
1182 CUNHA
(eg, preleukemia, multiple myeloma when other tests are negative). Bonemarrow biopsy is also useful in detecting intracellular infectious pathogensassociated with FUO (eg, disseminated histoplasmosis). Bone marrow bi-opsy is also useful diagnostically in cases of suspected miliary tuberculosis.Bone marrow biopsy is also helpful in a variety of miscellaneous disordersnot usually associated with abnormal bone marrow findings (eg, temporalarteritis) [42–44].
Exploratory laparotomy
With the advent of sophisticated serologic tests and imaging and variousimaging modalities, the necessity for exploratory laparotomy has beenlargely eliminated. Because the initial work-up of the FUO patient includesabdominal and pelvic CT and MRI scanning and total body gallium-indiumscanning, exploratory laparotomy is useful primarily to obtain lymph nodeor organ biopsies that are otherwise unobtainable. Blind exploratory lapa-rotomy has a low diagnostic yield. The clinical syndromic presentationand the pattern of physical and laboratory abnormalities determines the pat-tern of organ involvement, which should guide the surgeon to the appropri-ate tissue when doing an exploratory laparotomy [45,46].
Fever of unknown origin: approach to undiagnosed and recurrent disorders
after a focused evaluation
Even some rare disorders may be potentially diagnosed during the initialand focused FUO evaluation. The serum protein electrophoresis may sug-gest otherwise unsuspected sarcoidosis, hyperimmunoglobulinemia D syn-drome, or Schnitzler’s syndrome. The serum protein electrophoresis withincrease in IgD accompanied by a decrease in IgA should suggest hyperim-munoglobulinemia D syndrome. Schnitzler’s syndrome is suggested bya monoclonal increase in IgM antibodies. Polyclonal gammopathy seenon the serum protein electrophoresis should suggest previously undiagnosedcauses of FUO including sarcoidosis, lymphogranuloma venereum, or atrialmyxoma (if heart murmur was missed). The ESR may be helpful in indicat-ing trichinosis. Patients with trichinosis presenting as an FUO may no lon-ger have eosinophilia. There are no causes of FUO with polymyositis thatare associated with a subnormal ESR rate approaching 0 mm/h.
Highly elevated ESR rate accompanied by elevated fibrin split productsshould prompt further specific testing for deep vein thrombosis and smallpulmonary emboli. Isolated cervical adenopathy not previously diagnosedduring the initial or focused phases of FUO evaluation should raise the pos-sibility of toxoplasmosis, Kikuchi’s disease, or pseudolymphoma. Furtherspecific diagnostic testing and tissue biopsy can be done to diagnose defin-itively each of these entities. In patients with an appropriate reason and zo-onotic contact history, serologic tests may be sent for brucellosis, Q fever,
1183FUO: DIAGNOSTIC APPROACH
and leptospirosis. When presenting as FUOs, both Q fever and to a lesserextent brucellosis may present as ‘‘culture-negative endocarditis.’’ If rodentor rat bite exposure is present, then relapsing fever and rat-bite fever maybe diagnosed by blood smear or culture. Symptoms referable to the greatvessels in a patient with vasculitis should suggest the possibility of Takaya-su’s arteritis for testing with a positron emission tomography scan evalua-tion. Familial Mediterranean fever is suggested by migratory recurrenceserositis in patients of Mediterranean descent. Appropriate genetic studiesmay be sent to confirm the diagnosis of familial Mediterranean fever[47,48].
A head CT reveals obscure chronic sinusitis; relapsing mastoiditis; apicaldental abscesses; and primary, metastatic, and central nervous system neo-plasms. There may be abnormalities on the head CT and MRI to suggesttuberculous meningitis or hypothalamic abnormalities. Serologic tests forhistoplasmosis and coccidiomycosis should be obtained if the presentingsyndrome complex and exposure history to these endemic mycoses indicatesthe diagnostic possibility.
Two diagnoses that are not readily diagnosable during the initial and fo-cused FUO evaluation are Whipple’s disease and factitious fever. Whipple’sdisease is a particularly difficult diagnosis because intestinal biopsy is re-quired for confirmation for a tissue diagnosis and specific serologic testsmust be ordered to confirm the diagnosis. Whipple’s disease should be con-sidered as a cause of FUO in patients with prolonged fever, mental statuschanges, arthritis symptoms, and diarrhea or malabsorption. Whipple’s dis-ease may also present as true culture-negative endocarditis. Whipple’s dis-ease may be diagnosed by demonstrating periodic acid–Schiff positivematerial in macrophages in small intestinal biopsies. Polymerase chainreaction may be used to diagnose Whipple’s disease in tissue samples ofthe small intestine or heart valve (if the valve was replaced because ofendocarditis).
Even a focused diagnostic FUO work-up can miss certain rare causes ofFUO. Cervical carcinoma and colon carcinoma may be missed during theFUO evaluation. If FUO patients have persistent fever and no other diag-nosis, then these diagnoses should be considered in the appropriate patientsetting and diagnosis confirmed by endoscopy or biopsy of the cervix.
Factitious fever is rare cause of FUO. It occurs most commonly in youngadults, usually in medical personnel. Factitious fever may be suspected dur-ing the initial or focused phases of FUO evaluation on the basis of negativehistory, physical, and nonspecific laboratory findings. If these FUO evalua-tions are a noncontribution, then factitious fever should be suspected in theproper clinical context. Factitious fever may be diagnosed by comparingrectal and oral temperatures with temperature of urine. Urine reflects thecore temperature of the individual and in patients with factitious fevershould be normal, or the temperature obtained from other sites is elevatedby one means or another. Patients with factitious fever are inventive and
1184 CUNHA
have many ways to alter temperature recordings. An obvious clue to the fac-titious fever is a pulse-temperature deficit or relative bradycardia. In suchpatients, the rectal temperature should be taken under direct observationas should the urinary specimen collection to avoid maneuvers that alter tem-peratures [1,10,14,15,48].
The three-tiered focused diagnostic approach diagnoses all but the mostunusual causes of FUO. The only disorders missed by focused FUO evalu-ation are very rare entities. After a focused FUO diagnostic approach, how-ever, there are no more than half a dozen obscure disorders that may bepursued depending on the patient’s age, ethnicity, and so forth, or for exam-ple the periodic fever syndromes. The phased diagnostic approach elimi-nates ‘‘shotgun testing’’ and undue reliance on laboratory testing at theexpense of an FUO-relevant and detailed history and physical examination.Clinicians are often unaware of the diagnostic significance of certain phys-ical findings in evaluating FUO patients. The significance of nonspecific lab-oratory tests is enhanced when considered together to increase diagnosticspecificity. Although abnormalities are nonspecific, most entities responsiblefor FUO have several abnormalities, the significance of which taken to-gether may point to the diagnosis. Laboratory tests should not be excessive;rather, they should be focused and comprehensive in the category dealingwith differential diagnostic possibilities. A complete and detailed historyand physical examination not relevant to FUO evaluation is unhelpful. Cli-nicians should endeavor to become more familiar with the causes of FUOthat are not readily diagnosable by tests and that have subtle or uncommonfindings that may be the only clues to diagnosis [47–50].
Summary
FUOs usually are limited by their progression and are self-terminating orare terminated with effective therapy. Some causes of FUO are prone to re-currence. In the main, recurrent FUOs are most often caused by rheumatic-inflammatory etiologies. Patients with infectious FUOs usually resolve withor without therapy in less than a year. Neoplastic disorders usually presentthemselves in less than 1 year but some disorders may recur episodicallyover a prolonged period of time (eg, preleukemias, myeloproliferative disor-ders). Some infectious diseases are prone to recur (eg, relapsing fever). Asa general rule, the longer that an FUO remains undiagnosed, the less likelyit is caused by an infectious etiology [49,50].
Using a focused diagnostic approach a three-tiered system leaves very fewdisorders undiagnosed. Most of the common causes of FUO are diagnosedduring the initial FUO evaluation. The focused FUO evaluation should beable to diagnose less common and obscure disorders associated with pro-longed and perplexing fevers. The objective of the focused diagnostic evalua-tion is to prompt the clinician to order specific diagnostic tests to rule out or
1185FUO: DIAGNOSTIC APPROACH
confirm various causes of FUO in the differential diagnosis based on the clin-ical syndromic presentation. Definitive FUO evaluation should be diagnosticfor nearly all infectious, rheumatic-inflammatory, neoplastic, and miscella-neous causes of FUO. Some causes of FUO remain obscure even after sucha focused and relevant FUO work-up. Clinicians faced with obscure causesof FUO that remain undiagnosed should consult the FUO literature to eval-uate systemically each of these very rare diagnostic possibilities [14,15,28].