Show Summary Details
Page of

Fever of Unknown Origin 

Fever of Unknown Origin
Fever of Unknown Origin

Emily Shuman

Page of

PRINTED FROM OXFORD MEDICINE ONLINE ( © Oxford University Press, 2022. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Medicine Online for personal use (for details see Privacy Policy and Legal Notice).

date: 05 July 2022

  1. A. Introduction. Fever of unknown origin (FUO) is defined as a temperature greater than 38.3° C for more than 3 weeks that eludes diagnosis after 1 week of in-hospital evaluation. More recent authors have suggested that the duration of time in the hospital be modified to include patients whose illness remains undiagnosed after at least 3 outpatient visits or 3 days in the hospital. These definitions were designed to exclude common undiagnosed viral syndromes and to identify a homogenous population to describe in studies. This definition is arbitrary, and the following approach should be used in patients in whom the etiology of fever is obscure despite an initial workup.

  2. B. Common Causes of Fever of Unknown Origin

    Table 51.1 Common Causes of Fever of Unknown Origin

    Disease Category



    Bacterial, mycobacterial, viral, fungal, parasitic, spirochetal, rickettsial


    Lymphomas, leukemias, hepatic cancer, renal cancer, atrial myxoma

    Connective tissue disorders

    Systemic lupus erythematosus, rheumatoid arthritis, Still’s disease, vasculitides


    Inflammatory bowel disease, granulomatous hepatitis

    1. a. Infections should always be considered first when evaluating FUO.

      1. i. Bacterial infections

        1. 1. Intra-abdominal infections (e.g., in the liver, spleen, or kidney) are especially common causes of FUO.

        2. 2. Osteomyelitis and sinusitis are localized infections that may elude initial diagnosis.

        3. 3. Bacteremias from endocarditis, salmonellosis (typhoid fever), and brucellosis commonly cause FUO. Intravascular graft infections present with bacteremia and can be very difficult to diagnose.

        4. 4. Mycobacterial infections

          • a. Tuberculosis is a common cause of FUO and is the most common cause worldwide. HIV-infected patients have a higher rate of extrapulmonary tuberculosis.

          • b. Mycobacterium avium complex infections are the leading cause in the United States of FUO in patients with acquired immune deficiency syndrome (AIDS). CD4 counts are typically <50 in this setting.

        5. 5. Spirochetal infections. Secondary syphilis, leptospirosis, and Lyme disease should be considered.

        6. 6. Rickettsial and related infections include the typhus group (epidemic, endemic, and scrub typhus), the spotted fever group (Rocky Mountain spotted fever, rickettsial pox, and African tick-bite fever), and other similar illnesses (Q fever, trench fever, and ehrlichiosis).

      2. ii. Viral infections

        1. 1. Cytomegalovirus (CMV) infection is a common cause of FUO. Acute CMV infection in immunocompetent patients often presents without localizing signs.

        2. 2. Epstein-Barr virus causes mononucleosis (especially in adolescents and young adults where it is often called the “kissing” disease).

        3. 3. HIV. In HIV-infected patients, you should use the CD4 count to help predict the disease processes a patient is likely to have (see Chapter 55). Mycobacterium avium complex infections and CMV infections are common causes of FUO in AIDS patients with low CD4 counts (i.e., less than 100 cells/µL).

      3. iii. Fungal infections

        1. 1. Histoplasmosis and coccidioidomycosis can cause FUO in immunocompetent or immunocompromised patients.

        2. 2. Candidiasis and aspergillosis are opportunistic fungal infections that may cause FUO.

      4. iv. Parasitic infections. Amebiasis, malaria, and toxoplasmosis may cause FUO and are common causes of FUO in developing countries.

    2. b. Neoplasms. Lymphomas, leukemias, hepatic and renal cancers, and atrial myxomas are often associated with febrile syndromes. Concurrent infection should always be ruled out even if these conditions exist or are diagnosed during evaluation.

    3. c. Connective tissue disorders. Systemic lupus erythematosus (SLE), rheumatoid arthritis, Still’s disease, giant cell arteritis, and the vasculitides (see Chapter 75) can all cause FUO.

    4. d. Other causes. Inflammatory bowel disease, sarcoidosis, and granulomatous hepatitis may cause prolonged fever. Factitious fever and drug fever should also be considered.

  3. C. Approach to the Patient. Because the cause of an FUO is elusive (by definition), there is a tendency to try to “net the diagnosis” by ordering every test available. Proceeding in a logical, stepwise fashion will decrease the number of tests your patient must undergo, the cost of the total workup, and the incidence of false-positive results. Before delving into an extensive evaluation, rule out factitious (patient-induced) fever by observing the elevated temperature reading yourself.

    1. a. Go where the money is! The best bet for making a diagnosis is pursuing any symptoms, signs, or abnormal laboratory test results that present along with the fever. Make sure to always go down the list of possibilities from head to toe as outlined in Chapter 48. For example:

      1. i. Persistent headaches with a normal cerebrospinal fluid analysis. A computed tomography (CT) scan or magnetic resonance imaging (MRI) may be indicated to rule out an intracranial abscess or sinusitis.

      2. ii. Cardiac murmurs associated with negative blood cultures may be evaluated with echocardiography (transthoracic or transesophageal) to rule out culture-negative endocarditis.

      3. iii. Hemoptysis or chronic cough in the presence of an unremarkable chest radiograph may be further evaluated with a chest CT scan or bronchoscopy.

      4. iv. Hematuria may prompt a CT scan or ultrasound to evaluate for the possibility of renal carcinoma (which often causes fever).

      5. v. Lymphadenopathy may be evaluated by biopsy.

      6. vi. Bone pain can be pursued with radiographs and a bone scan or MRI to look for lytic lesions.

      7. vii. Cytopenias can be evaluated by bone marrow biopsy.

        Hot Key

        In HIV-infected patients, use the CD4 count to help predict the disease processes a patient is likely to have (see Chapter 55). Mycobacterium avium complex infections and CMV infections are common causes of FUO in AIDS patients with CD4 counts less than 100 cells/µL.

    2. b. Diagnostic studies

      1. i. Initial considerations

        1. 1. Abdominal and pelvic CT scan. There is no better place for an infection (e.g., an abscess) or malignancy (e.g., liver or renal carcinoma) to hide than in the abdomen or pelvis; therefore, performing this test early in the diagnostic workup may be valuable.

        2. 2. Chest CT. Although occult pulmonary infection is less common than occult abdominal abscess, unexpected findings (including infiltrates, cavitation, and lymphadenopathy) may lead to the diagnosis.

        3. 3. Screens for tuberculosis, Epstein-Barr virus, and syphilis may be performed with a purified protein derivative (PPD) test or interferon-γ‎ release assay (IGRA), a heterophile antibody (Monospot) test, and a serum rapid plasma reagin (RPR), respectively.

        4. 4. Rheumatologic workup. An antinuclear antibody (ANA) assay, rheumatoid factor (RF) assay, and erythrocyte sedimentation rate (ESR) are often requested early in the evaluation. Any rashes that are purpuric should be biopsied to rule out vasculitis (see Chapter 75).

        5. 5. Other tests may be ordered based on epidemiologic exposures. For example:

          • a. Thick and thin blood smears examined every 12–24 hours for a total of three sets are appropriate if the patient has traveled to areas where malaria is endemic.

            Hot Key

            Fever in a international traveler should be considered malaria until proven otherwise.

          • b. Stool for culture or ova and parasite (O&P) examination is also useful in patients with an appropriate travel history.

          • c. Amebic titers can help evaluate chronic intestinal amebiasis in patients with a remote travel history because O&P stool examination is usually negative in patients with chronic disease.

          • d. Toxoplasmosis titers are highly sensitive for end-organ toxoplasmosis.

          • e. Specific serologies or cultures for brucellosis, leptospirosis, Lyme disease, Rocky Mountain spotted fever, and other spirochetal and rickettsial diseases may be sent, depending on the geographic region, exposures to animals or the outdoors, and the clinical presentation.

          • f. Mycobacterium avium complex blood cultures may be performed in HIV-positive patients with low CD4 counts.

      2. ii. Secondary considerations

        1. 1. Radionuclide studies

          • a. Gallium scans, indium-labeled white blood cell (WBC) scans, and PET scans are often used when no etiology has been found. These tests are limited by low specificity.

          • b. Bone scans may be performed to rule out osteomyelitis and primary or metastatic malignancies of bone.

        2. 2. Invasive tests

          • a. Bone marrow biopsies are generally of low diagnostic yield but are often performed when the diagnosis is still in question.

          • b. Liver biopsies have a higher yield in patients with abnormal liver tests.

          • c. Endoscopy (upper and lower) should be performed earlier in the workup if there are any gastrointestinal symptoms, signs, or suggestive laboratory abnormalities (e.g., a positive fecal occult blood test, iron deficiency anemia).

          • d. Temporal artery biopsy may be revealing in older adult patients with FUO, even in the absence of localizing symptoms.

          • e. Bronchoscopy should be performed in patients with suggestive abnormalities on chest radiograph, on chest CT, or by symptoms.

          • f. Exploratory laparotomy in the absence of abnormal findings on imaging studies no longer plays a significant role in the evaluation of FUO.

Suggested Further Readings

Cunha BA, Lortholary O, Cunha CB. Fever of unknown origin: a clinical approach. Am J Med 2015;128:1138.e1–e15.Find this resource:

Mourad O, Palda V, Detsky AS. A comprehensive evidence-based approach to fever of unknown origin. Arch Intern Med 2003;163:545–51.Find this resource:

Petersdorf RG, Beeson PB. Fever of unexplained origin: report on 100 cases. Medicine 1961;40:1–30. (Classic Article.)Find this resource:

Weyand CM, Goronzy JJ. Giant-cell arteritis and polymyalgia rheumatica. N Engl J Med 2014;371:50–7.Find this resource: