Welcome to the all-new Vetlearn

  • Vetlearn is becoming part of NAVC VetFolio.
    Starting in January 2015, Compendium and
    Veterinary Technician articles will be available on
    NAVC VetFolio. VetFolio subscribers will have
    access to not only the journals, but also:
  • Over 500 hours of CE
  • Community forums to discuss tough cases
    and networking with your peers
  • Three years of select NAVC Conference
  • Free webinars for the entire healthcare team

To access Vetlearn, you must first sign in or register.


  Sign up now for:
Become a Member

Compendium January 2009 (Vol 31, No 1)

The Diagnostic Approach to Fever of Unknown Origin in Dogs

by Julie Flood

    CETEST This course is approved for 3.0 CE credits

    Start Test


    Identifying the cause of a fever of unknown origin (FUO) in dogs presents a considerable diagnostic challenge. The diagnostic workup can be frustrating for veterinarians and clients, especially when it fails to reach a final diagnosis after extensive testing. Fortunately, most causes of FUO can be found or treated successfully. This article discusses FUO in dogs and provides information about common causes, the diagnostic approach, and potential treatments.

    True fever (pyrexia) is defined as an increase in body temperature due to an elevation of the thermal set point in the anterior hypothalamus secondary to the release of pyrogens.1 With hyperthermic conditions other than true fever, the hypothalamic set point is not adjusted.1 Nonfebrile hyperthermia occurs when heat gain exceeds heat loss, such as with inadequate heat dissipation, exercise, and pathologic or pharmacologic causes.1

    Dogs with true fever typically have body temperatures between 103°F and 106°F (39.5°C to 41.1°C).2 Prolonged body temperatures above 106°F are dangerous and can result in organ failure, disseminated intravascular coagulation, systemic inflammatory response syndrome, and death.1,3 Such temperatures are usually seen with nonfebrile causes of hyperthermia rather than with true fever.4 Temperatures less than 106°F are unlikely to be harmful and may be beneficial because they constitute a protective response to inflammation.1,5

    The term fever of unknown origin (FUO) is used liberally in veterinary medicine.5 It should be used to identify a fever that does not resolve spontaneously, that does not respond to antibiotic treatment, and for which the diagnosis remains uncertain after an initial diagnostic workup.5 Along with a thorough history and physical examination, initial diagnostics should include a complete blood count (CBC), serum biochemistry profile, and urinalysis with antimicrobial culture. The cause of fever in most dogs is an infection that either is found during the initial workup or responds to antibiotic treatment; therefore, most dogs do not have a true FUO.5

    Differential Diagnosis

    The differential diagnosis for FUO in dogs is extensive, and development of an algorithm covering all causes is not feasible. Some causes of FUO in dogs are listed in Box 1 .2,4,5 Most FUOs are caused by a common disease presenting in an obscure fashion.6

    Current information in the veterinary literature regarding FUO in dogs is limited.1,5 Infectious, immune-mediated, and neoplastic diseases are all important and common causes.2,5,7,8 About 10% to 15% of FUOs in dogs remain undiagnosed despite thorough diagnostic evaluation.5 The prognosis for undiagnosed FUO in dogs is not known. However, a retrospective study7 revealed that in 13 of 14 dogs with undiagnosed FUO, the fever either resolved spontaneously or responded to antibiotics, NSAIDs, or corticosteroids.

    Clinical Approach

    The diagnostic approach must be tailored to the patient. It should be guided by history and physical examination findings, simple laboratory testing, and the potential causes common to the geographic location.9,10 A three-stage approach, such as the one presented in Box 2 , is commonly used.2,4,5 Communication with the owner is of utmost importance to ensure understanding of the time and financial commitment that may be required to obtain a definitive diagnosis. Fortunately, a diagnosis can be obtained in most circumstances, and many causes are treatable or manageable.8

    All medications should be discontinued to help rule out a drug-induced fever. If the fever persists beyond 72 hours after medication cessation, a drug reaction can be ruled out.11

    History and Physical Examination

    Obtaining a thorough history is the first step of a successful diagnostic approach. Clients should be questioned carefully about specific or subtle clinical signs (historical, intermittent, and current) because these may help localize the fever source. A history of stiffness may suggest joint disease, but fevers can present similarly.3 Often, diagnostic clues are not readily apparent on physical examination, so repeated detailed physical examinations are essential (by multiple clinicians, if possible).10 Careful attention should be paid to the whole body—pulses, skin, mucous membranes, oral cavity, lymph nodes, heart, abdomen, bones and joints, and rectum. Repeated fundic and neurologic examinations are also important to identify subtle changes. As the disease progresses, new clues may emerge to help guide the next diagnostic steps.

    Complete Blood Count and Serum Biochemistry Profile

    CBC and serum biochemistry profile abnormalities in dogs with FUO are generally nonspecific, but they may indicate a need for further diagnostic tests. Every CBC should be accompanied by a blood smear evaluation to detect morphologic changes and parasites. Frequently, multiple blood smears and careful scanning are necessary to find infectious organisms (Figure 1). Sometimes only one organism will be seen on an entire slide. It is wise to save serum for serologic testing or other special tests that may be crucial in the future. A serum bile acids assay may be indicated because fever may be the only predominant clinical sign in dogs with portosystemic shunts.12

    Urinalysis with Culture

    A urine sample obtained via cystocentesis (unless contraindicated) should be submitted for a complete urinalysis with bacterial culture even if sediment is inactive. These tests should be repeated, especially if there is a history of lower urinary tract disease, as a negative urine culture does not rule out infection. Further diagnostic testing could include urine protein:creatinine ratio if proteinuria is present with inactive sediment.


    Two-view abdominal and three-view thoracic radiographs should be obtained if the minimum database does not reveal the cause of the fever. Total body radiographs can help aid in the diagnosis of masses, pneumonia, pyothorax, or other infections. Joint radiographs can aid in the diagnosis of an erosive immune-mediated polyarthritis.3 Other anatomic areas to radiograph include long bones (especially in young dogs), the spine, and dental structures (tooth root abscesses, masses). Special contrast radiographic studies can focus on other body systems (urogenital, spinal, gastrointestinal).


    Abdominal ultrasonography allows for evaluation of organ parenchyma and can detect lesions not apparent on survey radiographs. It can also assist with fine-needle aspiration or biopsy if needed. Thoracic ultrasonography can be conducted if abnormalities (e.g., pleural effusion, cysts, masses) are detected on radiographs. When thoracic disease is not radiographically evident, ultrasonography is not rewarding because the lungs obscure intrathoracic anatomy.13 Ultrasonography can also be used to evaluate ocular (including retrobulbar), ventral cervical (thyroid/parathyroid, lymph node, salivary gland), and musculoskeletal (skin, subcutaneous, joint, muscle) regions if indicated.14-16


    Echocardiography should be conducted in dogs with FUO and a heart murmur, especially a new or diastolic murmur. Vegetative lesions must be differentiated from proliferative myxomatous valve degeneration. Dogs with infective endocarditis are usually medium to large breeds that do not tend to have myxomatous valve degeneration.17 Echocardiography can also be used to evaluate for a heart base mass if one is clinically suspected.

    Blood Culture

    Blood cultures should be conducted (preferably during a pyrexic episode) for all dogs with FUO, especially those with a heart murmur, bounding pulses, lameness (polyarthritis), back pain (diskospondylitis), or urinary tract infection, as the latter three conditions can be sequelae to endocarditis.17 It is common for dogs with positive blood cultures to have isolation of the same organism from other tissue or fluid sites (cardiac, urinary, spinal).18 Aseptic techniques for obtaining samples are described elsewhere.2,19,20 The volume of the blood sample is more important than the timing; larger volumes are associated with an increased diagnostic yield in human medicine.19,21,22 Patient size determines the amount of blood to be drawn. As a general guideline, 16 to 20 mL of blood should be obtained from large dogs, and 5 mL of blood should be obtained from cats and small dogs.2 The blood should be divided evenly and placed aseptically into aerobic and anaerobic blood culture vials (~70-mL vials for large patients, and ~20-mL vials for small patients).2 If the patient's size allows, a second blood sample can be obtained immediately from a different site and divided as described above.2 If the dog has recently received antibiotics, blood culture vials with resins that bind antibiotics should be used.2,19 Evidence suggests that recovery is improved in samples from blood culture resin vials because the resins may absorb inhibitors other than antibiotics23; therefore, use of these vials for all blood samples may be warranted. Bartonella spp are emerging as an important cause of culture-negative infective endocarditis in dogs; therefore, submission of samples for Bartonella polymerase chain reaction (PCR) testing as well as serology is recommended in suspected cases.24 Blood culture PCR techniques are being used in human medicine and may be valuable for use in dogs for detecting other infections in the future.25

    Cytologic Examination

    Fine-needle aspiration with cytology should be conducted on any suspicious masses or lymph nodes, fluid accumulations, or abnormal organs. Cytology can be rewarding in the diagnosis of many infections as well as in the identification of abnormal cells ( Figure 2 and Figure 3 ). Impression cytology (nasal planum, skin, feces) can also be conducted if indicated. Fluid samples should be submitted for bacterial culture if the sample quantity is sufficient.

    Bone Marrow Evaluation

    Bone marrow aspiration should be conducted early in the evaluation of dogs with FUO if CBC abnormalities consistent with bone marrow disease are present. It should be considered in later diagnostic stages if no definitive diagnosis has been made, even if the CBC is normal, because neoplasia and infectious diseases can be common causes of FUO in dogs.2,8


    Immune-mediated polyarthritis is a common cause of FUO in dogs even when no signs of arthritis are present (Figure 4).7,8 Arthrocentesis should be conducted on several joints and the samples submitted for cytologic evaluation (EDTA microcontainer) and possibly bacterial culture (aerobic, anaerobic, and mycoplasma). Infectious arthropathy needs to be ruled out if suppurative inflammation is seen on cytology because samples from septic joints do not always contain degenerate neutrophils, and a negative joint culture does not rule out infection.26 If only a small sample can be obtained, it should be used for direct cytology. Otherwise, it is recommended to submit synovial fluid samples in blood culture medium to improve the diagnostic yield.26 Synovial membrane biopsy with culture can also be considered.26 Bacterial endocarditis can cause true infective arthritis or immune-mediated arthritis, and the two conditions must be differentiated.27 Immune-mediated polyarthritis tends to involve the carpi and tarsi, whereas infective arthritis frequently involves larger joints (e.g., stifle, elbow, shoulder).27 If immune-mediated polyarthritis is suspected, serology for rickettsial disease and a heartworm test may be indicated.27


    Serum samples should be submitted for fungal and rickettsial disease testing if these diseases are clinically suspected and if patient history indicates possible exposure. These tests should not be used as screening procedures in the hope that something abnormal will be found. Because Toxoplasma and Neospora spp are ubiquitous protozoal parasites, paired serum antibody titers for IgG and IgM should be submitted if a diagnosis continues to be elusive. A single high IgM indicates active or recent infection, and a fourfold IgG rising titer confirms infection.28 It is important to remember that a negative antibody titer does not rule out infection and that a single positive titer implies either exposure (e.g., fungal disease, most rickettsial diseases) or previous infection (e.g., protozoal disease) and does not necessarily correlate with active disease or current clinical signs.28

    Immunodiagnostic Screening Panels

    Immunodiagnostic panels (antinuclear antibody, rheumatoid factor, Coombs) are typically unrewarding in dogs with FUO for several reasons, including the potential for false-positive results.2,7,8 Antiplatelet antibody tests and serum protein electrophoresis can be conducted if thrombocytopenia or hyperglobulinemia, respectively, is present.

    Other Diagnostic Tests

    Other diagnostic tests, such as prostatic wash, cerebrospinal fluid analysis, and bronchoscopy with bronchoalveolar lavage, should be considered if clinical abnormalities suggest prostatic, neurologic, or respiratory disorders, respectively. Samples should be submitted for cytologic evaluation as well as aerobic and anaerobic bacterial culture if quantity permits. Bronchoalveolar lavage samples should also be submitted for mycoplasma and slow-growing fungal cultures.

    Advanced Imaging

    Computed tomography (CT) and magnetic resonance imaging (MRI) should be used to help delineate diagnosed conditions or when the diagnosis remains equivocal.10 Nuclear scintigraphy is being used more frequently in veterinary medicine to detect infections and may be a valuable tool in the investigation of FUO in dogs.29 Another promising imaging modality being used in human medicine, called image fusion, is the combination of positron emission tomography (PET; a type of nuclear imaging) and CT. A few reports of the use of image fusion in dogs demonstrate that this technique could play an important role in investigating canine FUO.30-33


    If fine-needle aspiration cytology cannot provide a definitive diagnosis, a biopsy may be helpful. In one study,7 biopsy samples submitted for histopathology enabled a diagnosis in 15 of 17 dogs with FUO. Tissue samples can be obtained percutaneously (with or without imaging assistance); via endoscopy, laparoscopy, or thoracoscopy; or surgically during laparotomy. Submission of tissue samples for histopathology and possibly bacterial or fungal cultures is recommended. Exploratory celiotomy with biopsy is indicated only by the results of diagnostic testing.8 The diagnostic yield of exploratory celiotomy in a dog with no indication for surgery is unknown.


    Specific treatment is based on the definitive diagnosis, if found. Administering intravenous fluids or placing a fan blowing toward the cage can be used to reduce the temperature in hospitalized patients. In dogs for which extensive investigation yields no diagnosis, judicious use of antibiotic therapy may be warranted. The choice of antibiotic depends on the suspected bacterial agent. If no response is seen after 72 hours with appropriate dosing, another antibiotic that covers a different spectrum may be chosen.34 If a bacterial infection is suspected in a severely ill patient, the four-quadrant approach—choosing an antibiotic or combination of antibiotics that is effective against aerobic, anaerobic, gram-positive, and gram-negative organisms—is recommended.34 If antibiotic therapy is not successful, NSAIDs can be administered, keeping in mind the potential side effects.5,7 Fever can result in considerable malaise, dehydration, and anorexia; therefore, clinicians must decide in each case whether NSAIDs could be beneficial.4 Antipyretics (e.g., ketoprofen, flunixin meglumine, dipyrone) should be used with caution because fever can be beneficial, and many argue that antipyretic therapy can have a negative impact on the body by causing hypothermia and impairing the host's immune defenses.4,5,35 Fevers may increase the bactericidal effect of antibiotics and serum and can also decrease the pathogenicity of some pathogens.4,5,35 If an antipyretic is considered necessary, aspirin can be administered at a dosage of 10 mg/kg q12h PO.2 If there is clinical suspicion of a fungal disease, trial antifungal agents may be of use.

    If the fever does not respond to antibiotics or antifungals, options include waiting to see if new diagnostic clues arise and considering an immunosuppressive trial of corticosteroids. A dramatic improvement should be expected within 24 to 48 hours of corticosteroid therapy in dogs with an immune-mediated FUO.2,5 It is important to inform owners about the risks of trial corticosteroids, such as allowing a fungal or bacterial infection to disseminate or further decreasing the chance to diagnose the problem, as with lymphoma. Ideally, during a corticosteroid trial, the dog should be hospitalized and monitored closely for adverse effects.5 Initial improvement does not equal successful treatment. One study7 revealed that treatment 24 hours before referral was associated with a statistically significant increase in the time to diagnosis. Therefore, it is suggested that, when possible, therapy be withheld or withdrawn in dogs referred for investigation of FUO.


    The most common and important causes of FUO in dogs are infection, immune-mediated disease, and cancer. Using a logical diagnostic approach to FUO in dogs usually results in a definitive diagnosis. Sometimes, being patient and allowing new diagnostic clues to emerge by revamping historical information (via reassessing current information and possibly obtaining a more detailed history) and repeating physical examinations and simple laboratory tests is more desirable than proceeding with more invasive and expensive tests if the dog is stable. Communicating with the client is of utmost importance. A broad knowledge of the possible causative diseases and the ability to interpret specific diagnostic test results in the context of FUO in dogs are essential to diagnose the source of an FUO.


    The author thanks Leo "Ty" McSherry, DVM, DACVP, clinical pathologist at Antech Diagnostics in Irvine, California, for the cytology images.

    Read the companion article about fever of unknown origin in cats.

    Downloadable PDF

    1. Miller JB. Hyperthermia and fever of unknown origin. In: Ettinger SJ, Feldman EC, eds. Textbook of Veterinary Internal Medicine. Vol 1. 6th ed. St. Louis: Elsevier Saunders; 2005:9-13.

    2. Lunn KF. Fever of unknown origin: a systematic approach to diagnosis. Compend Contin Educ Pract Vet 2001;23(11):976-992.

    3. Bennett D. Diagnosis of pyrexia of unknown origin. In Pract 1995;17:470-481.

    4. Johannes DM, Cohn LA. A clinical approach to patients with fever of unknown origin. Vet Med 2000;95(8):633-642.

    5. Couto CG. Fever of undetermined origin. In: Nelson RW, Couto CG, eds. Small Animal Internal Medicine. 4th ed. St. Louis: Elsevier; 2009:1274-1277.

    6. Dunn JK, Gorman NT. Fever of unknown origin in dogs and cats. J Small Anim Pract 1987;28:167-181.

    7. Battersby IA, Murphy KF, Tasker S, et al. Retrospective study of fever in dogs: laboratory testing, diagnoses and influence prior to treatment. J Small Anim Pract 2006;47:370-376.

    8. Dunn KJ, Dunn JK. Diagnostic investigations in 101 dogs with pyrexia of unknown origin. J Small Anim Pract 1998;39:574-580.

    9. Mourad O, Palda V, Detsky AS. A comprehensive evidence-based approach to fever of unknown origin. Arch Intern Med 2003;163:545-551.

    10. Roth AR, Basello GM. Approach to the adult patient with fever of unknown origin. Am Fam Phys 2003;68:2223-2228.

    11. Johnson DH, Cunha BA. Drug fever. Infect Dis Clin North Am 1996;10:85-91.

    12. Wess G, Unterer S, Haller M, et al. Recurrent fever as the only or predominant clinical sign in four dogs and one cat with congenital portosystemic vascular anomalies. Schweiz Arch Tierheilkd 2003;145(8):363-368.

    13. Mattoon JS, Nyland TG. Thorax. In: Nyland TG, Mattoon JS, eds. Small Animal Diagnostic Ultrasound. 2nd ed. Philadelphia: Saunders; 2002:325-353.

    14. Samii VF, Long CD. Musculoskeletal system. In: Nyland TG, Mattoon JS, eds. Small Animal Diagnostic Ultrasound. 2nd ed. Philadelphia: Saunders; 2002:267-284.

    15. Mattoon JS, Nyland TG. Eye. In: Nyland TG, Mattoon JS, eds. Small Animal Diagnostic Ultrasound. 2nd ed. Philadelphia: Saunders;2002:305-324.

    16. Wisner ER, Mattoon JS, Nyland TG. Neck. In: Nyland TG, Mattoon JS, eds. Small Animal Diagnostic Ultrasound. 2nd ed. Philadelphia: Saunders; 2002:285-304.

    17. MacDonald KA. Infective endocarditis. In: Bonagura JD, Twedt DC, eds. Kirk's Current Veterinary Therapy XIV (Small Animal Practice). St. Louis: Elsevier Saunders; 2009:786-791.

    18. Hirsh DC, Jang SS, Biberstein EL. Blood culture of the canine patient. JAVMA 1984;184(2):175-178.

    19. Haggstrom J, Kvart C, Pedersen HD. Acquired valvular heart disease. In: Ettinger SJ, Feldman EC, eds. Textbook of Veterinary Internal Medicine. Vol 2. 6th ed. St. Louis: Elsevier Saunders; 2005:1022-1039.

    20. Calvert CA, Wall M. Cardiovascular infections. In: Greene CE, ed. Infectious Diseases of the Dog and Cat. 3rd ed. St. Louis: Elsevier Saunders; 2006:841-865.

    21. Li J, Plorde JJ, Carlson LG. Effects of volume and periodicity on blood cultures. J Clin Microbiol 1994;32(11):2829-2831.

    22. Karchmer A. Infective endocarditis. In: Braunwald E, ed. Heart Disease. 5th ed. Philadelphia: Saunders; 1997:1077.

    23. Lelií¨vre H, Gimenez M, Vandenesch F, et al. Multicenter clinical comparison of resin-containing bottles with standard aerobic and anaerobic bottles for culture of microorganisms from blood. Eur J Clin Microbiol Infect Dis 1997;16(9):669-674.

    24. MacDonald KA, Chomel BB, Kittleson MD, et al. A prospective study of canine infective endocarditis in Northern California (1999-2001): emergence of Bartonella as a prevalent etiologic agent. J Vet Intern Med 2004;18:56-64.

    25. Gebert S, Siegel D, Wellinghausen N. Rapid detection of pathogens in blood culture bottles by real-time PCR in conjunction with the pre-analytic tool MolYsis. J Infect 2008;57:307-316.

    26. Greene CE, Budsberg SC. Musculoskeletal infections. In: Greene CE, ed. Infectious Diseases of the Dog and Cat. 3rd ed. St. Louis: Elsevier Saunders; 2006:823-841.

    27. Goldstein RE. Swollen joints and lameness. In: Ettinger SJ, Feldman EC, eds. Textbook of Veterinary Internal Medicine. Vol 1. 6th ed. St. Louis: Elsevier Saunders; 2005:83-87.

    28. Houser G, Ayoob A, Greene CE. Laboratory testing for infectious diseases of dogs and cats. Appendix 5. In: Greene CE, ed. Infectious Diseases of the Dog and Cat. 3rd ed. St. Louis: Elsevier Saunders; 2006:1139-1168.

    29. Moon ML, Hinkle GN, Krakowka GS. Scintigraphic imaging of technetium 99m-labeled neutrophils in the dog. Am J Vet Res 1988;49(6):950-955.

    30. Peremans K, DeWinter F, Janssens L, et al. An infected hip prosthesis in a dog diagnosed with a 99mTC-ciprofloxacin (infection) scan. Vet Radiol Ultrasound 2002;43(2):178-182.

    31. Berry CR, DeGrado TR, Nutter F, et al. Imaging of pheochromocytoma in 2 dogs using p-[18F]fluorobenzylguanidine. Vet Radiol Ultrasound 2002;43(2):183-186.

    32. Ballegeer EA, Forrest LJ, Jeraj R, et al. PET/CT following intensity-modulated radiation therapy for primary lung tumor in a dog. Vet Radiol Ultrasound 2006;47(2):228-233.

    33. LeBlanc AK, Jakoby B, Townsend DW, et al. Thoracic and abdominal organ uptake of 2-deoxy-2-[18F]fluoro-D-glucose (18FDG) with positron emission tomography in the normal dog. Vet Radiol Ultrasound 2008;49(2):182-188.

    34. Lappin MR. Practical antimicrobial chemotherapy. In: Nelson RW, Couto CG, eds. Small Animal Internal Medicine. 4th ed. St. Louis: Elsevier; 2009:1291-1301.

    35. Klein NC, Cunha BA. Treatment of fever. Infect Dis Clin North Am 1996;10(1):211-216.

    References »

    NEXT: The Renin-Angiotensin-Aldosterone System Approaches to Cardiac and Renal Therapy

    CETEST This course is approved for 3.0 CE credits

    Start Test


    Did you know... A bright, alert, afebrile 7-day-old foal that is nursing readily and has soft feces likely has foal heat diarrhea and requires minimal intervention.Read More

    These Care Guides are written to help your clients understand common conditions. They are formatted to print and give to your clients for their information.

    Stay on top of all our latest content — sign up for the Vetlearn newsletters.
    • More