Pyrexia of Unknown Origin (PUO)

Written by: Alastair Savage from Aberdeen University,

Definition of pyrexia of unknown origin

 

Pyrexia of unknown origin (PUO) is a term which could be applied in many situations, the current definition however is as follows (1-3):

 

  • Fever of 38.3°C (101°F) or more.

 

  • Lasting for at least three weeks.

 

  • For which no cause can be identified after three days of investigation in hospital or after three or more outpatient visits.

 

This classification indirectly comes from an article written in 1961 by Petersdorf and Beeson, it has been revised to keep up with current medical practice.

Due to the nature of the disease, HIV associated PUO is usually considered separately.

Other types of PUO are Nosocomial (Developed in hospital) and Neutropenic (Immune Deficient).

Important features of the history and examination

 

It is especially important in cases of PUO to take a full and comprehensive history to aid diagnosis. In PUO due to the fact attempts at finding the cause have already been made and failed, it is especially important to be broad and thorough in history taking. It is important to stress to the patient that all symptoms present, transient and past may be relevant. It is only by taking this thorough history and by performing a thorough examination that it will be possible to narrow your diagnostic focus to the relevant area. History is often worth repeating several times if diagnosis remains elusive. It is important to document all symptoms and signs as relevant, as multiple clinicians are often involved. The important areas to remember are:

 

  • Co-morbid conditions as these may pre-dispose to particular causes.
  • Previously treated conditions particularly those which may re-occur such as TB.
  • History of fever, particularly onset, pattern and severity. This may suggest particular diseases, e.g. the pattern of fever may be characteristic in Malaria
  • Thorough surgical history including past complications, this may show increased likelihood of an intra-abdominal abscess or PID. It may be due to a recognised complication of a previous surgery or in the case of surgery overseas due to improperly sterilised conditions.
  • Travel history which countries they have visited and where previously resident. Particularly possible exposure while travelling, e.g. swimming in open water, and vaccination and anti-malarial compliance, as this may point to specific pathogens, such as ticks or other vector.
  • Sexual history and practice.
  • Drug use, increases the likelihood of specific diseases such as infective endocarditis.
  • Animal exposure, may suggest a zoonosis such as dog exposure may lead to hydatid disease (echinococcosis).
  • Occupation and Hobbies, these may expose patients to toxins or infections, for example leptospirosis is common in windsurfers.
  • Unusual dietary habits e.g. rare or undercooked meats, shellfish, well water or soft cheeses. This also helps to establish who else may have become exposed to any possible food borne diarrhoea etc.
  • Allergy history this is significant especially if to antimicrobials, whilst noting how this manifests is equally important.
  • Changes in Medication, as many may cause a fever themselves.
  • Systematic enquiry is particularly important in cases of PUO, note all symptoms including those no longer present.

 

The examination of a patient with PUO is a comprehensive examination of all systems, similar to most patients. A patient with PUO however should be re-examined at least once a day. A full thorough examination of all the systems is important because in cases of PUO one of the reasons a diagnosis may not have been made is because common signs were either missed or absent and uncommon signs not specifically looked for. The examination of a patient with PUO should include:

 

  • Full Respiratory examination
  • Full Cardiovascular examination
  • Full Abdominal/Gastro-intestinal examination
  • Full Neurological examination
  • Full Musculo-Skeletal examination
  • Full ENT examination
  • Full Dermatalogical examination
  • Full Lymphatic examination
  • Full Urogenital examination
  • Fundoscopy

 

All of these systems are important to examine thoroughly as signs may be more subtle, and as the locus of the problem is unknown it is important to assume that it may be in any or indeed every system.

The areas to look out for in particular are:

 

  • Detailed temperature chart to exclude factitious fever.
  • Skin changes; Rashes.
  • Abdominal tenderness.
  • New or altered cardiac murmurs.
  • New or altered arthritis.
  • Tenderness over the temporal artery.

Common causes of PUO (8)

Infection (~30%):

 

  • Abdominal abscess - swinging fever, abdominal mass
  • Extra pulmonary/disseminated TB
  • Infective Endocarditis - heart murmur, splinter haemorrhages, more common in intra-venous drug users.
  • Osteoarticular infections - pain, redness and swelling, may be in atypical joints.
  • Typhoid/enteric fevers - travel particularly to developing areas, diarrhoea and relative bradycardia.
  • Endemic mycosis
  • EBV infection - tonsilar enlargement, cervical lymphadenopathy and hepatosplenomegaly
  • CMV infection - as above
  • Brucellosis - travel to affected areas, exposure to animals or work in food preparation and ingestion of unpasteurised dairy or raw meats
  • Leishmaniasis - travel and exposure to insect bite (sand-fly), skin lesions, hepatosplenomegaly and pancytopaenia.
  • Prostatitis - urinary symptoms, lower back pain, recent UTI or STI, catheterisation
  • Malaria - Travel and exposure to insect bite (mosquito), recurring or intermittent fever, chills and rigors
  • Rickettsial infections - travel and exposure to insect bite (lice, fleas or tic), skin lesions or GI changes
  • Dental abscess - swelling or pain
  • Chronic sinusitis - tenderness on palpation of sinuses, mucosal inflammation.

 

Miscellaneous (~20%)

 

  • Drug fever - recent change in medication
  • Factitious fever
  • Mediterranean familial fever - at risk ethnic group, recurring periodic fever, usually begins when young
  • Deep vein thrombosis/Pulmonary embolism
  • Hyperthyroidism

 

Neoplasm (20%):

 

  • Lymphoma - weight loss, night sweats, anorexia, fatigue and lymphadenopathy
  • Hepatoma - hepatitis or alcoholism history, pruritus, jaundice, RUQ pain
  • Hepatic metastasis
  • Renal cell carcinoma - fatigue, weight loss, hypertension, may present only with fever
  • Leukaemia - bruising, fatigue, lymphadenopathy, hepatosplenomegaly, night sweats and weight loss
  • Colon cancer - weight loss, altered bowel habit, anaemia, mass
  • Pancreatic cancer - radiating abdominal pain/back pain, jaundice, acute pancreatitis, weight loss

 

Connective tissue disorders (~15%)

 

  • Systemic lupus erythematosus - many and varied, see other article
  • Adult onset Still's disease - arthralgia, 'salmon pink rash' and lymphadenopathy
  • Autoimmune hepatitis - fatigue, myalgia, nausea and mild pruritus
  • Systemic vasculitis - purpura, ulcers, signs of organ ischaemia
  • Mixed connective tissue disease - Raynaud's, lethargy, polyarthritis and vasculitic rash
  • Polymyalgia rheumatica - age > 50, shoulder or pelvic girdle pain, morning stiffness
  • Inflammatory bowel disease - diarrhoea, mouth ulcers, abdominal pain, anaemia
  • Sarcoidosis - night sweats, malaise, weight loss, usually lung involvement
  • Kikuchi's disease - lymphadenopathy, rash and headache

 

Idiopathic (~15%)

Important investigations

 

The investigation of the patient with PUO should, like all cases, be driven by the findings in the history and examination. Testing should be done for diseases which the patient is at higher risk of having or has been exposed to. 

 

There is no gold standard list of minimum investigation; however the following have been widely considered recommendable (8):

 

  • Full Blood Count - look for signs of infection, and any anaemia that may affect management
  • Urea & Electrolyte's - look for disturbances of the urinary tract or kidneys, which may be the result of disease or merely affect management
  • Creatinine
  • Erythrocyte Sedimentation Rate / C-Reactive Protein - elevated suggests infection or other inflammatory cause, e.g. auto-immune
  • Liver Function Test's - may suggest liver disease or just affect management plan
  • Urinalysis
  • Blood cultures x 3 - show bacteraemia, but contaminants may obscure accurate interpretation
  • Anti-Nuclear Antibody test, Antineutrophilic cytoplasmic antibody, Rheumatoid factor - All highly suggestive of auto-immune conditions
  • Chest X-ray - may show disease of the chest e.g. pneumonia or TB
  • Abdominal ultrasound or contrast CT - Shows abdominal pathology such as abscess or diverticula

 

Any lesions which are identified should be immediately biopsied to obtain culture samples to aid in the diagnosis of diseases such as TB.

If these tests fail to provide diagnostic clues then more specialist testing is required, such as echocardiography, serological tests (cryoglobulins, complement levels), positron emission tomography and further microbiological tests.

Causes of pyrexia of unknown origin

 

There are many causes of pyrexia of unknown origin. They are generally grouped into four main categories

 

  • Infection (~15-30%)
  • Neoplasia (~10-30%)
  • Connective tissue disease (~33-40%)
  • Miscellaneous (~20-30%)

 

Recent studies have shown in the developed world that cases of PUO due to infective and neoplastic causes are falling due to better culturing and imaging techniques. This does not however hold true in the less developed world.

Children and younger adults are more likely to suffer from infectious diseases, in particular viral infections. Whereas older adults are more likely to suffer more specific infectious diseases such as TB, intra-abdominal abscesses and infective endocarditis or non-infectious diseases such as neoplasms and polymyalgia.

It has been extensively proven that PUO is more often due to atypical presentation of a common disease than an uncommon or exotic disease (6,7).

Causes of PUO (4,5)

Management and prognosis of patients with Pyrexia of unknown origin

 

Management of pyrexia is usually focused on correcting the cause. For some patients however a definite cause is never found. In these cases when the patient is systemically well, a wait and see approach is often taken. During this time it is important to keep repeating the history and examination, and where appropriate investigations. It is often the case that cases of pyrexia of unknown origin in which no diagnosis can be made resolve spontaneously over time (9).

The mortality rate of PUO is roughly 30-40%, which is usually attributed to malignancy in older patients. Mortality rates among the idiopathic are usually low.

If no diagnosis can be made and the patient is unstable or neutropenic then it is appropriate to start antibiotic therapy. If the cause is suspected to be TB then tuberculosis medication should be given. Rifampicin may suppress fever regardless of cause. Empirical steroid treatment should be avoided.

Important points to remember

 

  • Fever of over 38.3 degrees, undiagnosed after at least 3 weeks or 3 outpatient appointments.
  • History and Examination needs to be broad and repeated.
  • More likely to be a common disease presenting atypically.
  • Treat based on suspicion if the patient is systemically unwell.

References

  1. Petersdorf RG. Fever of unknown origin. An old friend revisited. Arch Intern Med 1992;152:21-2.
  2. Knockaert DC, Vanderschueren S, Blockmans D. Fever of unknown origin in adults: 40 years on. J Intern Med 2003;253:263-75.
  3. Durack DT, Street AC. Fever of unknown origin—reexamined and redefined. Curr Clin Top Inf Dis 1991;11:35-51.
  4. Efstathiou SP, Pefanis AV, Tsiakou AG, Skeva II, Tsioulos DI, Achimastos AD, Mountokalakis TD. Fever of unknown origin: discrimination between infectious and non-infectious causes. Eur J Intern Med 2010;21:137-43.
  5. Hot A, Jaisson I, Girard C, French M, Durand DV, Rousset H, et al. Yield of bone marrow examination in diagnosing the source of fever of unknown origin. Arch Intern Med 2009;169:2018-23.
  6. Arnow PM, Flaherty JP. Fever of unknown origin. Lancet 1997;350:575-80.
  7. Bleeker-Rover CP, Vos FJ, de Kleijn EM, Mudde AH, Dofferhof TS, Richter C, et al. A prospective multicenter study on fever of unknown origin: the yield of a structured diagnostic protocol. Medicine 2007;86:26-38.
  8. BMJ 2010;341:c5470.
  9. Knockaert DC, Dujardin KS, Bobbaers HJ. Long-term follow-up of patients with undiagnosed fever of unknown origin. Arch Intern Med 1996;156:618-20.