TORONTO - When diagnosing recurrent fevers in children, one has to look at when the fevers occur. Do they occur at well-defined, fixed intervals vs. those that occur at irregular intervals?
"Fevers with a fixed interval have a fairly small differential diagnosis, so if you get a patient's history, you can really narrow the possibilities," said Chandy John, MD, assistant professor of pediatrics and internal medicine, Case Western Reserve University and Rainbow Babies and Children's Hospital, University Hospitals of Cleveland, at the 40th Interscience Conference on Antimicrobial Agents and Chemotherapy.
"When parents come to you, they often give you (the child's) history accurately. Fevers with no fixed interval have a much larger differential diagnosis, but again the focus during the physical exam and a few basic lab tests can give you an idea of the children for which you need to do a more extensive work-up."
Recurrent fever can be defined as two or more episodes, with an interval of at least seven days between episodes, and no clearly defined medical illness accompanying fever.
Periodic fever, aphthous ulcers, pharyngitis and cervical adenopathy are moderately common findings in children with recurrent fevers, according to Janet Gilsdorf, MD, professor of pediatrics and communicable diseases, University of Michigan in Ann Arbor.
With those symptoms, fever is regular, usually every 21 to 28 days, and parents can almost predict to the day when their child will present with the next fever. Children are well between febrile episodes, and fevers almost always present before 5 years of age and are non-familial.
Other illnesses that can be present with fixed interval recurrent fevers include cyclic neutropenia, relapsing fever and familial Mediterranean fever.
"Cyclic neutropenia on the other hand is rare," said Gilsdorf. "We have to include it in the differential diagnosis ... but in reality it's a rare entity. It is characterized by regular fevers, but also by neutropenia that is extremely regular, usually every 21 days, although the interval can vary. So fevers associated with infections related to neutropenia appear at similar, regular intervals."
According to Gilsdorf, those fevers are characterized by recurrent stomatitis, gingivitis, cutaneous infections, lymphadenopathy and neutropenia-related infections. They almost always present in children younger than 5 years and the condition may be familial, but does not have to be.
Gilsdorf said relapsing fever is rare, and comes in two forms: louse-borne and tick-borne, which is seen more commonly in the United States. This fever usually lasts for two to three days then abates, occurring every six to nine days on a fairly regular basis.
"These children have a high fever with systemic symptoms during the time they're ill," said Gilsdorf. "They have rigors, headache, arthralgia, lethargy, and they may have a petechial rash and hepatosplenomegaly as well."
Familial Mediterranean fever (FMF) and Hibernian fever are other rare entities with irregular fevers, although some children may present with regular fevers.
Despite an extensive work-up, there is often no definitive diagnosis for most children with fevers at irregular intervals, and the fevers resolve in most of these patients without serious sequelae, said John. The remaining patients with recurrent fevers most often have recurring viral infections causing their fevers. Only a small minority will have a medical condition other than recurrent viral infections diagnosed. In this small group, inflammatory bowel disease, specifically Crohn's disease, and systematic juvenile rheumatoid arthritis are among the more common causes of recurrent fever at irregular intervals.
"Other diagnoses are rare except in specific subsets of populations," said John. "Once you've excluded the more common causes, you should be aware of how some of the less common causes will present."
Recurrent fevers are more often seen in Crohn's disease than in ulcerative colitis in children, and symptoms are fairly non-specific. There is often a history of weight loss and malaise, and sometimes a history of diarrhea, loose stools or non-specific abdominal pain. Signs on exam include oral ulcers, perianal skin tags and uveitis, or no physical findings.
"Check stool," said John. "The majority of (children) will have heme-positive stool even if there's no gross blood streaking."
Patients also typically have anemia. With a two- to
three-month history, patients may also have moderately decreased serum protein
or albumen, John noted. "If you think you have a diagnosis, testing for these
Classic signs and symptoms of systemic juvenile rheumatoid arthritis are recurrent periods of high fever accompanied by a transient morbilliform rash. Fevers are classically hectic, greater than 40° C and decrease rapidly. Patients may have hepato-splenomegaly and diffuse lymphadenopathy. Lab tests show anemia, thrombocytosis and an elevated erythrocyte sedimentation rate.
"Systemic juvenile rheumatoid arthritis ... is much less likely to have a positive antinuclear antibody or rheumatoid factor, so getting a negative one does not in any way exclude this diagnosis."
John noted physicians should keep inflammatory bowel disease and juvenile rheumatoid arthritis high on the list of possibilities of disease associated with recurrent fever. A few simple laboratory tests will suffice to diagnose or rule out these serious illnesses. Other less common causes of recurrent fever at irregular intervals include hyper IgE syndrome, familial Mediterranean fever, and Behçet's disease. Malaria rarely presents with recurrent fevers as defined above, since the interval between fevers is almost always less than 72 hours.
Additional information may be necessary to diagnosis recurrent fevers, including family history, a child's present age, age at fever onset, past hospitalizations and past illnesses. "Then you need to really focus in on the fever and define it in the most careful way you can," said Gilsdorf. "How is it measured at home? How high does it get? How frequent is the fever? What is the duration? Has the child received antibiotics? If so, how did the fever respond? Did it go away and come back when the antibiotic was stopped?"
Gilsdorf also said the physician should ascertain if the child is ill or well between febrile episodes. Is the patient a well child with recurrent fevers or is he or she a chronically ill child with recurrent fevers?
"I'm a total believer in fever charts," she noted. "For every child we see with a story of recurrent fever, we send the parents home with instructions for how to measure the fever and how to record it and bring it back to us."
Looking for associated symptoms, medication history, travel history, especially to malarial endemic areas, and family history is vital. Ethnicity is also a critical part of evaluation, as well as evidence of chronic illness in family members.
On physical exam Gilsdorf looks for growth failure, something she calls an important piece of information about the general well being of the child. Healthy children with recurrent fever tend to grow well, but chronically ill children with recurrent fever and chronic illness do not grow well.
"Then you're looking for other physical signs: conjunctivitis, ulcers, adenopathy, heart murmur, pulmonary signs, abdominal signs, perianal skin tags, evidence of arthritis, mental status changes, and of course, rash."
In addition to a physical exam, laboratory evaluation should be guided by findings based on history and physical exam when developing a differential diagnosis.
For more information:
- John C, Gilsdorf J. Recurrent fevers in children. Session 184. Presented at the 40th Interscience Conference on Antimicrobial Agents and Chemotherapy. Sept. 17-20. Toronto.
- Long SS. Syndrome of periodic fever, aphthous stomatitis, pharyngitis, and adenitis (PFAPA) - What it isn't. What is it? J Pediatr. 1999;135:1-5.
- Thomas KT, Feder HM, Lawton AR, et al. Periodic fever syndrome in children. J Pediatr. 1999;135:15-21.
- Padeh S, Brezbiak N, Zemer D, et al. Periodic fever, aphthous stomatitis, pharyngitis, and adenopathy syndrome: Clinical characteristics and outcome. J Pediatr. 1999;135:98-101.
You can express your views on this article, or other relevant themes, in the Infectious Diseases in Children Specialty Forums.