How to treat periodic childhood fevers

October 6, 2007

"We all see childhood fever, an awful lot of fever," said Cormac O'Connor, MD, family physician at Naval Hospital Yokosuka, Japan. "And most of them are benign viral syndromes that we don't have to worry about once they leave the office."

"We all see childhood fever, an awful lot of fever," said Cormac O'Connor, MD, familyphysician at Naval Hospital Yokosuka, Japan. "And most of them are benign viral syndromes that wedon't have to worry about once they leave the office."

However, the fevers falling outside the "benign" category worry family physicians most. Somerecurring fevers are benign; some are not. Some can be treated, some cannot, and some should bereferred for specialist care.

"We're really looking at what distinguishes these period fevers," Dr O'Connor said during theAmerican Academy of Family Physicians 2007 Scientific Assembly on Saturday in Chicago. "What welearned in medical school really isn't much help here."

The first step is basic definitions, he added. Periodic fevers return on a regular schedulewith the same duration. Recurrent, or episodic fevers, are irregular and unpredictable febrileperiods.

Periodic fevers appear more dangerous than they actually are, he noted. Fevers that appear ona regular basis, last for a similar period, and are separated by episodes of good health are rarelyserious. Periodic fevers are occasionally caused by recurrent infection or neoplastic disorders, butthe longer the fever-healthy episode-fever pattern continues, the less likely the child is to have aserious condition.

Periodic fevers that continue for two years or longer are highly unlikely to be serious, DrO'Connor said.

"The real patients in these cases are the parents. It more about educating them that they donot need to be overly concerned about treating the fever. "

Take cyclic neutropenia, the recurring childhood fever most often mentioned in medicalschool. Febrile episodes usually begin in infancy, lasting three to five days and recur on a regularcycle about every 21 days.

"The fever is the chief complaint, not the occasional opportunistic infection that might popup during febrile episodes," Dr O'Connor noted.

The key to diagnosing cyclic neutropenia is a blood count that shows ANC

Cyclic neutropenia has an incidence of 1:1 million, which means most family physicians willnever see a case. But they are very like to see PFAPA, a look-alike condition that occurs inapproximately one in every few hundred to one in 2000 young children. The acronym stands for thefollowing signs and symptoms: Periodic, Fever, Aphthous ulcer, Pharyngitis, and Adenopathy.

"PFAPA is hands down the most common cause of periodic fevers in children," Dr O'Connor said."Fevers are brief, but kids fall like they were hit by lightening."

A typical PFAPA episode erupts in less than an hour, he said. The child becomes pale,lethargic, and feverish and may show signs of pharyngitis and adenopathy.The aphthous ulcers are shallow and not particularly painful. They tend to occur on the inside of thelips and are seldom seen unless the physician peels back the lip and looks."This is not something where the child is hurting and doesn't want to eat or drink," he said. "If youdon't look yourself, chances are pretty good that neither parent nor child will even know thoseulcers are there."

PFAPA fevers occur every four weeks and last only a few days. Between episodes, the child ishealthier than siblings or age cohorts. The key clinical difference between cyclic neutropenia andPFAPA is the absence of neutropenia in PFAPA. There is no systemic involvement of the skin,respiration, GI, or joints.

The cause of PFAPA remains unknown, Dr O'Connor said. Oral corticosteroids, such as prednisone, abortthe febrile episode, but observational reports suggest that steroids may also shorten the intervaluntil the next episode. Cimetidine can reduce the severity of episodes and a single retrospectivestudy claims that tonsillectomy cures the condition.

Because PFAPA usually resolves by age 8, Dr O'Connor suggested that no treatment is best unless thereare outside reasons, such as school attendance, to consider. Ibuprofen and acetaminophen can be usedfor symptomatic relief.

If the feverish child is of Mediterranean origin, Familiar Mediterranean Fever (FMF) may be theculprit. The disease is an autosomal recessive genetic malformation that is most prevalent amongArabs, Armenians, Sephardic Jews, and Turks, although it has been reported from all countriessurrounding the Mediterranean.

FMF produces irregular fevers lasting six to 96 hours. Patients usually present with acute abdominalpain and joint involvement. Most patients are between the ages of five and 20. Colchicine preventsfuture attacks in approximately 60% of patients. The long-term prognosis hinges on amyloidosis. If itis absent, the patient has a normal life expectancy. If amyloidosis is present, life expectancy isshortened.

Children who present with fever lasting longer than 30 days and joint involvement should be referredto a pediatric rheumatologist, Dr O'Connor said. The potential diagnosis is systemic juvenilerheumatoid arthritis.

Other rare but potential causes of episodic fevers include bacterial infection with Plasmodium,Borrelia, or Brucella or latent viral infections with HSV1, HSV2, or CMV.

"History is 85% of your diagnosis," Dr O'Connor said. "If the primary complaint is the fever, you'venarrowed it down to a couple of things. After two or three febrile episodes, it's time to startsuspecting things like PFAPA."