A Fever That Was Frightening
Toddler Fell Ill Every Three Weeks

By Sandra G. Boodman
Special to the Washington Post
Tuesday, June 24, 2008

Nighttime, Greg Licameli remembers, was always the worst.

Those were the hours when whatever was wrong with his baby daughter Claire seemed most terrifying, when alternating doses of Motrin and Tylenol did nothing to quell the fever that would spike as high as 106 degrees and when the surgeon and his wife, a nurse, would tersely confer about whether to pack their toddler in ice to try to stave off another seizure that would require a sprint to the emergency room.

For months, these nights occurred every three weeks or so like clockwork, preceded by 24 hours of dread, waiting for the onset of the mysterious fever that lasted four or five days. The prelude was always the same: A day before the fever, Claire would turn lethargic, and the small lymph nodes in her neck would become enlarged. Once when an episode was imminent, the family arrived at the Atlanta airport to discover that their flight home to Boston had been canceled.

"It was definitely frightening," Licameli said of the episodes that baffled at least half a dozen eminent pediatric specialists. "This was certainly outside of my scope of training."

Licameli did not know it at the time, but it was his training as an ear, nose and throat surgeon that ultimately helped him figure out how to help his daughter, now 6, and other children whose little-known malady has only recently been identified. Earlier this year Licameli, an attending physician at Boston's Children's Hospital, and his colleagues published a study in the Archives of Otolaryngology detailing the results of the treatment Claire and 26 other children underwent that cured all but one of them.

"I think it's very ironic that the cure is something I do frequently," Licameli said.

Before the mysterious cyclical fevers began, Claire, a healthy baby and the second of the couple's two children, had experienced the usual illnesses. When she suffered the initial episode, shortly after her first birthday, her pediatrician said a bad virus had probably caused her temperature to escalate from 101 to 105 degrees in less than an hour and had made her throat so sore it caused her to drool. After five long days, Claire recovered.

But every three weeks, it seemed, brought a repeat of the first episode. It was clear this was no ordinary virus -- or maybe not a virus at all, said Licameli, an assistant professor of otolaryngology at Harvard Medical School. Doctors prescribed antibiotics, thinking the problem might be bacterial, but they had no effect.

"She saw everyone I knew," her father recalled, in an attempt to figure out what was wrong. After Claire's fever got so high she had a seizure and was rushed to the ER, doctors performed an MRI to rule out a brain tumor or other lesion. They found nothing.

An infectious disease specialist thought the problem might be cyclic neutropenia, a rare immune disorder in which the white blood cell count drops dramatically every 21 days or so; testing ruled that out.

A pediatric rheumatologist suspected juvenile rheumatoid arthritis, another dead end. Licameli consulted experts at the National Institutes of Health to see whether his daughter might have a genetic fever syndrome. She didn't.

"One of the things that struck us was the periodicity of the fever," Licameli said. The only thing that seemed to reduce it was a dose of prednisone, which eliminated the fever in a number of hours, for reasons that remain unclear. But the corticosteroid, which reduces inflammation, had another, less desirable effect: It caused the fever to return in 11 days, rather than the usual 22.

No one seemed to know what was wrong.

One night, Licameli recalled, he sat down at his home computer and launched a thorough search to see what he could find. He unearthed 15 articles written more than a dozen years ago describing a syndrome called PFAPA, an acronym for its hallmark symptoms: periodic fever, aphthous (mouth) ulcers, pharyngitis (sore throat) and adenitis, or enlarged and often infected lymph nodes. The condition was also known as Marshall syndrome, named after the doctor who first described it in 1987.

Licameli said he was immediately struck by how similar the descriptions were to whatever was ailing Claire: The children in the articles were younger than 5, had abruptly developed high fevers that lasted four or five days every four to six weeks or so, along with a sore throat and enlarged lymph nodes in their necks. Some developed small ulcers in their mouths as well. Over-the-counter pain relievers had been useless in reducing the fever, and the children seemed normal between episodes and had no underlying illnesses that would account for the fevers.

One Italian study involved six children, four of whom had been cured of the recurrent fevers after their tonsils and in some cases adenoids were removed -- although the tissue showed no signs of infection.

"That's my bread and butter," said Licameli of tonsillectomy, one of the most common surgical procedures performed on children. Intrigued, he proposed performing the operation on selected patients who had been referred to Children's for recurrent fevers.

The hospital's institutional review board, a panel that monitors the safety of research studies, agreed, and Licamelis decided that Claire should be among the study's patients.

"I felt a little desperate at that point, and I had no idea whether she would be cured," said her father, whose colleague operated on Claire.

At 19 months, she was the youngest patient in the study, which involved 27 children. In 26 cases, including Claire's, the fevers stopped and have not recurred.

Although surgery appears to be a cure for many, no one knows why. The prevailing theory, according to Licameli, is that children who develop recurrent fevers may have a hypersensitive immune system and that the fevers represent an exaggerated response to an immunological trigger, possibly a chronic and as yet undetectable infection of the tonsils.

Licameli has treated 60 children with the disorder, some from as far away as California, nearly all of whom have been cured. He said he thinks the disorder may be more common than believed, because many physicians may see only one or two cases in their careers. In the past few years, teams at other hospitals across the country have reported success treating PFAPA with tonsillectomy.

"I think the syndrome is pretty well accepted," Licameli said, "but the treatment is not yet mainstream because people don't know about it."

"I used to think, 'Oh, geez, why my kid?' " he recalled. "Now I think things happen for a reason."

If you have a medical mystery that has been solved, e-mailmedicalmysteries@washpost.com. To read previous mysteries, visithttp://www.washingtonpost.com/health.