Petersdorf and Beeson's1 classic definition of a fever of unknown origin, or FUO, is a fever of at least 3-week duration, the cause of which remains unknown despite a reasonable workup, including 1 week in the hospital. However, 2 key factors have changed since the initial definition of FUO was set forth. The first factor is that medical technology has advanced a great deal since 1961. We now have computed tomography, magnetic resonance imaging, transesophageal echocardiograms, and the ability to perform a biopsy and aspirate many different places via the marvel of interventional radiology. Second, nowadays, there are many more patients with underlying immunocompromised states, including patients with cancer, human immunodeficiency virus (HIV), and transplant recipients.
Despite (or in some ways, because of) this changing landscape, FUOs remain one of our specialty's greatest challenges. Especially in my first few years, I never quite knew if I would make a diagnosis for these cases. As it turned out, my own experience with FUOs has run an interesting range, from miliary tuberculosis to lymphoma to Still disease, with a patient with Munchausen syndrome thrown in for good measure.
As infectious disease practitioners, we occasionally see true FUOs; however, more often, we see cases that nearly make the definition. For those cases that do not quite make the criteria of 3 weeks, perhaps we could use some alternative terminology. "Fever of elusive origin (FEO)" might fit the bill, although it would likely engender some confusion: some may shorten it to FEO, which may be mistaken for "pheo", short for pheochromocytoma, or perhaps more troublingly be misunderstood as "feo", Spanish for "ugly". I have had patients with FEOs from babesiosis, drug fever, anaerobic paraspinal abscess, abdominal tuberculosis, colon cancer, and as presenting signs of systemic lupus erythematosus, Behçet syndrome, Wegener granulomatosis, and Goodpasture syndrome. For that matter, I have even had a pheo causing a FEO.
Getting to the bottom of FUOs and FEOs requires good old fashioned detective work via compulsiveness, thorough history taking, logical methodology, reviewing the literature in the wee hours of the morning, sacrifice of sleep and sanity, and occasionally, good luck: for the human body is predictable only in its unpredictability. Hot on the trail of the fever, we follow leads, constantly trying to decide what path to reasonably take and which avenues take us down blind alleys and dead ends. We go over the same ground again and again, assembling our evidence, hoping that a consistent pattern emerges to support a single disease entity. Sometimes, it takes some time for this pattern to evolve. And, in the end, we either nail down a fascinating diagnosis, or less commonly, come away with a vague unsatisfied feeling, if we have not gotten all the answers. Sometimes, as in the case of Mrs T's mysterious fever, we are left with a little of both.
Mrs T was a lady in her late 70s, a sharp, proud, and independent woman. She liked to think she knew all the answers; in retrospect, she probably did. She was referred by her primary care physician to a pulmonologist for feelings of weakness and shortness of breath and was immediately admitted to a large university hospital. Her other complaints besides the dyspnea were severe fatigue, night sweats, and weight loss of some twenty pounds for 2 months. Her medical history was remarkable for monoclonal gammopathy of unknown significance. It was noted on her routine blood work that she had a normocytic anemia, with a hemoglobin level of 13.4 g/dL, mild renal insufficiency with a serum urea nitrogen of 25.0 mg/dL and a creatinine level of 1.3 mg/dL, normal findings on liver function tests, and an extremely high erythrocyte sedimentation rate. Her oxygen saturation level dipped to as low as 76% at night. In the hospital, she began spiking everyday to 102, unrelentingly. The pulmonary group then called in my group, along with a hematology group. Each group has 5 to 6 physicians. Although not all the physicians in the various groups rounded on her, certainly there were more than 10 physicians seeing her, all eagerly giving their input. Then, the million-dollar workup began.
Routine blood and urine cultures were negative, of course. An extensive pulmonary workup was undertaken, with essentially negative results. The patient was anergic. A bone marrow biopsy was consistent with monoclonal gammopathy of unknown significance, with no real difference from a prior biopsy. A computed tomography of the abdomen was done, which showed an enlarged spleen. My partner reminded us of the old story of a New England Journal of Medicine case, a patient with splenomegaly. A medical student managed to get the right diagnosis of splenic lymphoma; when he was asked how he had arrived so knowingly and convincingly at the answer, he had asked innocently, "What else causes splenomegaly?"
However, it was not to be that easy, of course. Everyday, we visited the patient and went over a careful but directed review of systems with her, along with a detailed physical examination. Her answers were rambling, imprecise, and occasionally tangential. Initially, I labeled her in my mind as another "poor historian", but over time, I believe she was actually trying to change the subject. But as she had more and more tests and became more ill for the next few days, she began to withdraw and often answered our questions in a flat monotone, just ignoring us altogether after awhile. This proved to be more than a little frustrating because somehow I felt that the answer lay somewhere in her history.
On Wednesday, nearly a week after her admission, the patient was lying in her bed, apathetic, and barely communicative with me, answering me in monosyllables. I spoke to her daughter, who was at her bedside, as I had done before; however, again, I asked if there were any significant past medical problems. For some reason, after her mother had been in the hospital for days and days, the daughter's mind was finally jogged; she related that her mother had been diagnosed with some immune disorder, where her CD4 cells were low, and she had been worked up at the Mayo Clinic for this. Apparently, she was HIV negative, but from the sounds of it, she may have had idiopathic CD4 lymphopenia. Here again was yet a new path to explore because it seemed that she had a T-cell deficiency of some sort, and her ensuing illness might represent either an opportunistic infection or some unusual tumor. I hurriedly ordered a CD4 count and other serologies, including an HIV test. The CD4 count was mildly depressed, and the HIV test was negative.
Everyday, as I bumped into some members of the pulmonary group in the hallway, we reviewed the entire case and ran over a differential: none seemed to add up completely, save perhaps one. When asked about his differential, the hematologist chuckled and answered, "Have you heard of the New England Journal case of splenomegaly, where the student…."
"Yes, yes, yes," I said wearily, but this was of limited use. No one in our institution felt that a needle biopsy of the spleen could be safely done; although it was mentioned numerous times, no one wanted to take out Mrs T's spleen unless there was absolutely no other way to make her diagnosis. Yet meanwhile, she continued to dwindle, spiking high nearly everyday, refusing to leave her bed, shrinking down to skin and bone, and talking less and less.
The patient had been in the hospital for nearly 2 weeks. Saturday on call, I noticed for the first time that she seemed slightly jaundiced. Someone had ordered liver function tests for that morning, and I was the first to check them. I noticed that her bilirubin level and transaminases, which had heretofore been normal, had risen dramatically over the course of a couple of days. Was this viral or autoimmune or, more likely, malignant in origin? I sent off various serologies and also an abdominal ultrasound and waited with bated breath. And there finally was the break in the case. Although a computed tomographic scan had not shown this a week and a half before, the ultrasound now showed multiple large hepatic nodules, along with splenic nodules that seemed most consistent with metastases. Wonderful, something easier than a biopsy of a spleen! I eagerly sought a gastroenterologist (who, with his partner, now made 17 physicians aware of the case). However, for the next few days, as the patient became weaker, she started to refuse all the tests. She claimed she was not strong enough to undergo a liver biopsy. This was vexing, as for the first time in 2 weeks, it seemed as if we were close to a diagnosis. For the next few days, the gastroenterologist held off on the liver biopsy, believing her to be too weak. Meanwhile, she started to become as yellow as a canary. I wanted a diagnosis. I picked up the phone and talked to the gastroenterologist, arguing that we had a small window of opportunity before she really became ill.
The liver biopsy was finally done. Seventeen days after her admission, a diagnosis of lymphoma was made, just as predicted by that famous medical student. For a moment, there was some relief that we had actually made a diagnosis but tempered by a sobering thought: the lymphoma was obviously very rapid and progressive. Several days later, Mrs T was intubated; her small form looking very frail was hooked up to the ventilator. The following weekend, Mrs T died.
We had done a thorough and professional job, generating a reasonable differential examination and working her up in a logical manner. As the consultants, we contributed to clinching the final diagnosis, which should have been satisfying to some degree.
Yet, cases are not merely intellectual exercises. As the patient's team of physicians was going through differentials and plotting the next course of action, she came to understand much sooner than we did that she was dying, and she had to take the little time she had left to prepare herself for this. Once the diagnosis was made, most of the 17 or so subspecialists on her case signed off, leaving her alone in her final hours.
Perhaps we concentrated too hard on finding her disease. She probably would have preferred us to find out what was truly in her heart.
The author thanks the Pulmonary and Intensive Care Specialists of New Jersey, consisting of Drs David Fischler, Tricia Gilbert, Andrea Harangozo, Douglas Hutt, Donna Klitzman, and Philip Schiffman, for allowing the author to write this case study about their patient.
© 2007 Lippincott Williams & Wilkins, Inc.