A surprisingly large and growing proportion of fevers have no identifiable source, as illustrated in this chart from a new NEJM Perspective on fevers of unknown origin (FUOs), by Harold Horowitz:
As the keeper of the antibiotics, should I be a conservative or a cowboy? Should the current antibiotics be continued, changed, or stopped? If there are no prescribed antibiotics, should I recommend some? These are interesting questions in the abstract, but there is a real patient suffering, a family with questions, and medical teams awaiting my opinion. There are no evidence-based studies and there is no guidance on which potential source of fever is the single appropriate one to treat. Frequently, the treatment approach is like playing Whac-A-Mole: positive cultures are treated sequentially — pneumonia, then catheter cultures, then urine cultures. When the fever persists, the cycle begins again.
I have it on good authority that this is what it is really like to practice medicine. Very few patients comprehend just how much uncertainty exists about many treatments and maladies, though not all of them.