ABSTRACT
Fever of unknown origin (FUO) amply illustrates the mot that “the three most important principles of medicine are diagnosis, diagnosis, and diagnosis.” Empiric therapy plays a limited role in FUO, if by FUO we mean prolonged FUO (3 weeks) (1). Cunha in 1996 recommended empiric therapy for only four situations: antibiotics for culture-negative endocarditis, low-dose corticosteroids for presumed temporal arteritis, antituberculous drugs for suspected military tuberculosis in elderly patients, and naproxen (Naprosyn) for suspected neoplastic fever (2). Our purpose is to review some general principles of empiric therapy and a handful of specific indications. Elsewhere in this volume, various authors discuss empiric therapy for specific subsets of patients with FUO.
