Abstract
The course of a patient's elevation in body temperature is not generally thought of as a diagnostic test. Yet it has a number of characteristics that are highly desirable in such a test: It puts the patient to no added expense, it holds no risk and little inconvenience, and the results are immediately available at the bedside. Assessments of its sensitivity and predictive potential are difficult to make. A given pattern is surely not highly specific. Even so, certain fever or fever-pulse patterns in some settings can, at a minimum, suggest a particular microbial agent. Because the first step in determining the optimum strategy for managing an infectious disease is to determine the most likely infecting microorganism, tests that yield such information are especially valuable. In this article we shall describe several types of fever or fever-pulse patterns that can be clinically significant. We shall begin with the two that we ourselves have found most useful and then mention some found useful by others.