We undertook a prospective study of children from three to 24 months of age with rectal temperatures of greater than or equal to 40.0 degrees C (104.0 degrees F) to determine if children whose fevers fail to respond to antipyretic therapy are more likely to be bacteremic than children whose fevers are lowered by antipyretic measures. Children from two clinical settings were studied: primarily black lower-class children at an inner-city hospital (n = 188) and primarily white middle-class children at a suburban hospital (n = 45). We found an overall prevalence of bacteremia of 7.3%, which was not statistically different between two hospitals. A response to antipyretic therapy, defined as a decrease in temperature of at least 1 degrees C, was seen in 83.7% of children. Children who did not respond to antipyretics had no more increased prevalence of bacteremia than did responders. We conclude that lack of fever response to antipyretics is not a clinical marker for bacteremia in children.