Although septic shock may be the most common cause of distributive shock, to our knowledge, no studies have defined the likelihood and type of infection among patients with distributive shock. We performed a retrospective study of 100 consecutive patients who were admitted to a city-county hospital with hemodynamic evidence of distributive shock. Forty-nine of 100 patients with distributive shock had microbiological documentation of infection. Six patients had clinical evidence of infection without microbiological documentation. Forty-five patients had no microbiological or clinical evidence of infection. Among patients with microbiologically documented infections, the incidence of infection due to aerobic gram-positive cocci equaled the incidence of infection due to aerobic gram-negative bacilli. Clinical parameters, such as the criteria for the systemic inflammatory response syndrome, were not useful in distinguishing the group with infections from the group without infections. In conclusion, many patients with distributive shock do not have evidence of infection.