[Errors in patients with infections].

Problems in the management of patients with infections are illustrated with three cases. The first case underlines the need for knowledge about antibiotic resistance pattern and mechanisms, as well as about clinical treatment studies for the correct management of endocarditis. The second case shows that results of microbiological samples may provide irrelevant data on which the treatment with suboptimal or wrong antimicrobial agents is based. The third case illustrates that the analysis of an infection should be a continuous process. Each microbiological result should initiate reevaluation, in order to optimize treatment. Criteria for adapting therapy are the in-vitro susceptibility of the infecting agent, the risk of emergence of resistance during treatment, results of published clinical studies, as well as toxicity and price of the antibiotic. All three cases illustrate the impact of an infectious disease specialist who should not only provide clinical knowledge, but also information about microbiology and epidemiology.