Polymyxin B (PMB) belongs to a class of antibiotics discovered more than six decades ago. PMB was used for various bacterial infection threatening, in particular to sepsis. Its use, however, was abandoned because of the observation of severe side effects. In the last years this view changed due to the appearance of multi-drug resistant Gram-negative pathogens, which were resistant to most available antibiotics, leading to a re-evaluation of the polymyxin antibiotics (PMB and PME). Although there is a large market potential for the development of drugs to fight sepsis, the available successful clinical strategies are very limited. The cause for this lies in the clinical failures of a number of drug candidates, which were tested in the last years. This was attributed to some extent to our elementary understanding of the pathophysiology of sepsis, to not optimally designed clinical trials and a lack of appropriate pre-clinical models to establish the proof of concept (POC). At that time there were just humble knowledge about the structural mechanisms involved in the advantageous aspects of PMB-endotoxin interactions to increase the knowledges outcome in sepsis therapy. Therefore, the current paper describes the clinical aspects of PMB application in bacteraemia and sepsis therapy. However, the focus of the presented paper lies in the structural and mechanistic aspects of PMB-endotoxin (LPS: lipopolysaccharide) recognition and how this knowledge can be applied for the development or improvement of new clinical drug candidates to support sepsis therapies. Due to chemical similarities between PME and PMB, certain aspects of the use of PME as an antimicrobial agent and in sepsis therapy are considered and compared to PMB.