Abstract

PIP: 2 topics of particular interest are addressed in this summary of data concerning antibiotics and endotoxic shock: 1) the antibiotic treatment of gram-negative bacillary infections complicated by shock differs from the treatment of similar infections not associated with shock; and 2) endotoxin per se is responsible for the occurrence of shock and other manifestation of gram-negative bacillary infections. With shock, inadequate tissue perfusion exists, requiring antibiotic administration be performed intravenously rather than by oral or intramuscular routes. In general, antibiotics act only to eradicate infection (i.e., kill bacteria) and in no way address the shock condition. Clinical results with polymyxins, implicated in some studies to neutralize endotoxin, assessed these drugs' clinical applicability by comparing the frequency of manifestations attributed to endotoxin, shock, and death occurring in patients with gram-negative bacteremia after treatment with polymyxin B or E. Treatment with 1 of the polymyxins failed to reduce the frequency of shock and death in gram-gegative bacteremia over that observed when other effective antibiotics were used initial therapy. Indeed, the frequency of complications was significantly greater in patients with ultimately fatal underlying diseases (P .05) and in all patients combined (P .001). Human studies have attempted to define or pinpoint the relationship between endotoxin and disease symptoms (chills, fever, etc.), assuming an apparent correlation between the presence of circulating endotoxin and the frequency of fever, shock, or death compared with similar patients without detectable endotoxin; though the results did not preclude a role of endotoxin in symptomatology, they do cast considerable doubt on the widely held concept that circulating endotoxin is primarily responsible for the pathophysiologic changes observed during gram-negative infection. Using Limulus test data, the majority of studies found no connection between circulating endotoxin and presence of pathophysiologic changes. Factors affecting treatment of bacteremia are also discussed.