Dr Jon Iredell - University Of Sydney - Westmead Hospital

More than half of nosocomial infections are due to antibiotic-resistant bacteria, and are clearly associated with excess mortality and health care costs (equivalent to approximately 2-7 billion Australian dollars, in our population), in turn largely due to increased length of stay and antibiotic usage. Delays in appropriate treatment with antibiotics may have serious consequences, including up to thrice the attributible mortality from pneumonia and bacteraemic infection. However, empiric broad spectrum antibiotics must remain the mainstay of therapy in the critically ill patient with suspected infection, since antibiotic prophylaxis is widespread and only a minority of prescriptions are informed by a timely and relevant microbiological diagnosis. The spiral of increased resistance and broader-spectrum antibiotic usage thus continues. Antibiotic usage is well documented to be associated with emergence of specific resistance as a direct result of antibiotic selection pressure. Certainly, the usage of multiple classes of antibiotics is a strong predictor of colonisation with extremely resistant Acinetobacter baumannii in an outbreak at our hospital. Consequently, most antibiotic resistance in hospitals appears to arise in the ICU, the very place in which its consequences are most dire. However, reduced usage of an antibiotic in the ICU setting has been shown to be associated with reduced resistance to that antibiotic in subsequently cultured isolates over a period of months, and this has led to attempts to 'cycle' antibiotics in an effort to reduce the effect of strong selection for resistance to a given class of antibiotic. However, while 'fitness' of a resistant mutant is often compromised, this varies with the mechanism involved and may be resolved by secondary mutations over time. Thus, maintained selection pressure may promote fitness in mutants and when resistance comes at no biological cost, there is no selection pressure against it. Drug inactivating enzymes encoded by single genes are the most transmissible and arguably the most important cause of resistance, and specialised gene acquisition and management systems are found in the great majority of resistant nosocomial bacteria. Much of the resistance which is amplified under selection pressure may come in with the patient as part of the endogenous microflora, but the extent to which this process contributes to the general pool of resistant microflora in an ICU environment is unknown. Nevertheless, it is clear that mobile elements are in constant flux and that this reflects selection pressure on that population of resistance genes. In this discussion, we deal with distinctions between 'breakthrough' infection, presenting infection (community flora), presenting infection (hospital flora), and the overall burden of resistance. We will discuss the genetic basis for silent gene transfer and management and place resistance phenotypes in the broader biological context. An alliance of microbiologists, molecular geneticists, infection diseases specialists, and intensivists are uniquely positioned to make new inroads into Intensive Care infection.

Time of Presentation:

Sunday 4 May 2003 - 1500-1530