Improving antimicrobial use in critically ill patients

Although delivery of prompt and appropriate antimicrobial therapy is of paramountnimportance for critically ill patients, few data exist to guide what doses of antimicrobialsnshould be used in critically ill patients. Presently available generic guidelines are derivednfrom healthy young adults and are unlikely to be appropriate for critically ill patients due tonthe major physiological changes that occur. Altered physiology can result in significantnchanges in antimicrobial pharmacokinetics and greatly affect dosing requirements. Therenis limited guidance to determine therapeutic antimicrobial doses due to only limitednpharmacokinetic data being available in these patients. Additionally, there will be greatnvariance between patients because of differences in co-morbidities, the invasiveness ofnprocedures and the disease status. Thus, it is very unlikely that a dose found to beneffective in non-critically ill patients will be optimal for the majority of critically ill patients.nSince available data do not appear to suitably guide dosing in these patientnsubpopulations, directed pharmacokinetic studies should be considered fundamental.nTherefore, the principal aims of this thesis were to evaluate the appropriateness ofnstandard doses in achieving pharmacokinetic/pharmacodynamic targets and clinicalnscenarios affecting this target in critically ill patients. Dynamic renal function is commonly seen in critically ill patients. For drugs cleared throughnthe renal route; while decreasing renal function trigger reduced dosing, elevated renalnfunction or augmented renal clearance should also trigger dosing increases. This is tonavoid sub-therapeutic concentrations. However, due to scarce data available, thisnphenomenon has not been clearly described. This led to a study in this thesis, conductednin Malaysian intensive care units to describe this clinical scenario. It was found that almostnhalf of the subjects recruited have augmented renal clearance. Significant bias andnimprecision was demonstrated when comparing estimated Cockcroft-Gault creatininenclearance and measured urinary creatinine clearance with the bias being larger innaugmented renal clearance patients and significant difference were found between thisntwo methods. This study supports previous data that equation-based estimates ofncreatinine clearance are unreliable for use in critically ill patients.n