Severe sepsis is a major burden in the intensive care unit (ICU) with persistently high mortality rates. Optimization of antibiotic dosing has been suggested as an intervention to improve clinical outcomes for critically ill patients with severe sepsis. However, current antibiotic dosing guidelines may not be appropriate for these patients, as they rarely consider the altered physiology and illness severity associated with this population. Optimizing antibiotic dosing using pharmacokinetic (PK) and pharmacodynamic (PD) principles can address these critical illness-related changes and promote therapeutic success. Due to their wide spectrum of antibiotic activity and excellent safety profile, beta-lactam antibiotics are commonly used for severe infections in the ICU. Two alternative dosing approaches to traditional intermittent bolus (IB) dosing, namely continuous infusion (CI) and extended infusion (EI), have been suggested to maximize the therapeutic potential of these antibiotics in critically ill patients. Collectively, the two dosing approaches can also be referred as prolonged infusion (PI).This Thesis aims to better characterize the pharmacokinetics/pharmacodynamics (PK/PD) of betalactam antibiotics to determine whether there is any therapeutic advantage associated with PI dosing (CI and/or EI) as compared to IB dosing. This Thesis comprises of eight chapters. Chapter 1 is an introductory chapter which provides an overview of the published literature on the area of research. The discussion in Chapter 1 presents a theoretical framework behind the Thesis objectives. Chapter 1 concludes with the specific aims of this Thesis.Chapter 2 reports the findings of a prospective PK study which aimed to describe the population PK of doripenem in critically ill patients with sepsis and perform dosing simulations to develop clinically relevant dosing guidelines for these patients. Twelve critically ill participants receiving 500 mg of doripenem 8-hourly as a 1-hr infusion were enrolled. The volume of distribution (Vd) and clearance (CL) of doripenem in this patient cohort were substantially different than those usually described in non-critically patients. As current dosing guidelines were mostly derived from the non-critically ill, findings from this study suggest that the licensed "one-dose-fits-all" dosing for doripenem is unlikely to achieve optimal exposures in critically ill patients. Empirical use of PI dosing should be considered to account for PK and illness severity differences, particularly when less-susceptible pathogens are involved.ii Chapter 3 incorporates a published systematic review which compares the PK/PD data and clinical outcomes between CI and IB dosing to describe any potential merits supporting either of the two dosing approaches for critically ill patients. The findings suggest that beta-lactam CI may not be advantageous for all critically ill patients and may be beneficial in patients with severe infections.Chapter 4 describes the findings of a post hoc analysis on the Defining...