Achieving and maintaining adequate levels of analgesia and sedation in critically ill patients is a fundamental part of ICU care. Understanding the clinical pharmacology of commonly used sedative agents (e.g., midazolam, lorazepam, and propofol) and opioids (e.g., fentanyl and morphine) enables clinicians to best dose these drugs to the desired clinical effect while minimizing the risk of excessive sedation and cardiopulmonary depression. This has significant safety and cost implications for patient care in the ICU. Simulations of plasma concentrations of these medications when administered to ICU patients provide useful insight into the clinical pharmacology of these agents. A number of points should be made with regards to the interpretation of these predicted plasma concentrations, however. First, it is important to remember that PK parameters for most of these agents, with the exception of midazolam and propofol, were derived from bolus or short-term infusions administered to healthy patients, and that the PK parameters for lorazepam, fentanyl, and morphine when administered as long-term infusions to critically ill patients may vary dramatically from these initial estimates. Specifically, their volumes of distribution and elimination half-lives may prove to be significantly larger and longer, respectively, when administered to patients in the ICU. This pharmacokinetic variability may result in even longer emergence times than predicted herein following discontinuation of continuous infusions of these agents. Until similar studies in ICU patients are performed for lorazepam, fentanyl, and morphine, the clinical pharmacology of these agents in ICU patients remains uncertain. Additionally, midazolam and morphine both have active metabolites that can accumulate in critically ill patients receiving long-term infusions. These metabolites add significantly to the sedative effects of the primary compound. Other drugs with sedative effects given concurrently with any of these agents (i.e., psychotropic agents, epidural opioids, etc.) may also contribute to the sedative effects of these drugs. These studies do not account for the development of tolerance (which can occur with both benzodiazepines and opioids) or changing kinetic profiles within an individual patient over time (i.e., due to changes in volume of distribution, protein binding, or clearance). Finally, there is a high degree of interpatient variability among critically ill patients, and medication dosing must be tailored to individual patients' needs (i.e., one dose does not fit all patients). Given the uncertainty of resulting plasma concentrations with long-term administration of these medications, the best ways to achieve and maintain optimal levels of sedation and analgesia while minimizing the risk of oversedation and side effects are to (1) initiate sedation in an incremental fashion until the desired level of sedation is achieved, then periodically (i.e., once a day) titrate the infusion rate of sedative-hypnotics and opioids downward until the patient begins to emerge from the sedative effects of these drugs; and finally gradually increase the infusion rate until the desired level of sedation is once again achieved; and (2) consider the use of a sedation scale to standardize the level of sedation to be maintained (see Table 3). The use of such a scale enables physicians to communicate to nursing staff the specific level of sedation to be achieved and maintained in an individual patient (i.e., titrate the midazolam infusion between 0 to 5 mg/hr to maintain a sedation score of 2-3; call MD for inadequate sedation, respiratory depression, or hypotension). Achieving optimal sedation and analgesia of patients in the ICU requires not only that the choice of medication(s) be appropriate for the clinical setting but also that there are specific clinical endpoints for the agents used (i.e., light versus deep sedation, continuous versus intermittent sedation, sedation with