Kollef et al1 published an early study on this topic in 1998. They reported a retrospective analysis of continuous IV sedation (93 patients) compared with no continuous IV sedation (149 patients). In this analysis, patients in the continuous group had longer ventilator times, longer lengths of stay (ICU and hospital), more organ system derangements, and more reintubations. At first glance, this study appears to support an intermittent approach. However, it is important to realize that this study reported IV sedation (93 patients) compared with no continuous IV sedation. In the no continuous IV sedation group of 149 patients, 85 received no IV sedation at all, whereas only 64 of the total 149 received intermittent sedation. Accordingly, it is not valid to conclude from this study that intermittent bolus sedation was better. Furthermore, 72% of the continuous infusion group patients received lorazepam infusions, with only two patients receiving propofol. As discussed in the “Final Comments” section, this heavy reliance on benzodiazepines is no longer a recommended strategy. Finfer et al2 reported a study of 40 patients randomized to intermittent diazepam vs continuous midazolam, with 31 patients completing the study (midazolam continuous infusion group, n=17). There were no statistical differences between the groups regarding hours to target sedation, hours within target sedation, or hours oversedated; the diazepam intermittent group had a significantly greater number of hours undersedated. Carson et al3 published a prospective randomized trial investigating intermittent lorazepam (based on the 2002 Society of Critical Care Medicine4 sedation and analgesia guideline consensus statement) vs propofol with a daily sedative interruption strategy. The continuous infusion propofol group had fewer days on a ventilator. The survivors in this group also had a reduced ICU length of stay. Mehta et al5 compared patients randomized to continuous midazolam infusions with daily interruption with intermittent midazolam based upon a nursing protocol and published a preliminary pilot trial. Despite being in the nursing protocol intermittent dosing group, the majority of these patients (93.8%) actually received midazolam infusions for at least a portion of their ICU stay. The investigators noted no differences in ventilator days, length of stay, or sedation target level achievement. The crossover seen in this 65-patient pilot study may confound the ultimate analysis of the data. The final results of this study were presented recently in abstract form.6 There was no difference in ventilator days or length of stay between the two groups, although frequency of adherence to the particular randomization scheme remains an area of concern. At the time of this writing, the study has not yet been published in a peer-reviewed format.