Can We Ventilate Patients Without Sedation?

Abstract & Commentary

By Andrew M. Luks, MD, Pulmonary and Critical Care Medicine, University of Washington, Seattle, is Associate Editor for Critical Care Alert.

Dr. Luks reports no financial relationship to this field of study.

Synopsis: This single-center randomized trial demonstrated that when compared with standard sedation practices and daily sedation vacations, a protocol of no sedation was associated with a shorter duration of mechanical ventilation in critically ill patients.

Source: Strom T, et al. A protocol of no sedation for critically ill patients receiving mechanical ventilation: A randomised trial. Lancet 2010;375:475-480.

While intravenous infusion of sedative medications and daily sedation vacations have become standard practice in most ICUs, an institution in Denmark has actually been ventilating patients without intravenous sedatives for several years. Strom and colleagues sought to determine if this strategy was associated with shorter duration of mechanical ventilation than the more standard approach of sedative infusions and daily interruption of sedation.

The authors conducted a single-center, randomized, non-blinded study of critically ill patients who were expected to need mechanical ventilation > 24 hours. Patients were excluded if they had elevated intracranial pressure, had medical indications for sedative infusions (e.g., status epilepticus), were pregnant, or met criteria for ventilator weaning. Patients in the intervention group received bolus sedation with morphine (2.5 or 5 mg) on an as-needed basis. Physical restraints were not used and patients suspected of having delirium were treated with haloperidol. Patients who were still uncomfortable despite haloperidol received propofol for 6 hours after which time a trial off sedation was restarted. In addition to bolus morphine administration, patients in the control group received a propofol infusion before being transitioned to midazolam after 48 hours and received daily interruptions of their sedative infusion. All patients were ventilated using pressure support and once ventilator settings for patients in the control group reached an inspired oxygen fraction of 0.4 and positive end-expiratory pressure of 5 cm H2O, sedation was discontinued unless their ventilator requirements worsened. The primary outcome measure was the number of days without mechanical ventilation in the 28 days following intubation. Secondary outcomes included total length of stay in the ICU, ICU and hospital mortality, need for brain imaging, unplanned extubation, and ventilator-associated pneumonia.

A total of 140 patients were enrolled in the study (67% male; 33% female) but 27 were excluded from the analysis because mechanical ventilation was stopped within 48 hours. Ten of the 55 patients (18%) in the no-sedation group required continuous sedation with propofol on more than two occasions, usually due to oxygenation problems in the setting of severe acute respiratory distress syndrome (ARDS). Patients in the intervention group spent 4.2 fewer days (95% confidence interval, 0.3-8.1) on the ventilator than patients in the control group and had a shorter length of ICU stay (13.1 vs 22.8 days) and hospital stay (34 vs 58 days). Morphine use was not different between the two groups (0.0048 mg/kg/hr of mechanical ventilation in the intervention group vs 0.0045 mg/kg/hr in the control group). There were no statistically significant differences in ICU or in-hospital mortality, need for brain imaging, unplanned extubation, ventilator-associated pneumonia, or a need for reintubation within 24 hours. Delirium was more common in the intervention group (20% vs 7%) and haloperidol was used more frequently in that group (19 vs 8 patients).

Commentary

Strom and colleagues demonstrate that it may be possible to move beyond the benefits of daily interruption of sedative medications and achieve further gains with a no-sedative strategy for many but not all, critically ill patients. In fairness, this was not truly a "no-sedation" strategy as all patients received bolus doses of morphine, a medication with sedative effects, but they did manage most of the patients without the need for propofol or midazolam. Of particular note was the fact that they were able to do so without seeing an increase in complication rates such as unplanned extubation or ventilator-associated pneumonia.

Before we rush to turn off the propofol and midazolam drips on all of our patients, however, there are several key issues that must be addressed. The first, and most obvious, issue is that this was a single-center study with a limited number of patients. More importantly, the protocol was implemented at an institution that already had almost 8 years of experience with the no-sedation strategy at the time the trial was conducted. As a result, both physician and nursing staffs were likely comfortable and facile with the approach, a situation that might not exist at other institutions where the use of sedation is strongly ingrained in ICU practice. For this reason, we will clearly need larger, multicenter trials to establish that this approach can be adopted safely at institutions with a tradition of different sedation practices. It is not difficult to imagine the culture shift that such a strategy will require and more data will be necessary to encourage such a shift.

Other aspects of the protocol also warrant attention and may limit attempts at implementation. The ICU in this study maintained a 1:1 patient-nurse ratio for all study participants and the investigators also used extra people to help calm the patients when they developed agitation. These are resource-intensive measures that may not be feasible at many institutions and it remains to be seen whether the cost savings associated with shorter duration of mechanical ventilation and ICU stay offset the costs of these resources.

Finally, it is worth noting that while the strategy was successfully implemented in many patients, 18% of those patients randomized to the no-sedation group ultimately required infusions of sedative medications. This tended to be in the patients with severe ARDS and indicates that the strategy might not be suitable for all individuals. Interestingly, the mean APACHE II scores were fairly high in the intervention and control groups (average score, 26 in both groups), suggesting that severity of illness alone may not predict the need for sedation. In fact, the more severely ill patients may be more encephalopathic and, as a result, may need less sedation than less ill patients who are more alert and, as a result, more uncomfortable with the endotracheal tube and other interventions. Follow-up studies of the no-sedation approach will need to address the question of which groups are appropriate for this particular strategy.