Post-intubation Sedative options

Does dexmedetomidine confer a mortality benefit in patients with sepsis who undergo post-intubation sedation over other sedative agents? 

Study #1: Kawazoe Y et al 2017 (DESIRE trial):

This was an open-label, multicenter RCT of adult patients with sepsis requiring mechanical ventilation for > 24 hrs. Eligibility criteria were age ≥ 20 years; requiring mechanical ventilation for > 24 hrs (which included non-invasive ventilation strategies such as face mask, head mask, nasal mask, and cuirass); and with sepsis (defined as SIRS + source of infection, including acute pancreatitis due to need for antisepsis treatments and fluid resuscitation). Exclusion criteria included severe chronic liver disease (Child-Pugh grade B or C), acute MI, and NYHA class IV heart failure. The objective of the study was to determine whether sedation with dexmedetomidine would improve mortality and ventilator-free days at 28 d in the patient population defined above. 203 patients were initially enrolled, and 201 were ultimately randomized. 100 were randomized to the dexmedetomidine group. Patients in the dexmedetomidine group received both the sedative and analgesia concurrently, with other sedatives added as needed. In the dexmedetomidine group, at least 29 received propofol and 21 received midazolam.101 were randomized to the usual care group, with 6 receiving dexmedetomidine during the trial as per the attending physician’s discretion; all 101 were included in the final analysis. Patients in the usual care group received a variety of sedatives (such as propofol and midazolam) without dexmedetomidine, but also with analgesia. Sedation targets were a RASS of 0 during the day and -2 at night for both groups, although acceptable RASS scores ranged from +1 to -3. Post-hoc, adverse events and coma-free days (with coma defined as RASS between -4 and -5 for 1 d) were analyzed. The study was adequately powered to detect a 20% difference in the primary outcomes, and calculations were included. Intention-to-treat was used. There was a 19% 28-d mortality in the dexmedetomidine group and a 28% 28-d mortality in the usual care group (p = 0.14). The investigators argue that while not statistically significant, this is clinically significant. The median number of ventilator-free days at 28 d was 20 in the dexmedetomidine group and 18 in the usual care group (p = 0.20). No subgroup analysis showed statistical significance in primary outcomes. While not a primary outcome, subjects in the dexmedetomidine group demonstrated a greater percentage of patients hitting sedation RASS targets (p = 0.01), although it did not demonstrate a greater percentage of patients without delirium or coma (p = 0.17). 

Study #2: Shehabi Y et al 2019 (SPICE III trial). 

This was an open-label, multicenter RCT of adult patients requiring mechanical ventilation through an ETT for > 24 hrs. Inclusion criteria were receiving ventilation through an ETT, expected to receive ventilation for at least the next calendar day, and receiving sedatives for comfort and safety. Exclusion criteria were age < 18 years, invasive ventilation in the ICU for > 12 hrs prior to enrollment, and suspected or proven primary brain injury. The objective of the study was to determine whether sedation with dexmedetomidine as the primary agent would improve all-cause mortality at 90 d in the patient population defined above. 2001 patients were randomized to receive dexmedetomidine, and 1999 were randomized to receive usual care. There was withdrawal of consent in both groups after randomization, and 1954 had initial data analyzed in the dexmedetomidine group, and 1964 had initial data analyzed in the usual care group. 96 patients were lost to follow-up or later withdrew from the study (2.4%), resulting in 1948 patients in the dexmedetomidine group and 1956 patients in the usual care group who had data analyzed at 90 d. Sedation targets were a RASS of +1 to -2, assessed q4hrs.  Some patients in the trial had deeper RASS scores as targets due to clinical situations. It was encouraged to use only dexmedetomidine in the dexmedetomidine group, although it was acceptable to use propofol as a second-line agent when RASS targets could not be achieved with dexmedetomidine alone. Benzodiazepines were discouraged in the dexmedetomidine group except for specific indications. Medications administered in the usual care group included propofol and midazolam, and rescue dexmedetomidine could be used in the usual care group if indicated. Antipsychotic medications were administered in both groups if sedation could not otherwise be achieved. Crossover was reported; in the dexmedetomidine group, 64.7% of patients received propofol, 2.9% of patients received midazolam, and 6.9% received both. In the usual care group, use of rescue dexmedetomidine was not reported. The study was adequately powered to detect a difference of 4.5% with an estimated loss to follow-up of 5%. The study was analyzed using an intention-to-treat principle. There was a 29.1% 90-d mortality in the dexmedetomidine group and a 29.1% 90-d mortality in the usual care group (p = 0.98). No statistical significance was found while stratifying by sepsis/not sepsis. The only subgroup analysis that was statistically significant is that usual care was favored when patients were ≤ median age (63). While not a primary outcome, 56.6% of patients in the dexmedetomidine group reached target RASS levels, while 51.8% of patients in the usual care group reached target RASS levels. 40% of patients in the dexmedetomidine group had RASS scores of -3 to -5, while 45.6% in the usual care group had similar scores. There was no breakdown by indication for deep sedation.

Conclusions: 

Dexmedetomidine does not confer a mortality benefit over other sedative agents for patients who undergo intubation and post-intubation sedation for sepsis. Sedation-related mortality is more likely influenced by the depth and duration of sedation rather than the agent used to achieve it. It is better to choose a sedative agent based on patient factors rather than because one has proved superior in reducing ICU mortality.

A reminder: 

Mortality is not the only patient-centered outcome that matters in the ICU. There is a complex triad of symptoms that persist after ICU stays called PICS (post-intensive care syndrome). This syndrome comprises emotional dysregulation, cognitive dysfunction, and ICU-acquired weakness that occur after an ICU stay. Sedation is thought to play an important role in the pathogenesis of PICS. Further research is required to assess whether the choice of sedative influences these patient-centered outcomes. 

References

• Farkas, Josh. Sedation. The Internet Book of Critical Care. Accessed on March 4th, 2021. Available at https://emcrit.org/ibcc/sedate/.
• Kawazoe Y et al. Effect of dexmedetomidine on mortality and ventilator-free days in patients requiring mechanical ventilation with sepsis. JAMA. 2017;317(13):1321-1328. doi: 10.1001/jama.2017.2088.
• Needham D, Davidson J, Cohen H et al. Improving long-term outcomes after discharge from an intensive care unit: report from a stakeholders’ conference. Critical Care Medicine 40, 502-509 (2012). doi: 10.1097/CCM.0b013e318232da75.
• Pandharipande PP et al. Effect of sedation with dexmedetomidine vs lorazepam on acute brain dysfunction in mechanically ventilated patients: the MENDS randomized control trial. JAMA 2007;298(22):2644-2653. doi: 10.1001/jama.298.22.2644
• Shehabi Y et al. Early sedation with dexmedetomidine in critically ill patients. NEJM. 2019;380(26):2506-2517. doi: 10.1056/NEJMoa1904710. 
• Tanaka LMS et al. Early sedation and clinical outcomes of mechanically ventilated patients: a prospective multicenter cohort study. Critical Care. 2014;18(4):R156. doi: 10.1186/cc13995
• Weingart, Scott. EMCrit 283 – Dexmedetomidine (Precedex) – You’d have to be Delirious Not to Use It. EMCrit Blog. Published on October 16, 2020. Accessed on March 4th, 2021. Available at https://emcrit.org/emcrit/dexmedetomidine.