MENDS2 is a multicenter, double-blind trial comparing dexmedetomidine to propofol for sedation of intubated adults with sepsis in the United States.1 The primary endpoint was days alive without delirium or coma. There were no differences in any of the endpoints:
That’s not shocking, because both propofol and dexmedetomidine are excellent sedatives. Prior studies have failed to find any definitive winner between the two. Although we’re all addicted to the horserace of evidence-based medicine (WHICH ONE IS BETTER?), the truth is probably that their performance is equivalent among heterogeneous groups of patients. Individual patients may nonetheless respond better to one or the other.
One peculiarity of the study may limit this conclusion, however. The doses of propofol and dexmedetomidine used were tiny (a median of 10 ug/kg/min and 0.27 ug/kg/hr, respectively). It’s hard to believe that 10 ug/kg/min of propofol is doing much of anything here (this is technically defined as a “whiff” of propofol). So what is going on?
Patients didn’t receive much sedative, but they did receive a fair amount of opioids. The median fentanyl dose was 60 ug/hour or ~1,440 mcg/day. This is roughly equivalent to ~144 mg/day of IV morphine. In comparison, two European studies involving intubated patients who received no sedation found that patients received ~10 mg of IV morphine daily.2,3
Why do the American patients receive so much more opioids? In the MENDS2 trial, it appears that fentanyl was often used for its sedative properties (rather than its analgesic properties). For example, the study protocol explicitly specifies that fentanyl may be used to provide rescue sedation (figure below). Fentanyl is a potent analgesic but a weak sedative, so if it is being used for its sedative properties then you will need a lot.
Alternatively, in the European studies, patients were sufficiently awake to communicate with caregivers regarding pain. When patients were able to communicate the difference between pain and anxiety, it turned out that the actual amount of pain was not very high (as reflected in the low morphine requirement). This suggests that we may often be overestimating the amount of pain that our patients are experiencing. Thus, when high doses of opioids are used to achieve comfort, these drugs may be acting via sedative and euphoric effects – not analgesic effects.
Ideally, we should be trying to sort out pain, anxiety, and delirium. Pain should be treated with analgesics, anxiety with sedatives, and delirium with antipsychotics. In reality, this is not always easy to achieve. Among patients who cannot communicate, these states of disquietude may all appear similar. When comparing the American and European approaches, it seems that a “pain first” approach, popularized in the United States with aggressive use of opioids may often be leading to overutilization of opioids, as a broad-spectrum “quiet-down” drug to treat anxiety or agitation.
The use of continuous fentanyl infusions is fraught with hazard (e.g., accumulation, dependence, and withdrawal – discussed further here). Over the past several years, I’ve largely eliminated these from my practice (by using a multi-modal analgesic approach and PRN opioids). After years of using high-dose opioid infusions, it wasn’t very difficult to stop using them. However, with lower doses of PRN opioids, a more meaningful dose of sedatives will usually be required (10 mcg/kg/min of propofol just won’t cut it).
The MENDS2 trial was intended as a study comparing dexmedetomidine to propofol. However, the doses of these medications used in the study weren’t high enough to be impactful. In retrospect, the study may actually be an investigation of how clinicians approach the agitated intubated patient – potentially revealing an over-reliance on opioid infusions.
- 1.Hughes CG, Mailloux PT, Devlin JW, et al. Dexmedetomidine or Propofol for Sedation in Mechanically Ventilated Adults with Sepsis. N Engl J Med. Published online February 2, 2021. doi:10.1056/nejmoa2024922
- 2.Strøm T, Martinussen T, Toft P. A protocol of no sedation for critically ill patients receiving mechanical ventilation: a randomised trial. Lancet. 2010;375(9713):475-480. doi:10.1016/S0140-6736(09)62072-9
- 3.Olsen H, Nedergaard H, Strøm T, et al. Nonsedation or Light Sedation in Critically Ill, Mechanically Ventilated Patients. N Engl J Med. 2020;382(12):1103-1111. doi:10.1056/NEJMoa1906759