Sometimes a question is posed on Twitter that generates a great discussion from colleagues ’round the globe. Such was the case for dexmedetomidine. Although benzodiazepines remain the standard of treatment for ethanol withdrawal, particularly seizures and delirium tremens, what’s all the hype about dexmedetomidine?
Alpha-2 agonists that reduce sympathetic output may be effective adjunct treatment modalities without suppressing respiratory drive. Some older studies using clonidine demonstrated possible benefit. Similarly, there are several case reports and four case series using dexmedetomidine (a parenteral alpha-2 agonist) in addition to benzodiazepines. In 2014, the first randomized, placebo controlled trial was published evaluating dexmedetomidine in this role. FDA-approved dosing is 0.2-0.7 mcg/kg/hr as a continuous infusion. 1
2012/2013 – Case Series
- A retrospective case series of 10 ICU patients demonstrated safety of dexmedetomidine for ethanol withdrawal. Only 3 patients needed intubation, less benzodiazepines were required, heart rate was reduced up to 10 bpm, and blood pressure was decreased up to 3 mm Hg. Maximum dexmedetomidine dose used was 1.2 mcg/kg/hr. 2
- A retrospective case series of 20 ICU patients demonstrated a 62% reduction in benzodiazepine dosing after initiation of dexmedetomidine and a 21% reduction in alcohol withdrawal severity score. Only one patient required intubation. Dexmedetomidine was stopped in one patient who had two 9-second asystolic pauses noted on telemetry. 3
- A prospective case series of 18 ICU patients further demonstrated safety of dexmedetomidine for ethanol withdrawal when administered for a mean 24 hours. No patients required intubation. Time to resolution of alcohol withdrawal was 3.8 days, with and ICU length of stay 7.1 days, and hospital length of stay 12.1 days. Maximum dexmedetomidine dose used was 1.5 mcg/kg/hr. 4
- A retrospective case series of 33 ICU patients used dexmedetomidine as an adjunct therapy to benzodiazepines titrating to a rate of 0.7 mcg/kg/hr. In the 12 hours after dexmedetomidine began, patients experienced a 20 mg reduction in median cumulative benzodiazepine dose used (P < 0.001), a 14 mmHg lower mean arterial pressure (P = 0.03), and a 17 bpm reduction in median heart rate (P < 0.001). Four (12%) patients experienced hypotension (systolic blood pressure < 80 mmHg) during therapy and there were no cases of bradycardia (heart rate < 40 bpm). 5
2014 – The First Randomized Trial
A randomized, double-blind trial evaluated 24 ICU patients with severe ethanol withdrawal. 6 Group 1 received: Lorazepam + Placebo, Group 2 received: Lorazepam + Dexmedetomidine (doses of 0.4 mcg/kg/hr and 1.2 mcg/kg/hr). Pertinent results include the following: 24-hour lorazepam requirements were reduced from 56 mg to 8 mg in the dexmedetomidine group (p=0.037); 7-day cumulative lorazepam requirements were similar; Clinical Institute Withdrawal Assessment or Riker sedation-agitation scale scores were similar between within 24 hours; and bradycardia occurred more frequently in the dexmedetomidine group.
My thoughts: In my opinion, a major limitation of this study is that patients had more than 24 hours of treatment before randomization. Eleven of the 24 patients were already intubated when the trial started. The best place to use dexmedetomidine is in the early treatment course to help avoid intubation altogether. I’m not sure this very exclusive trial (24 patients included/209 excluded over 4 years) provides any answers for the patients who may benefit most from this therapy. I really feel a reduction in benzodiazepines is not even the correct outcome to measure. That may lead to the erroneous notion that we don’t need benzodiazepines. We should be looking at reduction in intubations and ICU length of stay.
2015 – The Second Randomized Trial
Another RCT looked at a similar primary outcome: 24-h diazepam consumption and cumulative diazepam dose required over the course of the ICU stay. 7 36 patients received dexmedetomidine at a dose of 0.2-1.4 μg/kg/h and titrated to achieve the target sedation level (-2 to 0 RASS) with symptom-triggered BZD (10 mg diazepam bolus). The other 36 patients received only symptom-triggered 10 mg boluses of diazepam. Median 24-h diazepam consumption during the study was significantly lower in the intervention group (20 vs. 40 mg, p < 0.001), as well as median cumulative diazepam dose during the ICU stay (60 vs. 90 mg, p < 0.001). Bradycardia was a common adverse event in the dexmedetomidine group (10 vs. 2; p = 0.03).
My thoughts: Even if benzodiazepine reduction was a good outcome to measure (I don’t think that it is), a 20 mg decrease in diazepam between the groups is not a clinically meaningful change that warrants a high-cost medication such as dexmedetomidine. If patients only received 60-90 mg of diazepam during their ICU stay, then these clearly weren’t severe withdrawal patients.
2016 – A Retrospective Controlled Cohort Study
The same authors who published the first randomized trial 6 followed it up with a retrospective cohort study 8 . Twenty patients receiving dexmedetomidine were matched to 22 control patients. The mean 12-hour change in benzodiazepine requirement was significantly different for dexmedetomidine versus control (-20 vs -8.3 mg, P = .0455) with a trend toward significance at 24 hours (-29.6 vs -11 mg, P = .06). Patients receiving dexmedetomidine experienced significantly more bradycardia than controls (35% vs 0%, P < .01) but not hypotension.
My thoughts: Interestingly, the study periods of the two studies by the same authors overlap and it appears many of the same patients were in both studies. In fact, Dr Lewis Nelson (@LNelsonMD) and I published a letter to the editor in response to this study, citing several problems. 9 Although the study is titled “Early Dexmedetomidine Addition…” patients were enrolled up to 60 hours after hospital admission.
- Dexmedetomidine may be a useful adjunct to benzodiazepines for ethanol withdrawal patients (in the ED or ICU).
- Reduced benzodiazepine requirements have been observed in both RCTs, but only in the short-term (24 hours).
- Dexmedetomidine does not suppress the respiratory drive and can be administered to non-intubated patients. Bradycardia and possibly hypotension are the major adverse effects with dexmedetomidine use.
- Benzodiazepines are still first, second, and third line therapy for ethanol withdrawal. Other GABA agents such as phenobarbital and propofol remain good next line options after benzodiazepines. Even ketamine may have a future role, with the first study published in 2015.
- Future studies should stop looking at a reduction of benzodiazepine requirements as an endpoint and instead focus on outcomes such as reduction in intubations and ICU admissions. They should also evaluate dexmedetomidine earlier in the severe withdrawal course, but only after appropriately high benzodiazepines (or barbiturates) have been administered.
Translation to Clinical Care
Some are using adjunct alpha-2 agonists with success (with benzodiazepines).
Many institutions utilize nurse-driven alcohol withdrawal scales allowing them the autonomy to give patients symptom-triggered benzodiazepine therapy based on an objective score. ED nurses may not be as familiar with proper titration of dexmedetomidine, particularly in non-intubated patients. It may be prudent to have guidelines in place to ensure safe administration and good patient outcomes.
Original: February 7, 2013
Updated: April 30, 2014; May 4, 2016