Secondary Logo

Journal Logo

Myths in Emergency Medicine: Antiemetics of No Benefit for Patients Receiving Opioids

May, Natalie MBCHB, MPHE

doi: 10.1097/01.EEM.0000476251.38207.f1
Myths in Emergency Medicine

Dr. Mayis a consultant emergency physician with specialty training in pediatric emergency medicine at John Radcliffe Hospital and Horton General Hospital, both in Oxfordshire, UK. She writes regularly for Follow her on Twitter @_NMay.





Perhaps it's time we admitted it: We're scared of opioids.

We once were afraid of overdosing our patients so we used to undertreat their pain. (I'm sure that still happens, but less so, I hope.)

But opioids are not risk-free drugs. If we're honest, we're probably worried about making patients' suffering worse by causing nausea and vomiting. We in the UK routinely administer antiemetics to patients receiving intravenous morphine, particularly in the prehospital environment. But does it help? And is it even necessary?

Lambie, et al. randomized a group of 214 patients with musculoskeletal trauma slated to receive intravenous opioids to be given 10 mg metoclopramide or placebo prior to receiving morphine. In the subsequent two hours of observation, 3.7 percent of subjects vomited — 1.9 percent of the placebo group and 5.4 percent of the metoclopramide group. (Emerg Med Australas 1999;11[4]:240.) That's right; it's not a typo. Patients were more likely to vomit in the group given prophylactic antiemetic.

Back to Top | Article Outline

Nausea with Placebo

A year later, an American Journal of Emergency Medicine study randomized 122 patients to metoclopramide or placebo following intravenous opioid, and found a similarly low rate of overall nausea (5.7%) and vomiting (0.8%). (Am J Emerg Med 2000;18[6]:653.) They did see more nausea in the placebo group at 30 minutes (6.8% vs. 3.2% in the metoclopramide group) but higher rates of nausea in the metoclopramide group at 60 minutes (4.8% vs. 3.4% for the placebo group) with higher rates of other side effects — dystonia, dizziness, vertigo — in the metoclopramide group (7.9% vs. 3.4%).

Bradshaw and Sen also randomized 259 patients receiving intravenous opioids to metoclopramide or placebo in 2006, and found similarly low incidence of nausea and vomiting at 2.7 percent overall, with 1.6 percent in the metoclopramide group and 3.7 percent in the placebo group. (Emerg Med J 2006;23[3]:210.)

Simpson, et al. combined these studies into a meta-analysis, and their pooled data found no statistically significant difference in vomiting between metoclopramide or placebo groups, with an odds ratio of 0.72 (95% CI 0.11-4.51, p=0.73). (Emerg Med Australas 2011;23[4]:452.) They raise some concerns about relationships between morphine dosing and incidence, noting that patients in all three studies received less than 10 mg of morphine, and speculate that motion sickness might provide a potential confounder if studies are undertaken specifically in the prehospital population.

So maybe drugs don't work, but does it matter? Our perception that nausea and vomiting are commonly occurring events does not hold with the literature. Smith, et al. found that nausea occurred with opioid administration in about one-fifth to one-third of cases, with the incidence of vomiting occurring in approximately half of these cases. (Ann Palliat Med 2012;1[2]:121.) Much published literature around opioid-associated nausea and vomiting comes from post-operative patients who, as a group, are likely to have more medications co-administered than our typical ED patient in pain, making a direct comparison troublesome. The literature examining opioid administration in the emergency department seems skewed toward the lower end of their quoted range.

Back to Top | Article Outline

Low Incidence

Paolini and Talbot-Stern looked specifically at rates of opioid-associated nausea and vomiting in the ED population back in 2002, finding an incidence of vomiting of 1.5 percent at 30 minutes and 2.4 percent at 60 minutes after intravenous opioids, with corresponding nausea at 4.9 percent at 30 minutes and 9.3 percent at 60 minutes. (Am J Emerg Med 2002;20[7]:604.) They found more than 75 percent of patients rated their nausea as mild, and they make the case that prophylaxis probably is not indicated with such a low incidence.

Finally, Sussman, et al. conducted a multicenter, randomized, double-blinded, placebo-controlled trial in which 2574 patients presenting acutely with nonsurgical pain requiring opioids received placebo, low-dose ondansetron (8 mg), or higher dose ondansetron (16 mg). (Clin Ther 1999;21[7]:1216.) This was the only ED-based paper finding a similar incidence of nausea and vomiting even approaching the range given by Smith, et al., and even this was at their lower end — 520 patients (20.3%). Sussman and colleagues found a higher proportion of complete control of emesis during the 24-hour follow-up period in both ondansetron groups (62.3% for the 8 mg group, 68.7% for the 16 mg group) than in the placebo group (45.7%), with similar proportions of adverse events across all three groups.

So what should we do when we administer opioids in the emergency department?

  • Give opioids when patients need them.
  • Because of a low background rate, do not co-administer an antiemetic as prophylaxis against nausea and vomiting; it has no demonstrable benefit.
  • If the patient develops nausea or vomiting which is distressing them, 8 mg or 16 mg of ondansetron is likely to give symptomatic relief and is a more evidence-based choice than metoclopramide.

Share this article on Twitter and Facebook.

Access the links in EMN by reading this on our website or in our free iPad app, both available at

Comments? Write to us at

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.