I am not sure which I prescribe more often each shift: analgesics or antiemetics. Despite having medical directives for both, I am still interrupted multiple times every hour with requests to chart one or the other. To be fair, I am never bothered. Nausea and vomiting are horrible symptoms, and I would want to be treated rapidly myself. However, we now have multiple trials saying that antiemetics don’t do anything. Have I been using an expensive placebo? Should I give up? Is it time to descend into nihilism?
Meek R, Mee MJ, Egerton-Warburton D, et al. Randomized Placebo-controlled Trial of Droperidol and Ondansetron for Adult Emergency Department Patients With Nausea. Academic emergency medicine. 2019; 26(8):867-877. PMID: 30368981 Australian New Zealand Clinical Trials Registry ACTRN12617000224325
This is a triple-blind, randomized, controlled trial that was designed to demonstrate the superiority of two antiemetic drugs, droperidol and ondansetron, over placebo.
Patients 18 years and older with nausea rated as 4 or more out of 10, from any cause (except the cancer patients excluded below).
- Exclusions: allergy to ondansetron or droperidol, use of an antiemetic in the last 4 hours, too unwell to participate (at the discretion of the attending physician), contraindication to a normal saline infusion (e.g., fluid-restricted patients), Parkinson’s disease or restless leg syndrome, current use of a dopamine antagonist, cognitive impairment or language barrier compromising study understanding, pregnant or breastfeeding women, chemotherapy- or radiotherapy-induced nausea.
Intervention #1: Droperidol 1.25 mg IV
Intervention #2: Ondansetron 8mg IV
For reasons that are not entirely clear to me, all patients received 1 L of normal saline over 4 hours.
The primary outcome was the percentage of patients with a minimum of 8 mm improvement on the visual analog scale at 30 minutes (they increased this outcome from 5 mm after starting the trial based on as yet unpublished data).
They enrolled 215 patients (unfortunately, this was only about half way to the calculated sample size of 378 patients).
The primary outcome, an improvement of 8mm on the VAS, occurred in 75% of the droperidol group, 80% of the ondansetron group, and 76% of the placebo group (not statistically significant).
The mean change on the VAS was 29 mm with droperidol, 34 mm with ondansetron, and 24 mm with placebo (representing a 50%, 55% and 41% relative improvement respectively).
Interestingly, despite the above numbers being unchanged, the number of patients who subjectively “achieved the desired effect” was statistically higher in the droperidol group than placebo (77% vs 59%, ARR= 18%, 95% CI 3-33%, NNT = 5). It was also higher in the ondansetron group, but not quite statistically so (73% vs 59%, ARR= 14%, 95% CI -1-29%). Similarly, the droperidol group seemed to require less rescue medication (15%, 23%, and 30% respectively).
There was drowsiness with droperidol (37% vs 13% and 17%), but other adverse events look pretty similar (in this significantly under-powered trial).
The simple conclusion from this paper would be that antiemetics don’t work for emergency department patients. That would seem to fit with what we have seen from a number of recent trials. (Barrett 2011; Egerton-Warburton 2014) However, I am not sure that this trial really shows no benefit from the anti-emetics, and especially not droperidol.
First, the trial was stopped half way, and so is significantly underpowered. Second, the trial was originally powered to look at the percentage of patients who achieved what the authors considered a clinically important change: 5 mm on the VAS. The mean improvement was 29 mm with droperidol and 34 with ondansetron – so people did get better. Finally, although there was no difference in the nausea scores between the treatment groups and placebo, patients clearly knew which group they were in. More patients achieved the desired effect and fewer required rescue medications in the active treatment groups, and those results were statistically significant with droperidol. (Of course, unblinding because of the side effects of droperidol is a concern.)
That being said, I think you need to be really careful when interpreting data on rescue medications. The results may not be as straightforward as they seem. We expect that if you receive little relief from your first drug, that you will request a second. However, perhaps people who saw no response from the first medication didn’t ask for a second, because drugs aren’t working, so what’s the point of taking more? Or, the first medication may have caused significant side effects, lowering the desire for a second medication. Or maybe the fist medication, instead of being neutral, actually made patients feel worse. It’s complicated.
Previous research on antiemetics in the emergency depatment is somewhat mixed. There is RCT data that droperidol (but not metoclopramide or prochlorperazine) is better than placebo at reducing nausea. (Braude 2006) There is also a Cochrane review that demonstrates a statistical (and probably clinically important) decrease in nausea with droperidol, but not with any other anti-emetic. (Furyk 2015)
The data for ondansetron is a little more confusing. There are multiple prior studies that showed no difference between ondansetron (and metoclopramide) and placebo. (Barrett 2011; Egerton-Warburton 2014) However, despite not being any better than placebo at decreasing nausea, there is good evidence that ondansetron decreases vomiting, admissions, and failure of oral hydration in pediatric vomiting, with a very low NNT. (DeCamp 2008; Danewa 2016)
So how can anti-emetics be useless and so very useful at the same time?
Nausea improves significantly in all of these trials. The reason that the trials are negative is not that the antiemetic group is no better, but that the placebo group has improved just as much. Nausea may rapidly improve after the initial ED assessment, but nausea comes and goes in cycles. Most of these trials measured nausea at 30 minutes. Perhaps that is too early to see the benefit that anti-emetics have on the next wave of nausea? Or maybe the patients we are treating (and including in trials) are just too healthy?
Furthermore, placebo is not the same thing as no treatment in these trials. Patients are being treated for whatever brought them in. For example, a patient with cholelithiasis might also be getting NSAIDs and opioids to control their pain, and that reduction in pain could improve nausea, or simply make the same level of nausea seem less bad.
There is a difference between nausea and vomiting. From the pediatrics literature, we have a pretty good sense that ondansetron decreases vomiting. Perhaps ondansetron truly doesn’t decrease nausea, but still decreases vomiting (which I would still count as beneficial). Perhaps we are measuring the wrong thing.
We might also be over-simplifying nausea and vomiting. Nausea and vomiting are caused by a large variety of disease states. The underlying pathophysiology is often quite different. Vomiting from vertigo after a cerebellar stroke is not the same as vomiting from poison mushrooms or vomiting from a bowel obstruction. Perhaps the antiemetics are useful in some states, but futile in others. The various antiemetics we use are pharacologically quite different. Perhaps they need to be tailored to the pathophysiology of the patient. Our current approach, and the approach in these studies, is somewhat unsophisticated.
I think we are almost certainly over-using antiemetics. I probably should be more judicious in my use, but I hate leaving nausea untreated. I personally find nausea far more distressing than pain, and when I am sick, I certainly want to try something. That being said, all medications have side effects, and if these agents really aren’t effective, the patient will still feel awful in an hour, but with the additional side effects of the antiemetic. The routine use of these agents also suggests that patients need to rush to the emergency room every time they start vomiting.
So what should we do? From a research standpoint, I think we need to start looking at subgroups; trying to match the pharmacology of the agent to the pathophysiology of the patients. Clinically, the way forward is less clear. When I look at this evidence, I still see a suggestion that droperidol helps. We know that ondansetron works in pediatrics. On the other hand, I don’t want to harm my patients by prescribing a medication with no benefit. I think I may slightly curtail my use of these agents, but I don’t think this literature convinces me to make any drastic practice changes.
This study demonstrated no improvement in nausea with droperidol and ondansetron. There were some secondary outcomes suggesting a treatment effect from droperidol, but there were also side effects. It is very hard to know what to do with this information. We probably need to use these medications less often, but relieving suffering is a huge part of our job, and nausea is a significant cause of suffering. I don’t think I am changing my practice significantly yet.
Barrett TW, DiPersio DM, Jenkins CA, et al. A randomized, placebo-controlled trial of ondansetron, metoclopramide, and promethazine in adults. The American journal of emergency medicine. 2011; 29(3):247-55. [pubmed]
Braude D, Soliz T, Crandall C, Hendey G, Andrews J, Weichenthal L. Antiemetics in the ED: a randomized controlled trial comparing 3 common agents. The American journal of emergency medicine. 2006; 24(2):177-82. PMID: 16490647
Danewa AS, Shah D, Batra P, Bhattacharya SK, Gupta P. Oral Ondansetron in Management of Dehydrating Diarrhea with Vomiting in Children Aged 3 Months to 5 Years: A Randomized Controlled Trial. The Journal of pediatrics. 169:105-109.e3. 2016. PMID: 26654135
DeCamp LR, Byerley JS, Doshi N, Steiner MJ. Use of antiemetic agents in acute gastroenteritis: a systematic review and meta-analysis. Archives of pediatrics & adolescent medicine. 2008; 162(9):858-65. [pubmed]
Furyk JS, Meek RA, Egerton-Warburton D. Drugs for the treatment of nausea and vomiting in adults in the emergency department setting. The Cochrane database of systematic reviews. 2015; [pubmed]
Meek R, Mee MJ, Egerton-Warburton D, et al. Randomized Placebo-controlled Trial of Droperidol and Ondansetron for Adult Emergency Department Patients With Nausea. Academic emergency medicine. 2019; 26(8):867-877.
Justin Morgenstern. Nausea and vomiting in the ED: Does nothing work?, First10EM, 2019. Available at: