Date: January 4th, 2015

Guest Skeptic: Eve Purdy is a fourth year medical student from Queen’s University. Eve is interested in all things medicine but especially emergency medicine, rural medicine, medical education and social media. She is the creative force behind the excellent medical student blog Manu et Corde (Hand and Heart). She is also an editor for BoringEM. When not learning to be a doctor, Eve is usually running, playing guitar or planning an outdoor adventure.

Case: You are working in the emergency department when you pick up the chart for an otherwise healthy 35-year-old male with the chief complaint of abdominal pain and nausea. On history you learn that he has had crampy generalized abdominal pain for the past 24 hours associated with one episode of emesis. The pain started after he went out for wings night with some of his buddies and he hasn’t had much by mouth since then. He has been keeping down a bit of water. His last bowel movement was 12 hours ago and it was “loose” but was a normal colour with no blood. He hasn’t had any surgeries, he hasn’t been traveling, but is a grade 3 teacher and many of the kids in his class have had “gastro”. In the department he vomits while you are taking a history. His vital signs are normal and his abdomen is soft with some mild tenderness but no peritoneal signs. How will you treat his symptoms?

Background: Nausea and/or vomiting are common emergency department presentations. While investigating underlying cause and establishing a diagnosis are important, so too is the goal of relieving the patient’s symptoms. The success of pharmacologic anti-emetic strategies in oncology and post-operative patients (1, 2) has been extrapolated to support their use in patients with un-differentiated nausea and vomiting in the ED. Four studies (3, 4, 5, 6) have shown success of metoclopramide and/or ondansetron in reducing the severity of nausea in the ED but the only two placebo controlled studies showed no benefit of these medications over placebo (3, 4). Severity of nausea and vomiting is frequently measured using a visual analogue scale (VAS) and a minimally significant change has previously been defined as 15mm (7).

Clinical Question: What anti-emetic is most effective for Emergency Department patients with undifferentiated nausea?

Reference: Egerton-Warburton et al. Antiemetic Use for Nausea and Vomiting in Adult Emergency Department Patients: Randomized Controlled Trial Comparing Ondansetron, Metoclopramide, and Placebo. Annals of Emergency Medicine 2014.

  • Population: Adult patients with nausea and vomiting during emergency department care for which the physician prescribed intravenous anti-emetics
    • Exclusions: Hemodynamic instability, critical intervention needed, pregnancy or lactation, Parkinson’s disease, restless leg syndrome, use of antemtic in last 8 hours, previous IV fluids during emergency department stay, nausea and vomiting related to vertigo/chemotherapy/radiotherapy or previous allergy to a study medication.
  • Intervention: Metoclopramide 20mg IV (10mg/2ml x 2-2ml syringes) or Ondansetron 4mg IV (4mg/2ml x 1-2ml syringe + 0.9% saline x 1-2ml syringe )
  • Control:0.9% Saline 4ml (0.9% saline x 2-2ml syringes)
  • Outcomes:  
    • Primary Outcome: Mean change in severity rating on the visual analog scale 30 minutes after administration of study drug
    • Secondary Outcomes: Median change in severity on the numeric rating scale, adjectival description of change, change in number of vomiting episodes, need for rescue medication, patient satisfaction and adverse events

Authors’ Conclusions: “In summary, this study found that although 20 mg intravenous metoclopramide and 4 mg intravenous ondansetron resulted in slightly greater VAS score reductions than saline solution placebo, differences did not reach significance. Comparable majorities in each group also reported symptom improvement and satisfaction with treatment. This supports the findings of the other placebo- and nonplacebo-controlled studies, which also suggest that all antiemetic drugs, with the possible exception of droperidol, are similar.4….This adds weight to a recommendation that drug use not be routine and that condition-specific treatments, where possible, and other supportive measures, such as provision of intravenous fluids, be undertaken in the first instance.”

Quality Checklist for Randomized Clinical Trials:checklist-cartoon

  1. The study population included or focused on those in the ED. Yes. This study targeted ED patients specifically which is important because most of the data around anti-emetic use was not performed in this population.
  2. The patients were adequately randomized. Yes
  3. The randomization process was concealed. Yes
  4. The patients were analyzed in the groups to which they were randomized. Yes
  5. The study patients were recruited consecutively (i.e. no selection bias). No. 744 eligibility checked, 385 eligible, 270 randomized. Convenience sample (see commentary below).
  6. The patients in both groups were similar with respect to prognostic factors. Yes
  7. All participants (patients, clinicians, outcome assessors) were unaware of group allocation. Yes
  8. All groups were treated equally except for the intervention. No
  9. Follow-up was complete (i.e. at least 80% for both groups). Yes
  10. All patient-important outcomes were considered. No
  11. The treatment effect was large enough and precise enough to be clinically significant. No

Key Results:

  • The differences in mean VAS score change for ondansetron, metoclopramide, and placebo of 27 mm (95% CI 22 to 33 mm), 28 mm (95% CI 22 to 34 mm), and 23 mm (95% CI 16 to 30 mm), respectively, were not statistically significant between the 3 groups.
  • Less need for rescue medication in the metoclopramide group (18%) compared to ondansetron (35%) and placebo (36%). No statistically significant differences in the other secondary outcomes.
  • Nine adverse events were reported (3.5%) with six were in the metoclopramide group. Of those, two had akathisia, two had restlessness, one had muscle twitching, and one was diaphoretic. There were also two minor adverse events with ondansetron and one with placebo.

SGEM Commentary:

  1. Sampling: all physicians and nurses were trained in recruiting but a “convenience sample” of patients was recruited based on how busy the department was at a given time. There is no data provided to support that this sample was overall representative of patients coming through the ED. It sounds as though patients were unlikely to be recruited during busy ED times, which likely created some degree of sampling bias.
  2. Differential Co-Treatment: unknown whether groups were treated the same as we have no data on whether opioids/steroids/other medications were given differently to each group. Since there is no guarantee in the protocol that groups remained similar throughout ED stay, we are left to hope that proper blinding prevented against any systematic confounding bias towards or against a specific treatment.
  3. Patient Important Outcomes: The study endpoint was symptoms at 30 minutes, however, nausea often comes in waves rather than being a persistent phenomenon. As such, it would have been helpful to see comparison at a number of different evaluation time points (ie. 60 minutes, 120 minutes) to account for more realistic symptomatology and provide information that may be relevant when considering patient discharge.
  4. Medication Dosing and Type: The dosing of medications must be considered. The recommended dose of ondansetron is 0.15mg/kg so it could be argued that patients were actually under dosed in this trial by receiving 4mg. Conversely, metoclopramide is most often dosed at 10mg (rather than 20mg) so the increased number of side effects may have been attributable to that. Unfortunately this trial did not include anti-emetics delivered PO. IM or SL. We often administer medications this way to avoid an IV. We can’t extrapolate the results from this study for those alternate antiemetic strategies.
Eve Purdy

Eve Purdy

Comment on Authors’ Conclusion Compared to SGEM Conclusion: Overall this study shows that most adults that we see in the ED with nausea and vomiting do improve. This is a convenience sample of 270 ED patients that may have been selected with some degree of selection bias and the medications may not have been optimally dosed. It does not provide convincing evidence that anti-emetics have no effect in ED patients with nausea and vomiting but it certainly does question routine use of these medications.

The author’s conclusions are actually fairly balanced and agree that this paper simply adds to a growing body of evidence. The authors’ conclusions were that it “adds weight to a recommendation that drug use not be routine and that condition-specific treatments, where possible, and other supportive measures, such as provision of intravenous fluids, be undertaken in the first instance”

SGEM Bottom Line: Intravenous ondansetron and metoclopramide are not superiour than placebo at improving patient perceptions of nausea and vomiting along a visual analogue scale 30 minutes after administration but all three provide a clinically significant improvement in symptoms.

Case Resolution: You return to the 35 year old. You discuss with him that based on his current symptoms you are not worried that something dangerous is causing his vomiting. You let him know that there are options for IV medications to treat nausea but that they are no better than placebo for patients like him. He opts not to get poked. You highlight that the main key is that he needs to stay hydrated and encourage him to take small sips of electrolyte rich beverage and discharge him home with specific instructions about when to return.

Clinically Application: This study is a reminder that identifying a cause for nausea and vomiting then targeting treatment to that cause is likely more effective than a shotgun approach to all undifferentiated nausea and vomiting.

What Do I Tell My Patient? We will rule out worrisome/dangerous causes of nausea and vomiting. Most people improve with hydration and we are unsure if medications help. If you get worse, your symptoms change (fever, GI bleeding, pain, etc) or you are otherwise worried we are happy to see you again.

Keener Kontest: Last week’s keener/gunner winner was Dr. Johnny Nie a second year resident from Queen’s University. He knew approximately 80 million bacteria are transmitted through 10 seconds of intimate kissing (tonsil hockey).

Screen Shot 2014-12-17 at 12.26.47 AMListen to the show this week for the keener/gunner question. Send your answer to TheSGEM@gmail.com with “keener” or “gunner” in the subject line. The first correct answer will receive a cool skeptical prize.


Remember to be skeptical of anything you learn, even if you heard it on the Skeptics’ Guide to Emergency Medicine.