SGEM#318: Why Am I Throwing Up – Because You Got High

On examination, his vital signs are normal, and he is afebrile. The abdomen exam shows no peritoneal sign and normal bowel sounds are heard. Laboratory values are unremarkable. An ultrasound does not show any free fluids or any signs of an Ileus, appendicitis or gallbladder disease.

His pain and nausea are difficult to control with standard medications. You admit him to hospital and the next day he undergoes gastroscopy which is unremarkable.

In the afternoon the patient is seen by a nurse when he is smoking “weed” (cannabis) in the garden of the hospital. He admits to being a heavy cannabis user and his symptoms do seem to get worse when smoking weed. You suspect he has cannabis hyperemesis syndrome and discharge him home with the recommendation to stop smoking as much weed.

Background: Chronic marijuana use was recognized by Allen el al in 2004 to cause cyclical vomiting in patients from South Australia. Roche and Foster quickly reported in 2005 that this was not an isolated problem to the Adelaide Hills of South Australia. The medical condition became known as cannabinoid hyperemesis syndrome. We covered this on SGEM#46: Don’t Pass the Dutchie

Cannabis stimulates two receptors: CB1 and CB2. CB1 is also expressed in the GI-system and reduces motility and relaxes the esophageal sphincter tonus. If you conduct chronical cannabis abuse, it seems that the anti-nausea effect of cannabis vanishes and there is a continuous hyperstimulation of CB1. That’s why you have abdominal pain and nausea with continuous vomiting.

There are some criteria proposed for the diagnosis of cannabinoid hyperemesis. An essential feature is long term cannabis use (often daily). There are five major features for the diagnosis and five supportive features for the diagnosis. These are listed in the table. 

Clinical Question: Can haloperidol effectively treat patients with cannabis hyperemesis syndrome?

Reference: Ruberto et al. Intravenous Haloperidol Versus Ondansetron for Cannabis Hyperemesis Syndrome (HaVOC): A Randomized, Controlled Trial. Annals of EM 2020.

• Population: Adult patients 18 years of age and older presenting to the ED with a working diagnosis of hyperemesis cause by cannabis who had at least two hours of ongoing, witnessed emesis or retching. Patients needed to report at least three episodes of emesis in a cyclic pattern separated by greater than one month during the preceding two years, and near-daily to daily use of cannabis by inhalation for at least six months.
  • Exclusion: Daily opioid users, allergic to or intolerant of either study drug, deemed unreliable for follow-up, or unlikely to return for crossover, pregnant or received an antiemetic, anticholinergic or antipsychotic agent intravenously (other than up to 100mg of dimenhydrinate) in the previous 24 hours.
• Intervention: Haloperidol low-dose (0.05mg/kg) or high-dose (0.1 mg/kg)
• Comparison: Ondansetron 8mg IV
• Outcome:
  • Primary Outcomes: Average reduction from baseline in abdominal pain and nausea scores measured on a 10-cm visual analog scale (VAS) at two hours after treatment
  • Secondary Outcomes: Changes in either abdominal pain or nausea score over time, treatment success (ie, both abdominal pain and nausea <2 cm at !2 hours), being discharge ready at two hours, use of rescue antiemetics before discharge, time to discharge readiness, length of stay (LOS) greater than 12 hours, and unscheduled return visits within seven days. Any adverse events potentially related to the study drug and specifically any acute dystonia or moderate to severe akathisia.

Authors’ Conclusions: “In this clinical trial, haloperidol was superior to ondansetron for the acute treatment of cannabis-associated hyperemesis. The efficacy of haloperidol over ondansetron provides insight into the pathophysiology of this now common diagnosis in many EDs”.

Quality Checklist for Randomized Clinical Trials:

1. The study population included or focused on those in the emergency department. Yes
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. Unsure
5. The study patients were recruited consecutively (i.e. no selection bias). No
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. Yes
9. Follow-up was complete (i.e. at least 80% for both groups). Yes
10. All patient-important outcomes were considered. Yes 
11. The treatment effect was large enough and precise enough to be clinically significant. Unsure

Key Results: There were 313 unique individuals with nausea, vomiting and cannabis use mentioned. A total of 65 patients met inclusion criteria when the research team was notified and only 30 (46%) agreed to participate and were randomized. The mean age was 29 years with slightly more men than women.

Haloperidol was superior compared to ondansetron for pain and nausea at two hours post treatment.

• Primary Outcome: Haloperidol vs ondansetron change from baseline
  • Nausea: –5.0 vs –2.4, difference –2.5 (95% CI; –4.4 to –0.6)
  • Pain: –4.3 vs –2.1, difference –2.2 (95% CI; –4.4 to 0)
• Secondary Outcomes: 
  • Haloperidol was superior to ondansetron in all secondary outcomes.
  • Three patients had prespecified adverse events (one moderate akathisia and two return visits for acute dystonia) all after the higher dose haloperidol (0.1 mg/kg). All three patients were treated without difficulty, discharged, and withdrawn from further study eligibility for crossover, given the compromised blind.
  • There were four return visits (two for dystonia and two for ongoing nausea and vomiting) within the week after haloperidol compared with six return visits (all for ongoing nausea and vomiting) after ondansetron.

This was a difficult study to successfully complete due to the inclusion and exclusion criteria, the cross-over design and the target population. The researchers should be congratulated for the data they were able to obtain.

1) Convenience Sample: These were not consecutive patients but rather a convenience sample. Participants were recruited only when on-site research personnel were available to facilitate enrollment. Out of the 313 unique patients presenting with nausea/vomiting and cannabis use, 201 people were excluded because the research team was not notified, or it was after hours. This could have introduced some selection bias.

2) Selection Bias: Speaking of selection bias, another possible source of this type of bias was that less than 50% of eligible patients agreed to participate (30/65). The authors said this was often due to patients being skeptical regarding the diagnosis as disclosed during the consent process.

Also, if the patient was deemed unreliable for follow-up, or unlikely to return for crossover they were excluded. This is a subjective criterion that could have introduced selection bias into the process. The researchers did were not very successful considering only 9 of 30 patients (30%) returned the 24- and 48-hour VAS scores despite reminders by telephone, text, and e-mail.

3) Small Population: This was a relatively small study with only 30 patients. They discussed the difficulties they had in reaching their prespecified enrollment target. Part of the problem was the inclusion criteria of needing witnessed emesis, the exclusion due to IV ondansetron being given as a standing medical order for all-cause emesis before clinical assessment, lack of research personnel (Nerdy point#1) and a belief by some clinicians that their practice approach was superior to the study protocol.

Despite the small numbers, they did see a large effect size that was statistically and clinically significant.

Another small number we should consider was the low crossover rate of 25% (being treated more than once). This means they could only perform an ANOVA for the first period alone for the primary efficacy analysis.

4) Stopped Early: This trial was stopped early. We have talked about this issue before on the SGEM. Guyatt et al discussed the problem in a 2012 BMJ article. They said: “Guidance must include a high level of skepticism regarding the findings of trials stopped early for benefit, particularly when those trials are relatively small and replication is limited or absent.” 

However, they did explain why they stopped early. It was a blinded unplanned interim analysis before the trial was extended to enroll participants from other sites. The analysis a strong effect on the primary outcome in favor of haloperidol and the recommendation was to stop the trial early.

Some guidance has been published around the issue of stopping trials early due to efficacy (Tyson et al 2016). There are ethical considerations in stopping clinical trials early (Deichmann et al 2016). Consolidated Standards of Reporting Trials (CONSORT) has some guidelines regarding this issue.

• Interim analyses and stopping guidelines: When applicable, an explanation of any interim analyses and stopping guidelines
  • Multiple statistical analyses can lead to false-positive results, especially when using stopping guidelines based on statistical significance. Any interim analyses should be described, including which analyses were conducted (i.e. the outcomes and methods of analysis), when they were conducted, and why (particularly whether they were pre-specified [78]). Authors should also describe the reasons for stopping the trial, including any procedures used to determine whether the trial would be stopped early (e.g. regular meetings of a data safety monitoring board) [79].
• Reason for stopped trial: Why the trial ended or was stopped?
  • Authors should state why the trial was stopped. Trials might be stopped for reasons decided a priori (e.g. sample size reached and predetermined follow-up period completed) or in response to the results. For trials stopped early in response to interim analyses (Item 7b), authors should state the reason for stopping (e.g. for safety or futility) and whether the stopping rule was decided a priori. If applicable, authors should describe other reasons for stopping, such as implementation challenges (e.g. could not recruit enough participants) or extrinsic factors (e.g. a natural disaster). Authors should indicate whether there are plans to continue collecting outcome data (e.g. long-term follow-up).

Since we have brought up the CONSORT statement, there is an extension to the 2010 guideline for randomized crossover trials (Kerry et al). Crossover trials are different than parallel RCTs in a number of ways. One key feature is that each individual act as their own control. It has been reported that the quality of crossover trials is variable which can impact their use in clinical decision making. These new guidelines were not published until 2019 after this study was designed and completed.

5) Adverse Events: This randomized crossover trial was powered to look for efficacy and not harm. We also know that harms are systematically under reported in RCTs (Hodkinson et al 2013).  These authors report four return visits within seven days after haloperidol (two for dystonia and two for ongoing nausea and vomiting) versus six return visits after ondansetron (all for ongoing nausea and vomiting). These numbers are too small to make any conclusions about which treatment strategy has less harm.

Comment on Authors’ Conclusion Compared to SGEM Conclusion: The authors concluding that haloperidol is superior to ondansetron in treating cannabis hyperemesis syndrome in their small trial is reasonable.

SGEM Bottom Line: Haloperidol appears superior to ondansetron in treating cannabis hyperemesis syndrome, but the evidence is weak and needs to be confirmed.

Dr. Johannes Pott

Case Resolution:  The patient is offered a number of treatment options including gets ondansetron, haloperidol or topical capsaicin in a shared decision model. He decides to try capsaicin first, but there is no reduction in abdominal pain. Therefore, the patient asks to be given low-dose haloperidol. His symptoms get better within two hours and he is discharged home.

Clinical Application: Cannabis hyperemesis syndrome is a challenge for both the patient and clinician. Haloperidol is another treatment option with weak evidence of efficacy but does come with known potential harms. The evidence for topical capsaicin is also weak for potential benefit but does not have the same potential harms as haloperidol. Lacking clear evidence to support one method over another the patient will be asked what they would prefer.

Dr. Thorben Doll

What Do I Tell My Patient?  The amount of cannabis you are using is most likely causing your pain and vomiting. It can be hard to treat these symptoms once they have started. The best way is prevention by decreasing your cannabis use. The traditional medicines we use are not very effective. There is some weak evidence that a drug called haloperidol could work better, get you home sooner but it comes with a risk of uncommon but serious side effects. Another option would be to rub capsaicin cream on your abdomen. There is some weak data for this treatment, and it does not come with any known serious side effects (Dean et al 2020). What would you like to do?

Keener Kontest: There was no winner last week because I made a mistake. The question was supposed to be what was the name previously used for cyclooxygenase (COX) with the answer being prostaglandin H synthase (PGHS). I am sorry for the confusion and obviously there was no winner.

Listen to the SGEM podcast to hear this weeks’ question. Send your answer to TheSGEM@gmail.com with “keener” in the subject line. The first correct answer will receive a cool skeptical prize. 

Other FOAMed:

• Pin Up Docs: Cannabis
• Rebel EM: The HaVOC Trial 

The SGEM Content is free but there is a small fee for the CME

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

References:

• Sorensen CJ, DeSanto K, Borgelt L, Phillips KT, Monte AA. Cannabinoid Hyperemesis Syndrome: Diagnosis, Pathophysiology, and Treatment-a Systematic Review. J Med Toxicol. 2017 Mar;13(1):71-87. doi: 10.1007/s13181-016-0595-z. Epub 2016 Dec 20. PMID: 28000146; PMCID: PMC5330965.
• Fleming JE, Lockwood S. Cannabinoid Hyperemesis Syndrome. Fed Pract. 2017;34(10):33-36.
• Guyatt GH, Briel M, Glasziou P, Bassler D, Montori VM. Problems of stopping trials early. BMJ. 2012 Jun 15;344:e3863. doi: 10.1136/bmj.e3863. Erratum in: BMJ. 2014;348:319. PMID: 22705814.
• Tyson JE, Pedroza C, Wallace D, D’Angio C, Bell EF, Das A. Stopping guidelines for an effectiveness trial: what should the protocol specify?. Trials. 2016;17(1):240. Published 2016 May 10. doi:10.1186/s13063-016-1367-4
• Deichmann RE, Krousel-Wood M, Breault J. Bioethics in Practice: Considerations for Stopping a Clinical Trial Early. Ochsner J. 2016;16(3):197-198.
• Dwan Kerry, Li Tianjing, Altman Douglas G, Elbourne Diana. CONSORT 2010 statement: extension to randomised crossover trials  BMJ 2019;  366 :l4378
• Hodkinson A, Kirkham JJ, Tudur-Smith C, Gamble C. Reporting of harms data in RCTs: a systematic review of empirical assessments against the CONSORT harms extension. BMJ Open. 2013;3(9):e003436. Published 2013 Sep 27. doi:10.1136/bmjopen-2013-003436
• Dean D, et al. A Pilot Trial of Topical Capsaicin Cream for Treatment of Cannabinoid Hyperemesis Syndrome. Academic Emergency Medicine. 2020. PMID: 32569429