About the Rome IV Criteria for Cyclic Vomiting Syndrome
What defines cyclic vomiting syndrome is the shape of the illness over time, not the severity of any one attack. Rome IV requires stereotypical episodes of vomiting as to onset (acute) and duration (less than one week); at least three discrete episodes in the prior year *and* two in the past six months, occurring at least a week apart; and absence of vomiting between episodes, though milder interepisodic symptoms are permitted. Criteria are fulfilled over three months with onset at least six months ago. A personal or family history of migraine is a supportive remark, and it points toward the prophylaxis that works.
Formula
Cyclic vomiting syndrome = ALL of:
stereotypical episodes of vomiting, acute in onset and lasting under 1 week
AND at least 3 discrete episodes in the prior year
and 2 episodes in the past 6 months, at least 1 week apart
AND absence of vomiting between episodes
(milder interepisodic symptoms are permitted)
AND criteria fulfilled for the last 3 months, onset at least 6 months ago
Supportive: personal or family history of migraine headaches- Stereotypical
- The same episode each time in onset, duration and symptoms. Variability argues against the diagnosis and towards a structural or metabolic cause.
- 3 in a year and 2 in 6 months
- Rome IV states both counts explicitly, and requires episodes at least a week apart.
- Absence of vomiting between episodes
- Milder interepisodic symptoms are permitted; vomiting between episodes is not, and points to chronic nausea and vomiting instead.
- Rome IV specifies the episode counts explicitly and allows milder symptoms between episodes, which the paediatric-derived definition did not.
- A migraine history is supportive rather than required.
Interpreting the result
Meeting the criteria should immediately prompt two things. First, ask about cannabis in detail, because cannabinoid hyperemesis syndrome is defined by Rome IV as resembling this disorder and cannot be separated from it on the episodes alone — and it is far more common now than when these criteria were written. Second, split management into abortive and prophylactic. Abortive therapy during an attack includes triptans, anti-emetics and fluids, and works better the earlier it is given, which is an argument for a written plan the patient carries. Prophylaxis between attacks — commonly amitriptyline or topiramate — is what reduces episode frequency, and it is the half most often never started because each presentation is handled acutely and the patient is discharged. Ask about migraine, personally and in the family, since it supports the diagnosis and predicts response to prophylaxis. Where criteria are not met, check which one failed: interepisodic vomiting points to chronic nausea and vomiting syndrome, while unmet counts may simply mean the pattern has not yet declared itself and warrant review rather than exclusion.
| Score | Band | What it means | Action |
|---|---|---|---|
| All criteria met | Cyclic vomiting syndrome | Stereotypical acute episodes under a week, meeting the episode counts, with well periods in between | Ask about cannabis; start abortive therapy planning and prophylaxis. Ask about migraine, personal and family |
| Vomiting between episodes | Criteria not met — continuous pattern | Vomiting between attacks is incompatible with the cyclic pattern | Assess for chronic nausea and vomiting syndrome instead |
| Episode counts unmet | Criteria not met — pattern not yet established | Fewer than three episodes in the prior year or two in the past six months | Review rather than exclude; the pattern may declare itself with time. Keep an episode diary |
What the Cyclic Vomiting Syndrome needs (4 inputs)
- Timing
- Criteria fulfilled for the last three months with symptom onset at least six months before diagnosis.
- Stereotypical episodes of vomiting as to onset (acute) and duration (less than 1 week)
- Stereotypical is the operative word: the episodes resemble one another closely in how they start, how they feel and how long they last. That consistency is the diagnostic hallmark and is what distinguishes this from recurrent vomiting of varied cause.
- At least three discrete episodes in the prior year AND two in the past 6 months, at least 1 week apart
- Both counts are required, not either. The one-week separation exists to stop a single prolonged illness with fluctuation being counted as several episodes.
- Absence of vomiting between episodes (milder interepisodic symptoms permitted)
- Vomiting between attacks moves the diagnosis to chronic nausea and vomiting syndrome. Milder symptoms such as background nausea are explicitly allowed, and are common in adults — Rome IV notes this as a difference from the paediatric presentation.
What it returns
- Criteria met or not met
- All four requirements including the timing rule must hold simultaneously.
- Which criteria remain outstanding
- Named explicitly. Failure is most often on the episode counts or on interepisodic vomiting, and the two have different implications.
How it is calculated
The criteria are built entirely on pattern recognition across time, which is unusual among the Rome definitions and reflects what the disorder actually is. There is no biomarker, no characteristic finding during an attack, and nothing about the vomiting itself that differs from vomiting of any other cause. What differs is the architecture: acute onset, self-limiting duration under a week, close resemblance between attacks, and complete or near-complete wellness in between. The episode counts turn that observation into something checkable. The mechanistic account links the disorder to migraine — shared triggers, a shared response to migraine prophylaxis, and a high rate of personal and family migraine history — which is why Rome IV lists migraine as a supportive remark and why the treatment borrows directly from migraine practice.
Facts & figures
| Unmet criterion | What it points to |
|---|---|
| Vomiting between episodes | Chronic nausea and vomiting syndrome — a continuous rather than cyclic disorder |
| Episodes not stereotypical | Recurrent vomiting of varied cause; look for a different explanation for each attack |
| Episode counts not reached | Possibly early in the course — review rather than exclude, and keep a diary |
| Episodes lasting over a week | Falls outside the definition; reconsider obstruction, metabolic and central causes |
Failing on the counts is not the same as failing on the pattern. A patient with two identical episodes six months apart may simply be early, and should be reviewed rather than told they do not have the disorder.
| Phase | Approach |
|---|---|
| Abortive — during an attack | Triptans, anti-emetics, fluids and electrolyte correction. Works better the earlier it is given, which favours a written plan the patient carries |
| Prophylactic — between attacks | Commonly amitriptyline or topiramate; reduces episode frequency and is the half most often never started |
| Trigger management | Stress, sleep deprivation and menstruation are the usual identifiable triggers |
Because each attack presents acutely and is managed acutely, prophylaxis is frequently never initiated. That gap is the main reason the diagnosis is worth making explicitly rather than treating each episode on its own terms.
Evidence
Derivation — Rome Foundation, gastroduodenal disorders committee
2016Consensus criteria from the Rome IV gastroduodenal disorders committee, published in Gastroenterology in 2016.
Consensus-derived. Rome IV specified the episode counts explicitly — at least three in the prior year and two in the past six months — and added the allowance for milder interepisodic symptoms, which reflects that adult presentations differ from the paediatric pattern the disorder was first described in.
Management guidance — ANMS and CVSA 2019
2019Guidelines on the management of cyclic vomiting syndrome in adults from the American Neurogastroenterology and Motility Society and the Cyclic Vomiting Syndrome Association.
Set out the two-phase approach of abortive and prophylactic therapy, endorse migraine-directed prophylaxis, and recommend assessing cannabis use in every patient presenting with this pattern.
How it compares
Cyclic Vomiting Syndrome vs Cannabinoid hyperemesis syndrome
Indistinguishable on the episodes — Rome IV defines cannabinoid hyperemesis in terms of this disorder, and only cannabis exposure and response to cessation separate them.
Any patient meeting cyclic vomiting criteria should be asked about cannabis before the label is settled. The two disorders share onset, duration, frequency and character of episodes; what differs is a prolonged habitual cannabis history, the compulsive hot bathing many patients with cannabinoid hyperemesis adopt, and above all resolution on sustained abstinence. The consequence of getting it wrong is substantial, because cannabinoid hyperemesis responds to cessation and not to migraine-directed prophylaxis, while cyclic vomiting syndrome responds to prophylaxis and not to anything about cannabis.
Cyclic Vomiting Syndrome vs Chronic nausea and vomiting syndrome
Separated by pattern, not severity — cyclic vomiting has discrete episodes with well periods, while chronic nausea and vomiting syndrome is continuous.
The discriminating question is what happens between attacks. Cyclic vomiting syndrome requires the absence of vomiting between episodes, though milder symptoms such as background nausea are allowed; chronic nausea and vomiting syndrome is defined by bothersome nausea at least one day a week or weekly vomiting without an episodic structure. Both can be severe, and severity does not distinguish them. The distinction matters because the two-phase abortive-and-prophylactic approach that works in cyclic vomiting syndrome does not map onto a continuous disorder.
Cyclic Vomiting Syndrome vs Rumination syndrome
Different mechanism entirely — rumination is effortless post-prandial regurgitation without retching, not episodic vomiting.
Patients frequently describe rumination as vomiting, which is why Rome IV requires it to be excluded before diagnosing the nausea and vomiting disorders. The separating question is whether retching precedes the event: in rumination it does not, and the material is recognisable food brought up within minutes of eating rather than gastric contents expelled during an acute attack. The distinction is worth making explicitly because rumination responds to diaphragmatic breathing and to nothing in the cyclic vomiting armamentarium.
Pearls & pitfalls
- Ask about cannabis in every patient meeting these criteria. Cannabinoid hyperemesis syndrome is defined by Rome IV as resembling this disorder and cannot be separated on the episodes alone.
- Both episode counts are required — three in the prior year AND two in the past six months. Meeting one and not the other does not satisfy the criterion.
- The one-week separation matters. It prevents a single prolonged fluctuating illness being counted as multiple discrete episodes.
- Milder symptoms between cycles are permitted and are common in adults. Excluding a patient because they have background nausea misapplies the criteria.
- Vomiting between episodes is different and does exclude the diagnosis — that pattern is chronic nausea and vomiting syndrome.
- Ask about migraine, personal and family. It is a supportive remark rather than a requirement, but it predicts response to prophylaxis.
- Prophylaxis is the half of management most often never started, because each attack is handled acutely and the patient discharged. Make the plan between episodes, not during them.
- Abortive therapy works better given early, so patients benefit from a written plan they carry rather than starting again at each presentation.
- Failing the counts is not the same as failing the pattern. A patient early in their course should be reviewed with a diary rather than told they do not have the disorder.
Critical actions
- Take a detailed cannabis history — frequency, duration and recent change — before settling on this diagnosis.
- Establish that episodes are genuinely stereotypical and count them against both thresholds explicitly.
- Ask about vomiting between attacks, since its presence moves the diagnosis elsewhere.
- Ask about migraine in the patient and the family.
- Exclude structural, metabolic and central causes appropriate to the presentation, particularly at first diagnosis.
- Write an abortive plan the patient carries, so treatment starts early in the next attack rather than at presentation.
- Start prophylaxis between episodes — this is the intervention that reduces frequency and the one most often omitted.
- Identify triggers where possible: stress, sleep deprivation and menstruation are the usual ones.
- Correct fluid and electrolyte disturbance during attacks, which can be substantial.
Why this score exists
The addition of explicit episode counts is the practical heart of the Rome IV revision. Cyclic vomiting syndrome had been recognised for over a century — first in children — but its adult definition had been loose enough that it was applied to almost any recurrent vomiting, which made both diagnosis and research unreliable. Requiring three episodes in a year and two in six months, at least a week apart, converts a clinical impression into something two clinicians can check the same way. The allowance for milder symptoms between cycles was the other significant change and came from adult data: the paediatric description of complete wellness between attacks turned out not to hold in adults, many of whom have background nausea, and the earlier wording was excluding patients who plainly had the disorder.
About the creator
First author, Rome IV gastroduodenal disorders committee
Chaired the committee that produced the Rome IV gastroduodenal criteria.
Co-author; nausea and vomiting disorders
Co-authored the nausea and vomiting disorders section of the Rome IV gastroduodenal chapter.
First author, ANMS/CVSA adult management guidelines
Led the adult management guidelines that translate these criteria into the abortive and prophylactic treatment framework.
Limitations
- Entirely pattern-based with no biomarker, so the diagnosis depends on an accurate longitudinal history that is often not available at an acute presentation.
- Cannot be distinguished from cannabinoid hyperemesis syndrome on the episodes themselves, so misclassification is inevitable in cannabis users.
- The episode counts are consensus judgements rather than empirically derived thresholds, and exclude patients early in their course.
- 'Stereotypical' requires a subjective judgement of similarity between attacks, which is difficult when episodes are reported retrospectively.
- The allowance for milder interepisodic symptoms blurs the boundary with chronic nausea and vomiting syndrome.
- The episode counts were set by agreement, not fitted to data, so nothing establishes that three episodes a year is the right boundary rather than two or four.
- Says nothing about severity, and episodes range from manageable at home to requiring admission for dehydration and electrolyte disturbance.
- Provides no guidance on the extent of investigation required at first presentation, which in practice varies widely.
If you are the patient
Cyclic vomiting syndrome means repeated attacks of severe vomiting that come on suddenly, last anywhere from hours to a few days, and then stop — with periods in between where you are well or nearly well. The attacks tend to be very similar to each other each time, which is one of the main clues doctors look for. To meet the formal definition you need at least three attacks in the past year and two in the past six months, at least a week apart. Two things are worth knowing. First, your doctor should ask about cannabis use, because heavy long-term cannabis use causes an almost identical pattern, and if that is the cause then stopping is the treatment — so it is important to answer honestly, and it is asked of everyone, not because of any assumption about you. Second, there are two separate parts to treatment, and people often only get one. There is treatment for an attack, which works better the earlier it is started — so it is worth having a written plan you carry so you are not starting from scratch each time. And there is preventive treatment taken between attacks to make them less frequent, which is the part that often gets missed because each attack is dealt with as an emergency and then everyone moves on. Ask about it specifically. Many people with this condition also get migraines, or have family members who do, and the preventive medicines used are often the same.
Frequently asked questions
What are the Rome IV criteria for cyclic vomiting syndrome?#
Stereotypical episodes of vomiting as to onset (acute) and duration (less than one week); at least three discrete episodes in the prior year and two episodes in the past six months, occurring at least one week apart; and absence of vomiting between episodes, though milder symptoms may be present between cycles. Criteria must be fulfilled for three months with onset at least six months earlier.
How many episodes are needed to diagnose cyclic vomiting syndrome?#
Both counts must be met: at least three discrete episodes in the prior year AND at least two in the past six months, separated by at least a week. Meeting one and not the other does not satisfy the criterion. The one-week separation prevents a single prolonged fluctuating illness being counted as several episodes.
Can you have symptoms between attacks?#
Milder symptoms such as background nausea are explicitly permitted, and are common in adults — Rome IV added this allowance because the paediatric description of complete wellness between attacks did not hold in adult patients. Vomiting between episodes is different and does exclude the diagnosis, pointing instead to chronic nausea and vomiting syndrome.
How is cyclic vomiting syndrome different from cannabinoid hyperemesis?#
Not by the episodes. Rome IV defines cannabinoid hyperemesis syndrome as resembling cyclic vomiting syndrome in onset, duration and frequency, so they cannot be separated on the attacks themselves. The differences are prolonged habitual cannabis use, the compulsive hot bathing many patients with cannabinoid hyperemesis adopt, and resolution on sustained cessation. Every patient meeting these criteria should be asked about cannabis.
Is cyclic vomiting syndrome related to migraine?#
Closely. A personal or family history of migraine is listed in Rome IV as a supportive remark, the disorders share triggers such as stress and sleep deprivation, and the prophylactic drugs that work — amitriptyline and topiramate — come directly from migraine practice. The relationship is strong enough that asking about migraine is a routine part of the assessment and predicts response to prophylaxis.
How is cyclic vomiting syndrome treated?#
In two phases. Abortive therapy during an attack — triptans, anti-emetics, fluids — which works better the earlier it is given, so a written plan the patient carries is valuable. And prophylaxis between attacks, commonly amitriptyline or topiramate, which reduces episode frequency. The prophylactic half is the one most often never started, because each attack is managed acutely and the patient discharged.
What if a patient has had fewer episodes than the criteria require?#
Review rather than exclude. Failing the episode counts is not the same as failing the pattern — a patient with two identical stereotypical episodes may simply be early in their course. Keeping an episode diary and reassessing is more useful than recording that the criteria are not met, particularly since the diagnosis rests entirely on a pattern that takes time to declare itself.
How long do cyclic vomiting episodes last?#
Less than a week, by definition — that upper bound is part of the criteria. Episodes lasting longer fall outside the definition and should prompt reconsideration of obstruction, metabolic and central causes rather than being absorbed into the diagnosis.