About the Rome IV Criteria for Paediatric Cyclic Vomiting Syndrome
At least two discrete episodes of unremitting paroxysmal vomiting lasting hours to days within a six-month period; episodes stereotypical in that individual; and episodes separated by weeks to months with complete return to baseline health in between. Rome IV publishes this twice with deliberately different wording. The child and adolescent version (H1a) requires intense, unremitting nausea alongside the vomiting and adds an explicit fourth criterion that another medical condition has been excluded. The neonate and toddler version (G3) omits both — nausea because a toddler cannot report it, and the exclusion clause because it was left implicit. Return to baseline between episodes is what separates this from chronic nausea and vomiting.
Formula
H1a (child / adolescent) = ≥ 2 periods of intense unremitting nausea and paroxysmal
vomiting lasting hours to days within 6 months
AND stereotypical episodes
AND separated by weeks to months with return to baseline
AND another medical condition excluded after appropriate evaluation
G3 (neonate / toddler) = ≥ 2 periods of unremitting paroxysmal vomiting,
with or without retching, lasting hours to days within 6 months
AND stereotypical episodes
AND separated by weeks to months with return to baseline- Stereotypical
- The strongest positive feature. Families can usually describe the sequence precisely — the hour it starts, how long it lasts, what precedes it — and that precision is itself diagnostic.
- Return to baseline
- Complete wellness between episodes. This is the criterion that separates cyclic vomiting from chronic nausea and vomiting, and it is the one to ask about most carefully.
- Within a 6-month period
- Two episodes is a low bar deliberately — waiting for more delays the diagnosis, and the pattern is usually recognisable by the second identical attack.
- Rome IV publishes this disorder in both paediatric chapters with different wording; the age band changes which criteria apply.
- Nausea is required in the child and adolescent version and absent from the toddler version, because it cannot be reported reliably at that age.
- The neonate and toddler criteria allow vomiting with or without retching; the child criteria do not mention retching at all.
- Episodes often begin in the early morning or wake the child from sleep, which is characteristic but not a formal criterion.
Interpreting the result
Meeting criteria should change three things immediately. First, treat episodes early: intravenous fluids containing dextrose, antiemetics and a quiet dark environment given in the first hours are far more likely to abort an attack than the same treatment given after twelve hours of vomiting, and families should have a written plan that gets them past triage without repeating the diagnostic conversation. Second, identify and address triggers, which mirror migraine triggers and are often modifiable — irregular sleep and skipped meals are the two most commonly found. Third, consider prophylaxis where episodes are frequent or disabling, following the NASPGHAN consensus, with the agent chosen by age. In an adolescent, ask directly and specifically about cannabis: cannabinoid hyperemesis produces an almost identical pattern, is increasingly common, and resolves only with sustained cessation — compulsive hot bathing is the feature most likely to give it away. Where criteria are not met, the failing criterion directs you: episodes that are not stereotypical, or a child who is not fully well in between, both warrant a broader search including metabolic disease, intermittent obstruction such as malrotation with volvulus, raised intracranial pressure and urinary tract obstruction.
| Score | Band | What it means | Action |
|---|---|---|---|
| Criteria met — H1a (child / adolescent) | Paediatric cyclic vomiting syndrome | Four criteria including intense nausea and exclusion of other conditions | Written episode plan with early aggressive treatment; trigger identification; prophylaxis if frequent; ask about cannabis |
| Criteria met — G3 (neonate / toddler) | Paediatric cyclic vomiting syndrome | Three criteria; nausea is not required and no explicit exclusion clause is listed | Same management principles; a metabolic work-up carries more weight at this age |
| Criteria not met — no return to baseline | Not a cyclic pattern | Persistent symptoms between episodes | Consider chronic nausea and vomiting syndrome, or an ongoing organic cause |
| Criteria not met — episodes not stereotypical | Pattern does not fit | Episodes differ in character, timing or duration | Search for metabolic, neurological and surgical causes before any functional label |
What the Paediatric Cyclic Vomiting Syndrome needs (5 inputs)
- Age band — neonate/toddler (G3) or child/adolescent (H1a)
- Rome IV writes separate criteria for the two age bands. The child version requires nausea and adds an exclusion clause; the toddler version has neither, because nausea cannot be reported at that age.
- Two or more periods of unremitting paroxysmal vomiting lasting hours to days, within a 6-month period
- The child and adolescent wording adds 'intense, unremitting nausea' to this; the neonate and toddler wording allows vomiting with or without retching and says nothing about nausea.
- Episodes are stereotypical in each patient
- Same time of onset, same duration, same associated symptoms every time. Episodes that vary in character argue against the diagnosis and should prompt a broader search.
- Episodes separated by weeks to months, with return to baseline health in between
- Complete recovery between attacks is the single most discriminating feature. Persistent background nausea between episodes points to chronic nausea and vomiting instead.
- After appropriate evaluation, the symptoms cannot be attributed to another medical condition
- Stated explicitly only in the child and adolescent criteria, though the same work-up is expected in younger children.
What it returns
- Criteria met or not met
- Four criteria in the child and adolescent band, three in the neonate and toddler band.
- Which criteria set was applied
- Reported explicitly, since the two differ and the difference matters when comparing against published series.
How it is calculated
Cyclic vomiting behaves like a paroxysmal disorder rather than a gastrointestinal one, and its closest relative is migraine. The links are numerous: a family history of migraine is present in most children, many go on to develop typical headache migraine in adolescence or adulthood, the same triggers apply — sleep deprivation, fasting, excitement, infection, travel — and the same prophylactic agents are used. What actually drives an episode is not settled, with hypothalamic-pituitary-adrenal axis activation, mitochondrial dysfunction and autonomic dysregulation all implicated. Rome IV sensibly declines to pick between these and defines the disorder by its temporal architecture instead: discrete, stereotypical, self-limiting episodes with complete recovery in between. That architecture is what makes the diagnosis possible without any test, and it is also why the diagnosis cannot be made from a single episode however typical it looks.
Facts & figures
| Criterion | G3 — neonate / toddler | H1a — child / adolescent |
|---|---|---|
| Nausea | Not mentioned | Intense, unremitting nausea required |
| Retching | Explicitly 'with or without retching' | Not mentioned |
| Number of episodes | ≥ 2 within 6 months | ≥ 2 within 6 months |
| Stereotypical episodes | Required | Required |
| Return to baseline | Required | Required |
| Exclusion of other conditions | Not listed as a criterion | Listed explicitly as criterion 4 |
| Total criteria | 3 | 4 |
The differences are deliberate rather than editorial: a toddler cannot report nausea, and can retch visibly in a way an older child describes differently.
Evidence
Derivation — Rome Foundation, paediatric committees
2016Consensus criteria from the Rome IV committees on childhood functional gastrointestinal disorders, published in Gastroenterology in 2016 as two separate papers covering the neonate/toddler and child/adolescent age bands.
Consensus-derived, with the same disorder deliberately given different criteria in each age band to reflect what can and cannot be reported by the child.
Consensus statement — NASPGHAN 2008
2008North American Society for Pediatric Gastroenterology, Hepatology and Nutrition consensus statement on the diagnosis and management of cyclic vomiting syndrome.
Sets out the diagnostic approach, the red flags requiring further investigation, and prophylactic and abortive treatment recommendations stratified by age — the source most paediatric management is built on.
Adult guideline — ANMS / CVSA 2019
2019American Neurogastroenterology and Motility Society and Cyclic Vomiting Syndrome Association guideline on management in adults.
Relevant to adolescents transitioning to adult services, and the source of the current position on cannabinoid hyperemesis as a distinct entity requiring sustained cessation.
How it compares
Paediatric Cyclic Vomiting Syndrome vs Cannabinoid hyperemesis syndrome
Nearly identical presentation in adolescents — the discriminators are exposure, hot bathing, and response to sustained cessation.
Cannabinoid hyperemesis produces the same stereotypical episodic vomiting with wellness in between, and Rome IV defines it partly in terms of cyclic vomiting criteria. The differences are historical rather than clinical: regular, usually prolonged cannabis use, relief from hot bathing or showering to the point of compulsion, and resolution after sustained cessation. The last is diagnostic but slow — improvement can take weeks to months, which means a short abstinence trial that fails to help does not exclude it. This has to be asked about explicitly in every adolescent, and the questions are more productive without a parent in the room.
Paediatric Cyclic Vomiting Syndrome vs Chronic nausea and vomiting syndrome
The same symptoms in a different temporal shape — episodic with full recovery, versus continuous.
Chronic nausea and vomiting is defined by persistent symptoms rather than discrete attacks, and the two are separated by what happens between episodes rather than by how bad the episodes are. Severity does not distinguish them: a child with cyclic vomiting can be profoundly unwell for two days and entirely normal on the third. Asking 'how is she between the attacks?' is the single most useful question, and it is often not asked because the attacks dominate the consultation.
Paediatric Cyclic Vomiting Syndrome vs Abdominal migraine
Same family of paroxysmal disorders, differing in which symptom dominates — vomiting or pain.
Both are stereotypical, episodic, separated by weeks to months, and both share migraine's triggers, family history and prophylactic agents. Rome IV separates them by the dominant symptom: abdominal migraine requires pain lasting an hour or more that is the most severe and distressing feature, while cyclic vomiting is defined by the vomiting. Children can move between the two phenotypes over time, and many go on to develop typical headache migraine, so the distinction matters less for prognosis than for how an acute episode is treated.
Pearls & pitfalls
- The diagnosis lives in the pattern across episodes, not in any single one — which is why it is typically made years late.
- Complete wellness between episodes is the discriminating feature. Background nausea points to chronic nausea and vomiting instead.
- Treat early. An episode caught in the first hours often aborts; the same treatment at twelve hours usually does not.
- Give the family a written plan they can present at triage, so each attendance does not restart the diagnostic conversation.
- In any adolescent, ask about cannabis specifically. Cannabinoid hyperemesis mimics this closely and only resolves with sustained cessation.
- Compulsive hot bathing during episodes points strongly towards cannabinoid hyperemesis.
- Family history of migraine is present in most children and supports the diagnosis.
- Episodes that are not stereotypical are a red flag — look for metabolic disease, malrotation, raised intracranial pressure and urinary obstruction.
- The toddler criteria have three items and the child criteria four; using the wrong set changes the answer.
- Intravenous fluids should contain dextrose — fasting is itself a trigger and prolongs the episode.
Critical actions
- Take a detailed history of at least two episodes, establishing whether they were identical in onset, duration and symptoms.
- Ask explicitly whether the child is completely well between attacks.
- Take a family history of migraine.
- In adolescents, ask directly about cannabis use and about hot bathing behaviour during episodes.
- Exclude red flags: bilious vomiting, severe abdominal pain, neurological signs, and episodes triggered by fasting or intercurrent illness suggesting a metabolic disorder.
- Provide a written individualised episode plan covering early antiemetics and dextrose-containing fluids.
- Identify and address triggers, particularly irregular sleep and skipped meals.
- Consider prophylaxis where episodes are frequent or disabling, choosing the agent by age.
- Arrange follow-up between episodes rather than only in crisis, since that is when prophylaxis and trigger work can actually be done.
Why this score exists
Writing the same disorder twice, with different criteria, is a choice worth pausing on. Rome IV could have produced one definition with an age note, as most classifications do. It did not, because the two versions are not the same clinical exercise. A fifteen-year-old can tell you that the nausea is unbearable and continuous, and that report is diagnostically valuable enough to be made a criterion. A two-year-old cannot, and demanding it would simply mean the diagnosis is never made in toddlers — so the toddler version substitutes what can be observed, including retching, for what can only be described. The exclusion clause appears in one and not the other for a less principled reason, and the committee expects the same work-up regardless. The general lesson is that paediatric criteria are not adult criteria with the numbers lowered; they are rewritten around what the patient is able to communicate, and treating them as scaled-down versions of the adult sets misses the point of having them.
About the creator
First author, Rome IV child/adolescent functional gastrointestinal disorders committee
Chaired the committee that produced the child and adolescent criteria, including H1a.
First author, Rome IV neonate/toddler functional gastrointestinal disorders committee
Chaired the committee responsible for the G3 neonate and toddler criteria.
Lead author, NASPGHAN consensus statement on cyclic vomiting syndrome
Produced the consensus statement that most paediatric management of this disorder still follows.
Limitations
- Consensus criteria with no validation cohort, and the two-episode threshold is a pragmatic choice rather than a measured one.
- The diagnosis requires a retrospective account of at least two episodes, which delays it in practice by months to years.
- 'Stereotypical' is judged rather than defined, and depends on the quality of the history taken.
- The two age bands have different criteria for the same disorder, so published series are not directly comparable across ages.
- The neonate and toddler criteria omit the exclusion clause, which risks a functional label being applied before a metabolic work-up.
- Nothing in the criteria distinguishes cyclic vomiting from cannabinoid hyperemesis, which has to be pursued separately.
- No severity grading, so a child with two episodes a year and one with monthly episodes receive the same diagnosis.
- The criteria say nothing about when to investigate, which is left to the NASPGHAN consensus.
If you are the patient
Cyclic vomiting syndrome means having bouts of severe vomiting that last hours or days, with completely normal health in between. The bouts tend to be strikingly similar each time — starting at the same time of day, lasting about the same length, with the same warning signs — and many families can predict them closely. Between bouts the child is entirely well, which is the feature that separates this from other causes of vomiting. It is closely related to migraine. Most children with it have a family history of migraine, the same things set it off, and many later develop ordinary migraine headaches. That is also why the medicines used to prevent it are usually migraine medicines. Two practical things make the biggest difference. The first is treating an attack early: medication and a fluid drip given in the first few hours often stop an attack that would otherwise run for a day or more, so it helps to have a written plan from your specialist that you can show at the hospital rather than explaining the condition each time. The second is looking at triggers — missed meals, poor or irregular sleep, excitement and illness are the common ones, and small changes here can reduce how often attacks happen. For teenagers, doctors will ask about cannabis use. This is not an accusation; regular cannabis use can cause an almost identical illness, and it only settles if the cannabis is stopped completely for a period of weeks, so it is important to ask.
Frequently asked questions
What are the Rome IV criteria for cyclic vomiting syndrome in children?#
Two or more periods of unremitting paroxysmal vomiting lasting hours to days within six months; stereotypical episodes; and episodes separated by weeks to months with return to baseline health in between. In children and adolescents, Rome IV additionally requires intense unremitting nausea and explicit exclusion of another medical condition, giving four criteria rather than three.
Why are there two different sets of criteria for the same disorder?#
Because what a patient can report changes with age. The child and adolescent version requires nausea, which a fifteen-year-old can describe and a two-year-old cannot. The neonate and toddler version substitutes what can be observed — vomiting with or without retching — and drops the nausea requirement so that the diagnosis remains possible at that age.
What distinguishes it from chronic nausea and vomiting?#
Return to baseline health between episodes. Cyclic vomiting is episodic with complete wellness in between; chronic nausea and vomiting is continuous. Severity does not separate them — a child with cyclic vomiting can be extremely unwell during an attack and entirely normal two days later.
How is an episode best treated?#
Early and aggressively. Antiemetics, dextrose-containing intravenous fluids and a quiet dark environment given within the first hours are considerably more likely to abort an attack than the same treatment given late. A written individualised plan the family can present at triage removes the delay caused by re-explaining the diagnosis at each attendance.
Is cyclic vomiting syndrome related to migraine?#
Closely. Most affected children have a family history of migraine, the triggers overlap almost entirely, many develop typical headache migraine later in life, and prophylaxis uses the same agents. Rome IV does not classify it as a migraine variant, but clinically it behaves as one.
Why ask an adolescent about cannabis?#
Because cannabinoid hyperemesis syndrome produces an almost identical picture and is increasingly common. The clues are regular prolonged use, compulsive hot bathing or showering during episodes, and resolution only after sustained cessation. Improvement can take weeks to months, so a brief abstinence trial that fails does not rule it out. These questions are usually more productive without a parent present.
What red flags argue against the diagnosis?#
Episodes that are not stereotypical, failure to return to baseline in between, bilious vomiting, severe or localised abdominal pain, neurological signs, and episodes reliably triggered by fasting or intercurrent illness. These point towards intermittent obstruction such as malrotation with volvulus, raised intracranial pressure, urinary tract obstruction, or a metabolic disorder.
When should prophylaxis be considered?#
Where episodes are frequent or disabling enough to interfere with schooling and family life. The NASPGHAN consensus sets out agent choice by age. Prophylaxis is best discussed at a follow-up appointment between episodes rather than during an attack, when the priority is aborting the episode.
References
Original / primary references
- Hyams JS, Di Lorenzo C, Saps M, Shulman RJ, Staiano A, van Tilburg M. Childhood Functional Gastrointestinal Disorders: Child/Adolescent. Gastroenterology. 2016;150(6):1456-1468 (Rome IV, H1a).
- Benninga MA, Nurko S, Faure C, Hyman PE, St James Roberts I, Schechter NL. Childhood Functional Gastrointestinal Disorders: Neonate/Toddler. Gastroenterology. 2016;150(6):1443-1455 (Rome IV, G3).
Clinical practice guidelines
- Li BUK, Lefevre F, Chelimsky GG, Boles RG, Nelson SP, Lewis DW, et al. North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition Consensus Statement on the Diagnosis and Management of Cyclic Vomiting Syndrome. J Pediatr Gastroenterol Nutr. 2008;47(3):379-393.
- Venkatesan T, Levinthal DJ, Tarbell SE, Jaradeh SS, Hasler WL, Issenman RM, et al. Guidelines on management of cyclic vomiting syndrome in adults by the American Neurogastroenterology and Motility Society and the Cyclic Vomiting Syndrome Association. Neurogastroenterol Motil. 2019;31(Suppl 2):e13604.