Health · Explainer

Hyperemesis gravidarum and why some pregnancies bring severe vomiting

The label hyperemesis gravidarum, usually shortened to HG, covers the most extreme end of pregnancy sickness, and work combining human genetics with mouse experiments now gives a putative causal role to GDF15, a hormone made mostly by the foetus and placenta, with some women unusually sensitive to it (Fejzo et al., 2024; Jansen et al., 2021). A 2024 review counts between 0.3% and 3% of pregnant women as affected (Vinnars et al., 2024). Evidence on consequences for mother and child is largely observational, so it shows links rather than proof (Veenendaal et al., 2011; Fiaschi et al., 2018).

What follows explains the current definition, how symptoms and their intensity are scored and where HG departs from routine queasiness. The later sections cover hormone and gene research on its origins and the outcomes recorded for women and newborns, drawing on an international consensus, family and population genetics, several meta-analyses and a hospital cohort of over eight million English pregnancies.

Wooden bedside table holding a glass of water, a plate of plain crackers and a folded blanket beside a bed
A bedside table with a glass of water, crackers and a folded blanket Editorial illustration

Hyperemesis gravidarum explained through its modern definition

HG means sickness intense enough to cause weight loss, dehydration, upset body salts and missing nutrients (Nana et al., 2026). Researchers place it at the far end of a range whose gentler forms most expectant mothers know (Fejzo et al., 2019). No other cause sends as many women to hospital early on (Vinnars et al., 2024). American data rank it second among reasons for admission across the whole of gestation (Fejzo et al., 2018).

The Windsor consensus criteria

Agreement on a definition took a long time; a 2019 review still reported none (Fejzo et al., 2019). In 2021, scientists, clinicians and women who had lived through HG settled on criteria through a two-round online survey and an in-person meeting in Windsor, England (Jansen et al., 2021). Four elements make up the result: onset before week 16, nausea or vomiting with at least one severe, being unable to eat or drink as usual, and a heavy toll on everyday life (Jansen et al., 2021). Dehydration counts as supporting, not required (Jansen et al., 2021).

Earlier research leaned on different markers, among them losing over 5% of pre-pregnancy weight and ketones in urine (Fejzo et al., 2018). A 2014 meta-analysis showed that urinary ketones told doctors nothing reliable about whether HG was present or how bad it was, even though most studies had picked cases by them (Niemeijer et al., 2014).

Estimates of frequency

Counts differ. A 2024 review settles on 0.3% to 3% of pregnant women (Vinnars et al., 2024). A recurrence review gives roughly 1% of pregnancies (Dean et al., 2020). National registries have ranged from 0.3% in Sweden up to 10.8% in China (Zhang et al., 2011).

Origin can matter too. A Norwegian registry study, cited by a family study, found HG in 2.2% of women of Pakistani origin and 1.9% of women of Turkish origin, against 0.9% among Norwegian women (Zhang et al., 2011).

Hyperemesis gravidarum symptoms and how severity is assessed

Unrelenting queasiness, frequent vomiting and failure to keep down food or drink form the core picture (Vinnars et al., 2024). Within days or weeks the body runs short of water, sheds weight and loses its electrolyte balance (Vinnars et al., 2024). A 2026 review warns that women can become so ill in body and mind that the risk of ending a wanted pregnancy and of suicidal thinking rises (Nana et al., 2026).

Symptoms beyond ordinary nausea

Daily function sets HG apart. Eating and drinking become a struggle and routine tasks grind to a halt (Jansen et al., 2021). The older clinical portrait adds intractable vomiting with dehydration, weight loss, nutritional deficiencies and electrolyte disturbances (Fejzo et al., 2018).

Severity scores used in studies

Many studies rate intensity with the PUQE score, introduced in 2002 to replace a longer 8-item questionnaire (Koren et al., 2002). It asks only three things: daily vomiting episodes, hours of nausea per day and retching episodes (Koren et al., 2002). Scores agreed closely with the long version, at a correlation of 0.904 (Koren et al., 2002). Its purpose is to grade intensity in pregnancy sickness generally (Koren et al., 2002).

The three PUQE questions capture symptoms over a defined period, but they do not directly measure nutritional intake, the disruption of daily activities or a person's ability to drink normally. This creates an important distinction between grading nausea and vomiting and deciding whether a case meets the Windsor definition of hyperemesis gravidarum.

Jansen et al. (2021) included inability to eat or drink normally and a major limitation of everyday activities among its defining features. Koren et al. (2002), by contrast, tested agreement between a short symptom questionnaire and a longer one. A close correlation between the scores does not make them a complete assessment of the condition.

When fluid and weight losses mount

Prolonged losses carry real danger. Weeks of vomiting can rapidly exhaust stores of thiamine, also called vitamin B1 (Oudman et al., 2019). That shortage can trigger Wernicke's encephalopathy, a disorder of the brain (Oudman et al., 2019; Vinnars et al., 2024). Among 177 published cases, vomiting had lasted a median of 7 weeks before it appeared (Oudman et al., 2019).

Warning signs in those reports included double vision in 37.4% and blurred vision in 27.4%, with sickness present in every case (Oudman et al., 2019). The figures describe 177 published cases of the brain disorder, not women with HG overall (Oudman et al., 2019).

Hyperemesis gravidarum vs morning sickness: where the line falls

Routine queasiness is the norm. Estimates reach up to 70% of expectant mothers, and up to 85% at some stage of gestation (Fejzo et al., 2019; Vinnars et al., 2024). HG is far less frequent and is generally seen in research as the extreme of that same condition (Fejzo et al., 2019).

Degree and fallout separate them. Milder queasiness is unpleasant, while HG can bring heavy physical and psychological illness and badly erode quality of life (Dean et al., 2020).

Without shared criteria until recently, earlier studies drew the boundary loosely (Vinnars et al., 2024). The Windsor criteria now set it where normal eating and drinking stop and daily life becomes heavily restricted (Jansen et al., 2021). Other problems of early gestation are discussed in an article on ectopic pregnancy warning signs.

What causes hyperemesis gravidarum: hormones and genes

Its origin long puzzled researchers. In 2018, one team argued that clustering in families and twin data made genetics worth pursuing (Fejzo et al., 2018). A 2019 review suggested a placental contribution, citing variants in genes for placental proteins and hormone receptors (Fejzo et al., 2019).

Pregnancy hormones studied earlier

Suspicion first fell on the hormones of gestation. A 2014 meta-analysis of 81 articles found chorionic gonadotropin, thyroid hormones, oestradiol, progesterone and leptin all linked to HG inconsistently (Niemeijer et al., 2014). Before the latest GDF15 work, a 2020 review said no biomarker could diagnose HG or forecast its severity (Dean et al., 2020). Hormonal shifts at a different life stage are described in an article on perimenopause and hormone changes.

GDF15 and sensitivity before pregnancy

A 2018 genome scan, whose discovery cases were customers of a genetic testing company reporting intravenous fluids in an online survey, tied HG to two genes, GDF15 and IGFBP7, which play parts in placental growth and appetite (Fejzo et al., 2018). Its authors made clear that the results did not prove cause (Fejzo et al., 2018).

Placental cells produce GDF15 at their highest rate, and maternal blood levels climb through the first two trimesters (Fejzo et al., 2018). A 2024 paper that also reported mouse experiments notes from earlier work that the hormone's receptor exists only in the hindbrain, and activating it brings on nausea and vomiting (Fejzo et al., 2024).

Drawing on human data and work in mice, that study traced most GDF15 in a mother's blood to the foetus and placenta (Fejzo et al., 2024). Low GDF15 outside pregnancy raised the risk (Fejzo et al., 2024). Genetically, each standard deviation of higher pre-pregnancy GDF15 carried an odds ratio of 0.70 for HG (Fejzo et al., 2024). An interviewer-run survey at a single Sri Lankan hospital found that about 5% of women with beta-thalassaemia, whose GDF15 stays high, had any queasiness or vomiting, compared with greater than 60% of controls (Fejzo et al., 2024).

The team proposes a putative causal role for foetal GDF15, with the mother's sensitivity shaping severity (Fejzo et al., 2024). Some evidence comes from mice, and certain family results were called hypothesis generating (Fejzo et al., 2024). It is not a diagnostic test (Fejzo et al., 2024). The first author is paid to advise two biotechnology firms and sits on the board of an HG charity, and a therapy patent drawing on part of the work is pending (Fejzo et al., 2024).

Family history and genetic risk

HG clusters among relatives. In a survey recruited largely through a patient charity, sisters of affected women had an odds ratio of 17.3 (Zhang et al., 2011). A third of the cases, 33%, named an affected mother, against 7.7% of controls (Zhang et al., 2011).

All family details were self-reported, and the authors accept that a shared environment may play a part (Zhang et al., 2011). Earlier reports connect multiple pregnancies, trophoblastic disease and a female foetus with higher incidence (Zhang et al., 2011).

Effects on mother and baby reported in studies

Most outcome data are observational. An English study of 8,211,850 pregnancies contrasted women whose HG led to a hospital stay with women who had no such stay, with adjustment for other maternal and pregnancy factors (Fiaschi et al., 2018). Its figures show associations and cover only admitted cases (Fiaschi et al., 2018).

Hospital admission identifies a specific group of affected women, not everyone experiencing hyperemesis gravidarum. The English cohort used hospital records, so its estimates are most directly applicable to pregnancies with HG severe enough to result in admission.

Fiaschi et al. (2018) adjusted their comparisons for measured maternal and pregnancy characteristics, but statistical adjustment cannot guarantee that all unmeasured differences were removed. This matters when interpreting higher odds of pre-eclampsia, thrombosis or low birthweight. The reported associations describe events in the studied population; they do not establish that HG caused each outcome or predict what will happen in an individual pregnancy.

Complications for the mother

Admitted women showed raised odds of anaemia, at 1.28, pre-eclampsia, at 1.16, and eclampsia, at 1.84 (Fiaschi et al., 2018). Clots before delivery were also likelier, with odds ratios of 1.94 for deep vein thrombosis and 2.54 for pulmonary embolism (Fiaschi et al., 2018).

A meta-analysis of case-control studies found far higher depression and anxiety scale scores among women with HG, although studies differed greatly (Mitchell-Jones et al., 2017). Which came first cannot be told from that design. Death of the mother is uncommon, though grave complications such as Wernicke's encephalopathy do occur (Vinnars et al., 2024).

Outcomes for the baby

A 2011 meta-analysis tied HG to more daughters, lower birthweight, babies small for their gestational age and earlier delivery (Veenendaal et al., 2011). Not every finding points one way. It saw no link with Apgar scores, birth defects or perinatal death (Veenendaal et al., 2011). In the English cohort, stillbirth odds were lower, at 0.77, and absolute risks of the adverse outcomes stayed small (Fiaschi et al., 2018).

Outcome

Odds ratio

Source

Female child

1.27

Meta-analysis, 2011

Low birthweight

1.42

Meta-analysis, 2011

Small for gestational age

1.28

Meta-analysis, 2011

Preterm birth

1.32

Meta-analysis, 2011

Low birthweight

1.12

English cohort, 2018

Preterm birth

1.11

English cohort, 2018

Stillbirth

0.77

English cohort, 2018

Note: the odds ratios compare pregnancies with and without the condition in observational studies, so they show associations, not effects proven to result from it (Veenendaal et al., 2011; Fiaschi et al., 2018). Absolute risks in the large English cohort were small, and long-term effects on children are little studied (Fiaschi et al., 2018; Veenendaal et al., 2011).

Settled findings and open questions on severe pregnancy vomiting

Several conclusions now look firm. HG has an agreed international definition, it gathers in families, and genetic studies connect its risk with GDF15 (Jansen et al., 2021; Zhang et al., 2011; Fejzo et al., 2018). Mass spectrometry of maternal blood showed most GDF15 there comes from the foetus and placenta, while the same study's mouse experiments looked at desensitisation (Fejzo et al., 2024).

This severe form of pregnancy sickness has an agreed definition, and human genetic and mouse studies point to foetal GDF15 and the mother's sensitivity to it, although proof of cause and treatment evidence remain limited (Fejzo et al., 2024; Vinnars et al., 2024).

Much is still unsettled. Assessing 25 trials, a 2024 review rated the evidence behind each treatment as low or very low in certainty, while cautioning that this does not show current treatments are useless (Vinnars et al., 2024). Reported return rates span 15% to 81% across low-quality studies, with no pooled estimate (Dean et al., 2020). A 2011 meta-analysis found little known about long-term effects on children, and a 2024 review points to some evidence of later sleep problems and anxiety (Veenendaal et al., 2011; Vinnars et al., 2024).

Frequently asked questions

Does hyperemesis gravidarum come back in later pregnancies?

It can, but estimates vary widely. A 2020 systematic review found five studies reporting recurrence in 15% to 81% of women, with no pooled figure because the studies differed so much (Dean et al., 2020). Registry data may undercount cases, while the highest figure came from a small, self-selected online group (Dean et al., 2020).

When does hyperemesis gravidarum usually begin?

By the international definition agreed in 2021, symptoms start early in pregnancy, before 16 weeks (Jansen et al., 2021). In a review of 25 randomised treatment trials, the mean stage of pregnancy at which women joined ranged from 8 to 11 weeks (Vinnars et al., 2024).

Is hyperemesis gravidarum more common with twins?

Studies suggest so. A family study listed multiple pregnancies, along with trophoblastic disease, foetal chromosomal abnormalities and a female foetus, among factors reported with an increased incidence (Zhang et al., 2011). These are associations from observational reports, and the size of the effect for twins is not given in that study (Zhang et al., 2011).

Is hyperemesis gravidarum linked to the baby's sex?

Yes, an association has been reported. A 2011 meta-analysis found that women with the condition were more likely to have a girl, with an odds ratio of 1.27 (Veenendaal et al., 2011). This describes a statistical link in observational studies; it does not mean the condition can predict a baby's sex in any individual pregnancy (Veenendaal et al., 2011).

Does hyperemesis gravidarum affect mental health?

Studies report a strong association. A meta-analysis found much higher depression and anxiety scale scores in women with the condition than in controls, although results varied widely (Mitchell-Jones et al., 2017). A 2026 review adds that severe illness can raise the risk of suicidal thoughts and of ending a wanted pregnancy (Nana et al., 2026).

Sources

  1. Fejzo MS, Sazonova OV, Sathirapongsasuti JF, Hallgrímsdóttir IB, Vacic V, MacGibbon KW, et al. Placenta and appetite genes GDF15 and IGFBP7 are associated with hyperemesis gravidarum. Nature Communications. 2018;9:1178. DOI 10.1038/s41467-018-03258-0.
  2. Fejzo M, Rocha N, Cimino I, Lockhart SM, Petry CJ, Kay RG, et al. GDF15 linked to maternal risk of nausea and vomiting during pregnancy. Nature. 2024;625(7996):760–767. DOI 10.1038/s41586-023-06921-9.
  3. Nana M, Painter R, Williamson C, Nelson-Piercy C. Hyperemesis gravidarum. The Lancet. 2026;407(10523):78–89. DOI 10.1016/S0140-6736(25)01454-0.
  4. Jansen LAW, Koot MH, van't Hooft J, Dean CR, Bossuyt PMM, Ganzevoort W, et al. The Windsor definition for hyperemesis gravidarum: a multistakeholder international consensus definition. European Journal of Obstetrics & Gynecology and Reproductive Biology. 2021;266:15–22. DOI 10.1016/j.ejogrb.2021.09.004.
  5. Fejzo MS, Trovik J, Grooten IJ, Sridharan K, Roseboom TJ, Vikanes Å, Painter RC, Mullin PM. Nausea and vomiting of pregnancy and hyperemesis gravidarum. Nature Reviews Disease Primers. 2019;5(1):62. DOI 10.1038/s41572-019-0110-3.
  6. Koren G, Boskovic R, Hard M, Maltepe C, Navioz Y, Einarson A. Motherisk-PUQE (pregnancy-unique quantification of emesis and nausea) scoring system for nausea and vomiting of pregnancy. American Journal of Obstetrics and Gynecology. 2002;186(5 Suppl):S228–S231. DOI 10.1067/mob.2002.123054.
  7. Zhang Y, Cantor RM, MacGibbon K, Romero R, Goodwin TM, Mullin PM, Fejzo MS. Familial aggregation of hyperemesis gravidarum. American Journal of Obstetrics and Gynecology. 2011;204(3):230.e1–230.e7. DOI 10.1016/j.ajog.2010.09.018.
  8. Dean CR, Bruin CM, O'Hara ME, Roseboom TJ, Leeflang MM, Spijker R, Painter RC. The chance of recurrence of hyperemesis gravidarum: a systematic review. European Journal of Obstetrics & Gynecology and Reproductive Biology: X. 2020;5:100105. DOI 10.1016/j.eurox.2019.100105.
  9. Fiaschi L, Nelson-Piercy C, Gibson J, Szatkowski L, Tata LJ. Adverse maternal and birth outcomes in women admitted to hospital for hyperemesis gravidarum: a population-based cohort study. Paediatric and Perinatal Epidemiology. 2018;32(1):40–51. DOI 10.1111/ppe.12416.
  10. Veenendaal MVE, van Abeelen AFM, Painter RC, van der Post JAM, Roseboom TJ. Consequences of hyperemesis gravidarum for offspring: a systematic review and meta-analysis. BJOG. 2011;118(11):1302–1313. DOI 10.1111/j.1471-0528.2011.03023.x.
  11. Niemeijer MN, Grooten IJ, Vos N, Bais JMJ, van der Post JA, Mol BW, et al. Diagnostic markers for hyperemesis gravidarum: a systematic review and metaanalysis. American Journal of Obstetrics and Gynecology. 2014;211(2):150.e1–150.e15. DOI 10.1016/j.ajog.2014.02.012.
  12. Oudman E, Wijnia JW, Oey M, van Dam M, Painter RC, Postma A. Wernicke's encephalopathy in hyperemesis gravidarum: a systematic review. European Journal of Obstetrics & Gynecology and Reproductive Biology. 2019;236:84–93. DOI 10.1016/j.ejogrb.2019.03.006.
  13. Mitchell-Jones N, Gallos I, Farren J, Tobias A, Bottomley C, Bourne T. Psychological morbidity associated with hyperemesis gravidarum: a systematic review and meta-analysis. BJOG. 2017;124(1):20–30. DOI 10.1111/1471-0528.14180.
  14. Vinnars MT, Forslund M, Claesson IM, Hedman A, Peira N, Olofsson H, Wernersson E, Ulfsdottir H. Treatments for hyperemesis gravidarum: a systematic review. Acta Obstetricia et Gynecologica Scandinavica. 2024;103(1):13–29. DOI 10.1111/aogs.14706.
Ellen Marsh

Ellen Marsh — author

Ellen writes Health & Medicine explainers for Cabdirect, with a focus on public health, disease prevention and the interpretation of medical research for general readers. Her academic background combines biomedical sciences with postgraduate study in populatio...

This article is for general information and is not medical advice. It was reviewed for accuracy by a qualified clinician; decisions about your health should be made with your own doctor.

Related