Find synonyms Find exact match
Disclosures: Judith A Smith, PharmD, BCOP, CPHQ, FCCP, FISOPP Grant/Research /Clinical Trial Support: Prostraken [Pharmacology/fetal toxicology studies (granisetron)]. Jerrie S Refuerzo, MD Nothing to disclose. Susan M Ramin, MD Consultant/Advisory Boards: Member of the FDA Ob/Gyn Devices Panel. Employment: Baylor College of Medicine. Charles J Lockwood, MD, MHCM Nothing to disclose. Vanessa A Barss, MD, FACOG Employee of UpToDate, Inc. Equity Ownership/Stock Options: Merck; Pfizer; Abbvie.
Contributor disclosures are reviewed for conflicts of interest by the editorial group. When found, these are addressed by vetting through a multi-level review process, and through requirements for references to be provided to support the content. Appropriately referenced content is required of all authors and must conform to UpToDate standards of evidence.
INTRODUCTION — Nausea with or without vomiting is common in early pregnancy. Severe vomiting resulting in dehydration and weight loss is termed hyperemesis gravidarum and occurs infrequently. Symptoms usually resolve by midpregnancy regardless of severity and therapy.
Management of affected women depends upon the impact of her symptoms on her health and quality of life, and the safety of maternal treatment for the fetus. Treatment approaches include dietary/lifestyle changes, oral or rectal medication, and hospitalization for parenteral fluids and therapies in women who fail to respond to outpatient management and continue to lose weight . Enteral or parenteral nutrition may be required.
The treatment of nausea and vomiting of pregnancy will be reviewed here. The clinical features and evaluation of this disorder are discussed separately. (See "Clinical features and evaluation of nausea and vomiting of pregnancy".)
GOAL OF TREATMENT — The treatment goals in patients with nausea and vomiting of pregnancy are to:
●Reduce symptoms through changes in diet/environment and by medication
●Correct consequences or complications of nausea and vomiting (eg, fluid depletion, hypokalemia, and metabolic alkalosis)
●Minimize the fetal effects of maternal nausea and vomiting and their treatment
INITIAL APPROACH — Generally, treatment begins with advice about diet, avoidance of triggers, and non-pharmacologic interventions, such as acupressure; oral or rectal medications are added if symptoms do not improve (algorithm 1).
Diet — Meals and snacks should be eaten slowly and in small amounts every one to two hours to avoid a full stomach . Women with nausea should eat before, or as soon as, they feel hungry to avoid an empty stomach, which can aggravate nausea . A snack before getting out of bed in the morning can be helpful.
Clinicians commonly recommend ingestion of frequent small carbohydrate meals, such as soda crackers or dry toast, based primarily on historical anecdotal reports . Although scientific data on the effect of dietary components on nausea are sparse, there is some evidence that protein-predominant meals/snacks produce quantifiable decreases in nausea .
Woman should figure out what foods they tolerate best and try to eat those foods. Dietary manipulations that help some women include eliminating coffee and spicy, odorous, high fat, acidic, and very sweet foods, and substituting snacks/meals that are protein dominant, salty, low fat, bland, and/or dry (eg, nuts, pretzels, crackers, cereal, toast) [2,3,6].
Fluids are better tolerated if cold, clear, and carbonated or sour (eg, ginger ale, lemonade, popsicles), and if taken in small amounts between meals . Liquids can be taken in small amounts with a straw. Small volumes of electrolyte-replacement sports drinks replace both fluids and electrolytes. Aromatic therapies involving lemon (lemonade), mint (tea), or orange have also been described as useful. Fluids should be consumed at least 30 minutes before or after solid food to minimize the effect of a full stomach .
Patients whose symptoms are related to delayed gastric emptying should improve with a diet comprised of liquids and low fat solids since these foods are more readily emptied by the stomach; however, it is not known to what degree gastric emptying and dysfunction account for symptoms in women with nausea and vomiting of pregnancy.
Drinking peppermint tea or sucking peppermint candies can reduce postprandial nausea .
Avoidance of triggers — The cornerstone of nonpharmacologic therapy of hyperemesis gravidarum is avoidance of environmental triggers . Examples of some triggers include: stuffy rooms, odors (eg, perfume, chemicals, food, smoke) , heat, humidity, noise, and visual or physical motion (eg, flickering lights, driving) . Quickly changing position and not getting enough rest, particularly after eating, may also aggravate symptoms . Lying down soon after eating and lying on the left side are additional potentially aggravating factors because these actions may delay gastric emptying . Cold solid foods are tolerated better than hot solid foods because they have less odor and require less preparation time (ie, shorter exposure to the trigger if the woman is preparing her own meal) . Brushing teeth after a meal , spitting out saliva, and frequently washing out the mouth can be helpful.
Supplements containing iron should be avoided until symptoms resolve, as iron causes gastric irritation and can provoke nausea and vomiting . Taking prenatal vitamins before bed with a snack, instead of in the morning or on an empty stomach, may also be helpful .
Acupuncture and acupressure — P6 acupressure wristbands (picture 1) do not require a prescription and have become a popular self-administered intervention . A 2014 systematic review of randomized trials did not find P6 acupuncture or acupressure wristbands significantly more effective than placebo . One reason may be that a strong placebo effect has been observed in patients who receive sham therapy [16-19]. Self-administered nerve stimulation therapy over the volar aspect of the wrist at the P6 acupressure point using a commercial device also showed some promise in a randomized, controlled trial . P6 acupuncture or acupressure has not been associated with any adverse effects on pregnancy outcome.
Hypnosis — Hypnosis has been reported to be helpful in some patients . Psychotherapy can also be a useful adjunctive therapy, particularly if psychological sources of anxiety are identified and can be ameliorated [22,23].
Pharmacologic treatment — Historically, pregnant women have been excluded from most clinical drug trials. Thus, there are limited data from pregnant women to support the safety and efficacy of agents used to treat nausea and vomiting. A number of reports have demonstrated that antiemetic drug therapy is more effective than placebo and does not increase the incidence of congenital anomalies . However, there is little evidence from well-designed comparative trials upon which to base a treatment plan for women with nausea and vomiting of pregnancy.
If non-pharmacologic interventions fail, a reasonable approach is to consider the addition of pharmacotherapy with drugs that are reported to be effective and have the best maternal-fetal safety profile. If the initial drug is ineffective, then other drugs are added in a step-wise progression (algorithm 1) [1,25,26]. We typically continue a patient on a drug for several days to determine whether her nausea and vomiting are improving. If her symptoms persist, then we add another drug to the existing regimen; however, if she is experiencing side effects, then we substitute another drug in its place.
These patients require frequent assessment of their medical status and response to therapy; women with mild to moderate nausea and vomiting of pregnancy are evaluated daily, whereas those with severe disease are evaluated from hour to hour.
Complementary and alternative medications (CAM)
Ginger — In a 2014 systematic review and meta-analysis of randomized trials, ginger improved nausea compared with placebo, but did not significantly decrease the episodes of emesis . A common dose is powdered ginger 1 to 1.5 grams orally divided over 24 hours.
We suggest use of ginger containing foods (eg, ginger lollipops, ginger tea) or ginger supplements (eg, 250 mg capsules orally four times a day) for mild nausea and vomiting. Standard pharmacologic-grade ginger preparations are not readily available ; we do not prescribe powdered ginger.
Other — Most CAM approaches have not been studied rigorously for efficacy or safety, and should be avoided for this reason. As an example, in 2009, the United States Food and Drug Administration (FDA) notified healthcare professionals and pregnant or breastfeeding women to avoid consuming Nzu, a traditional African remedy for morning sickness, because of potential health risks from high levels of lead and arsenic . Nzu may be sold under such names as Calabash clay, Calabar stone, Mabele, Argile, or La Craie.
Pyridoxine (vitamin B6) — Pyridoxine is a water-soluble B-complex vitamin that is a necessary coenzyme in the metabolism of lipids, carbohydrates, and amino acids. It can be used as a single agent  or in conjunction with doxylamine succinate for the treatment of nausea of pregnancy (see 'Doxylamine succinate and pyridoxine' below).
Systematic reviews of randomized and/or controlled trials have shown that pyridoxine (vitamin B6) improves mild to moderate nausea, but does not significantly reduce vomiting [31,32]. Thus, it is most useful for women with mild disease rather than hyperemesis. The mechanism for the therapeutic effect of pyridoxine in women with nausea is unknown. Hypotheses to explain the beneficial effects of pyridoxine include prevention/treatment of vitamin B6 deficiency, intrinsic antinausea properties, and/or synergy with the antinausea properties of antihistamines . Although vitamin B6 levels decrease as gestation advances, there is no proven correlation between maternal vitamin B6 levels and incidence or severity of nausea .
As a single agent, the initial dose of pyridoxine is 25 mg orally every six to eight hours; the maximum dose suggested for pregnant women is 200 mg/day, but cumulative doses up to 500 mg/day appear to be safe . Pyridoxine has a good safety profile with minimal side effects and is easy to obtain; therefore, it is a reasonable first-line treatment for nausea and vomiting of pregnancy, either alone or in combination with doxylamine succinate. Where available, we suggest pyridoxine-doxylamine succinate combination therapy for initial pharmacologic treatment of nausea of pregnancy (see 'Doxylamine succinate and pyridoxine' below).
Antihistamines (H1 antagonists) — Only a few antihistamines have been studied for the treatment of nausea and vomiting of pregnancy. The most commonly used antihistamine is doxylamine in combination with pyridoxine. Other antihistamines that have been used independently to treat nausea and vomiting of pregnancy include meclizine, dimenhydrinate, and diphenhydramine. There are no data on use of the scopolamine patch for nausea and vomiting of pregnancy.
The efficacy of antihistamines was illustrated in an analysis of pooled data from controlled trials that found use of these agents significantly reduced pregnancy-related nausea and vomiting (RR 0.34, 95% CI 0.27-0.43); however, these studies used a variety of antihistamines and measured different outcomes .
The safety of antihistamines (specifically H1-receptor blockers) was affirmed in a meta-analysis that examined the association between antihistamine use and major malformations . This review of 24 controlled studies, including over 200,000 first trimester exposures, found that H1-receptor blockers appeared to have a protective effect on risk of malformations (OR 0.76, 95% CI 0.60-0.94).
The primary mechanism of antihistamines in treatment of nausea and vomiting of pregnancy is direct inhibition of histamine at the histamine 1 (H1) receptor; the secondary mechanism is an indirect effect on the vestibular system by decreasing stimulation of the vomiting center. In addition, these agents inhibit the muscarinic receptor, which may mediate the emetic response.
Side effects include sedation, dry mouth, lightheadedness, and constipation.
Doxylamine succinate and pyridoxine — As discussed above, pyridoxine (vitamin B6) improves mild to moderate nausea, but does not significantly reduce vomiting [31,32]. (See 'Pyridoxine (vitamin B6)' above.)
Doxylamine succinate is an antihistamine that is usually taken with pyridoxine. The combination doxylamine-pyridoxine appears to improve efficacy and was the formulation for Bendectin. Bendectin was voluntarily withdrawn from the market in 1983 due to lawsuits alleging teratogenicity, although scientific evidence supports its safety [37-39] and efficacy [15,24,40-42]. A meta-analysis of controlled studies on outcome of pregnancies exposed to Bendectin reported no increase in the incidence of birth defects . Evidence-based reviews of controlled studies concluded doxylamine/pyridoxine is modestly effective for treatment of nausea and vomiting in pregnancy [15,24].
Formulations of doxylamine-pyridoxine similar to Bendectin are available under various names and have been proven effective in placebo-controlled randomized trials . Diclectin is available in Canada and in 2013 Diclegis was approved in the United States by the FDA for treatment of nausea and vomiting of pregnancy in women who have not responded adequately to dietary and lifestyle modifications [44,45]. We recommend these formulations, where available, for initial pharmacologic treatment of nausea of pregnancy.
Initially, two delayed release tablets (a total of doxylamine 20 mg and pyridoxine 20 mg) are taken at bedtime. The dose may be increased to four tablets per day as needed for more severe nausea (one tablet in the morning, one tablet in the afternoon, two tablets at bedtime).
In the United States, doxylamine is available in some over-the-counter sleeping pills (eg, Unisom Sleep Tabs) and as a prescription antihistamine chewable tablet (Aldex AN). One-half of the 25 mg over-the-counter tablet or two chewable 5 mg tablets can be used off-label as an antiemetic; pyridoxine 25 mg three to four times per day should be prescribed with it. The 10 mg dose of pyridoxine is not commercially available in the United States.
Women without dehydration — Additional oral medications can be initiated in women who can tolerate them. Non-oral routes of administration are useful for other women (algorithm 1).
Antihistamines (H1 antagonists) — As discussed above, only a few antihistamines have been studied for the treatment of nausea and vomiting of pregnancy (see 'Antihistamines (H1 antagonists)' above). We suggest the following drugs as first-line agents for secondary therapy because they either have fewer maternal side effects or fetal safety is more well-established than other drugs.
Treatment with doxylamine is stopped before starting a different antihistamine.
Diphenhydramine — Diphenhydramine 25 to 50 mg can be taken orally every four to six hours, as needed. It can also be given intravenously 10 to 50 mg every four to six hours, as needed.
Meclizine — Meclizine 25 mg can be taken orally every four to six hours, as needed. Meclizine has caused cleft palate in rats, but at exposures far higher than those used therapeutically. Human data of an association between facial clefts and meclizine have been mixed, but three large studies did not show an increased risk of malformations [46-48].
Dimenhydrinate — When oral medications are tolerated, dimenhydrinate 25 to 50 mg can be taken orally every four to six hours, as needed. Otherwise, 50 mg dimenhydrinate is administered intravenously over 20 minutes or 50 to 100 mg is administered rectally (where available) every four to six hours, as needed; the total dose should not exceed 400 mg/day. If the woman is also taking doxylamine, the total dose of dimenhydrinate should not exceed 200 mg/day.
Dopamine antagonists — Based on their safety profile, we consider these drugs second-line agents for secondary treatment of nausea and vomiting of pregnancy.
During nausea and vomiting, dopamine receptors in the stomach mediate the inhibition of gastric motility and, therefore, may provide a site of action for antiemetic dopamine receptor antagonists. Dopamine, specifically at the dopamine 2 receptors, is also implicated in emetic signaling through the chemoreceptor trigger zone.
Several types of dopamine antagonists can be used for the treatment of nausea and vomiting of pregnancy. The three main classes of dopamine receptor antagonists are phenothiazines (promethazine and prochlorperazine), butyrophenones (droperidol), and benzamides (metoclopramide).
Prochlorperazine and chlorpromazine — Prochlorperazine 5 to 10 mg orally, intravenously, or intramuscularly every six hours or 25 mg per rectum twice per day, as needed, appears to benefit some patients . Safety information is limited: although case reports of malformations in exposed infants have been published, larger series have not reported an increased risk of birth defects. Results from animal studies vary depending on the animal exposed .
We reserve chlorpromazine for refractory cases, given its profile of adverse side effects (see 'Refractory patients' below).
Metoclopramide — Metoclopramide 10 mg orally, intravenously, or intramuscularly (ideally 30 minutes prior to meal and at bedtime) every six to 8 hours is commonly prescribed for nausea and vomiting related to pregnancy . In randomized trials evaluating this drug in women with hyperemesis, metoclopramide 10 mg was as effective as promethazine 25 mg  and ondansetron 4 mg .
A study of intravenous metoclopramide (1.2 to 1.8 mg/hour intravenously) plus diphenhydramine (50 mg every six hours) reported vomiting improved in 36 percent of patients and was more effective than the combination of droperidol (0.5 to 1 mg/hour) plus diphenhydramine, which had been used in previous patient cohorts. In another series, combination therapy with metoclopramide and pyridoxine was superior to monotherapy with either prochlorperazine or promethazine in decreasing the number of vomiting episodes .
In large cohort studies, infants exposed to metoclopramide during the first trimester of pregnancy did not have a significant increase in risk of major congenital malformations, miscarriage, or stillbirth compared with nonexposed infants [54,55]. In the largest of these studies, over 45,000 fetuses were exposed to a median of 40 doses of metoclopramide in utero, beginning at a median of 57 gestational days .
Maternal side effects are a concern, especially with long-term use. Metoclopramide accounts for almost one third of all drug-induced movement disorders and female sex is a risk factor for development of these side effects . However, in the randomized trial discussed above, dystonia was more common with promethazine than metoclopramide (14/73 [19.2 percent] versus 4/70 [5.7 percent]) . Upon discontinuation of the drug, metoclopramide-induced tardive dyskinesia (involuntary and repetitive movements of the body) can be irreversible in some cases. Early detection and discontinuation of the metoclopramide are important for the prevention of permanent tardive dyskinesia. Use of metoclopramide with diphenhydramine or hydroxyzine may mask a dystonic reaction.
The use of subcutaneous pumps for timed release of medications, in particular metoclopramide, has been reported for outpatient management of nausea and vomiting in pregnancy with some benefit . However, the experience is limited; we do not use them.
Domperidone is another promotility agent, but there is no information on its safety or efficacy for treatment of nausea and vomiting of pregnancy.
Promethazine — Promethazine is primarily an H1 receptor blocking agent, but is also a weak dopamine antagonist. Promethazine 12.5 to 25 mg can be given orally, per rectum, or intramuscularly every four hours. Oral or rectal administration is preferred. Intraarterial, intravenous, and subcutaneous administration is contraindicated, as inadvertent intraarterial injection can result in gangrene of the affected extremity and subcutaneous injection may result in tissue necrosis. (See "Characteristics of antiemetic drugs".)
Fetal safety  and maternal efficacy in relief of both nausea and vomiting have been demonstrated in large groups of patients [24,36,52,58-60]. Disadvantages include prominent sedation and risk of dystonic reactions. These risks are elevated under conditions of prolonged use and high dosing. Use of promethazine appears to lower the seizure threshold, which may be important late in pregnancy in women with preeclampsia. There are conflicting reports regarding a potential risk of neonatal respiratory depression following the administration of promethazine during labor . Neonatal platelet aggregation also may be impaired when the drug is given intrapartum, but this does not appear to increase the need for intervention in the newborn.
Droperidol — Droperidol is used as an antiemetic in surgical and diagnostic procedures. In one study, patients with hyperemesis gravidarum treated with droperidol-diphenhydramine had significantly shorter hospitalizations (3.1 versus 3.8 days), fewer days per pregnancy hospitalized for hyperemesis (3.5 versus 4.8 days), and fewer readmissions with this diagnosis (15.0 versus 31.5 percent) than women treated with other parenteral therapies . No congenital anomalies were reported in 108 pregnancies . However, maternal side effects are a concern, especially with long-term use. Droperidol has been associated with QT prolongation and/or torsades de pointes when used in doses higher than those typically used for treatment of nausea and vomiting . A discussion of a new drug warning for droperidol can be found separately. (See "Alternatives and adjuncts to moderate procedural sedation for gastrointestinal endoscopy".)
We rarely use this drug to treat nausea and vomiting of pregnancy because of these safety concerns.
Serotonin antagonists — Ondansetron, granisetron, and dolasetron are selective antagonists at the 5-HT3 serotonin receptor. This class of drug has a favorable efficacy-safety profile in non-pregnant individuals with nausea and vomiting of various etiologies and severities.
The safety of ondansetron in pregnancy is discussed below. There are no human data on the safety of granisetron, dolasetron or other 5-HT3 antagonists in pregnancy; animal studies did not show adverse pregnancy effects .
Ondansetron — Hyperemesis is a common unlabeled indication for use of ondansetron. In a small randomized trial, use of ondansetron resulted in clinically significant reductions in both nausea and vomiting compared with the combination of doxylamine and pyridoxine .
Ondansetron 4 to 8 mg can be taken orally every eight hours, as needed, or administered intravenously by bolus injection every eight hours, as needed. A single report described use of subcutaneous ondansetron via a microinfusion pump in 521 women with severe nausea and vomiting, with improvement of symptoms to mild to moderate in 50 percent of women within three days of therapy . Based on this report and data in other populations, administration of ondansetron via a microinfusion pump appears to be a reasonable alternative route for treating severe nausea and vomiting of pregnancy.
Headache, fatigue, constipation, and drowsiness are the most common drug-related side effects. Ondansetron can cause QT prolongation, particularly in patients with underlying heart conditions, such as congenital long QT syndrome; patients with hypokalemia or hypomagnesemia; and those taking other medications that lead to QT prolongation. ECG monitoring is recommended in these patients. Serotonin syndrome is a potentially life-threatening condition associated with use of serotonergic agents and manifested by increased serotonergic activity in the central nervous system. (See "Serotonin syndrome".)
Animal data on ondansetron are reassuring as to its safety in pregnancy. Human data on safety or efficacy of ondansetron for treatment of hyperemesis are based on case reports, small series, and a historical cohort of 1849 women in Denmark exposed during pregnancy [24,65-71]. In this large record linkage study (1233 first trimester exposures), ondansetron was not associated with an increased risk of major congenital anomalies, miscarriage, low birth weight, or small for gestational age when used for treatment of nausea and vomiting of pregnancy . Although another study using the same database reported an association between filling a prescription for ondansetron during the first trimester of pregnancy and heart defects in the offspring, these data have only been presented in abstract form and cannot be adequately evaluated . Analysis of data from the National Birth Defects Prevention Study (NBDPS) showed a possible association with isolated cleft palate (adjusted OR 2.37, 95% CI 1.18–4.76), but data were limited to only 55 first trimester exposures .
Acid reducing agents — Acid reducing medications can be used as adjunctive therapy. In women with heartburn/acid reflux and nausea/vomiting of pregnancy, an observational study found that acid-reducing pharmacotherapy (eg, antacids, H2 blockers, proton pump inhibitors) combined with anti-emetic therapy resulted in significant improvement in symptoms and well-being three to four days after beginning therapy .
Antacids containing aluminum or calcium are safe and preferable to those containing bismuth or bicarbonate, which may have adverse fetal/neonatal effects .
The greatest experience with pharmacologic acid-suppressive therapy in pregnant women has been with the H2 receptor antagonists ranitidine and cimetidine, which appear to be safe during pregnancy. The efficacy of ranitidine 150 mg orally twice daily for treatment of acid reflux was demonstrated in a placebo-controlled, double-blind trial of 20 pregnant women whose heartburn was refractory to conservative measures; heartburn severity was reduced 44 percent compared to placebo .
There is less experience using proton pump inhibitors (eg, lansoprazole or esomeprazole 30 or 40 mg intravenously or orally every 24 hours) during pregnancy, but they are probably safe [67,68]. (See "Medical management of gastroesophageal reflux disease in adults", section on 'Pregnancy and lactation'.)
Women with dehydration, electrolyte abnormalities, acid-base abnormalities — Women who are unable to maintain adequate hydration, normal electrolyte levels, and acid-base balance with the initial interventions described above should be admitted to the hospital for intravenous fluids and medications. We admit women to the hospital if they have persistent vomiting after rehydration and fail outpatient management. Women who have severe vomiting, weight loss, ketonuria, dry mucous membranes, poor skin turgor, dehydration, hypotension, alkalosis from hydrochloric acid loss, hypokalemia, or nutritional deficiencies are admitted to the hospital. In women with persistent vomiting after hospitalization, it is important to exclude underlying diseases that can cause hyperemesis. (See "Clinical features and evaluation of nausea and vomiting of pregnancy", section on 'Differential diagnosis'.)
These patients and their families often need emotional support to help deal with stress and anxiety about the maternal illness and its effect on the fetus, and the disruption to their home- and work-related activities . In some cases, psychiatric consultation can be helpful to teach the patient relaxation and coping techniques and address underlying psychopathology, if present .
Fluids and nutrition — Many patients respond to intravenous hydration and a short period of gut rest, followed by reintroduction of oral intake and pharmacologic therapy.
Fluids — Dehydration occurs when fluid output exceeds fluid intake and is often associated with electrolyte abnormalities, fatigue, dizziness, and weakness. We correct dehydration with up to 2 L intravenous Ringer’s lactate infused over three to five hours, supplemented with appropriate electrolytes and vitamins. Subsequently, the infusion rate is adjusted to maintain a urine output of at least 100 mL/hour and the solution is changed to dextrose 5% in 0.45% saline. The optimum replacement fluid regimen has not been studied. It is prudent to avoid use of dextrose in the initial rehydration fluid because of the theoretical concern of Wernicke’s encephalopathy with dextrose infusion in a thiamine-deficient state . We delay dextrose infusion until after the patient has received thiamine in her initial rehydration fluid (see 'Vitamins and minerals' below). A single small study did not observe adverse effects in 102 patients who received intravenous thiamine followed by 5% dextrose 0.9% saline for persistent nausea and vomiting of pregnancy; however, only 60 percent of these women had severe disease (ie, weight loss ≥5 percent body weight, ketonuria 4+) .
Relief of symptoms is common within one to two days of rehydration . Hospitalization, as well as replenishment of fluids and electrolytes, may contribute to palliation of symptoms.
Vitamins and minerals — If the patient is experiencing persistent vomiting, it is important to replenish low levels of vitamins (especially thiamine), electrolytes, and minerals (ie, magnesium, calcium, and phosphorous) . We provide thiamine (vitamin B1) supplementation by giving 100 mg intravenously with the initial rehydration fluids and another 100 mg daily for the next two or three days. Early administration of thiamine is important to prevent a rare maternal complication, Wernicke's encephalopathy [7,81].
Each day we administer a multivitamin (MVI) intravenously: MVI (10 mL) plus 0.6 mg folic acid (to bring the folic acid total to 1 mg) in one liter and vitamin B6 25 mg in every liter. The intravenous fluid is usually dextrose 5% in 0.45% saline with 20 mEq potassium chloride given at 150 mL/hour.
Hypomagnesemia is a common cause of hypocalcemia. We first correct the low magnesium level by giving 2 grams (16 meq) magnesium sulfate infused as a 10% solution over 10 to 20 minutes, followed by 1 gram (8 meq) in 100 mL of fluid per hour. The repletion of magnesium is continued if the serum magnesium level is less than 0.8 meq/L (1 mg/dL or 0.4 mmol/L). Once serum magnesium levels are restored, we reassess the calcium level. If serum calcium is still low, we administer 1 to 2 grams calcium gluconate in 50 mL of 5% dextrose solution over 10 to 20 minutes.
Diet — A diet that attempts to minimize nausea and vomiting can be resumed after a short period of gut rest. We usually begin women on a diet consisting of bananas, rice, applesauce, and toast (BRAT diet) and then advance their diet as tolerated. Consistent protein intake is key in helping prevent nausea. Additional dietary manipulations are described above. (See 'Diet' above.)
Patients who have not eaten for several days may develop edema when resuming feeding with carbohydrates . This results from the retention of sodium during fasting combined with enhanced sodium resorption due to the actions of insulin once carbohydrates are reintroduced . No intervention is required; the edema will gradually resolve.
Oral and non-oral medication — Oral medications can be initiated in women who can tolerate them. Non-oral routes of administration are useful for other women. These medications are the same as those used to treat women without dehydration (algorithm 1). (See 'Women without dehydration' above.)
Chlorpromazine — We have found chlorpromazine (25 to 50 mg intravenously or intramuscularly or 10 to 25 mg orally every four to six hours) to be helpful in refractory cases. A rectal suppository may be available in some countries. Adverse effects include extrapyramidal reactions, orthostatic hypotension, anticholinergic effects, and altered cardiac conduction.
Glucocorticoids — Glucocorticoids have been used in women with severe and refractory hyperemesis, although the mechanism of action is not well understood [84-87]. There is a paucity of evidence that glucocorticoids are effective [24,88]. The largest placebo controlled trial included 110 women with severe hyperemesis and reported that women who received glucocorticoid therapy had a similar clinical course and need for rehospitalization as those given placebo .
Glucocorticoid use has been associated with a slightly increased risk of oral clefts when the drug is administered before 10 weeks of gestation; therefore, ideally, use of glucocorticoids should be avoided in the first trimester [90-94]. If administered after 10 weeks, the palate has formed and is not at risk for developing defects.
An effective dose is methylprednisolone (16 mg) intravenously every 8 hours for 48 to 72 hours . Methylprednisolone can be stopped abruptly if there is no response, and tapered over two weeks in women who experience relief of symptoms. After intravenous therapy, we use an oral prednisone taper regimen of 40 mg oral prednisone per day for one day, followed by 20 mg per day for three days, followed by 10 mg per day for three days, and then 5 mg per day for seven days. This regimen may be repeated up to three times over a six week period.
Glucocorticoids should be reserved for treatment of refractory nausea and vomiting of pregnancy or hyperemesis gravidarum, given the risk of maternal side effects and uncertain efficacy. (See "Major side effects of systemic glucocorticoids".)
Enteral and parenteral nutrition — Women who are refractory to all pharmacologic and nonpharmacologic interventions should be supported with enteral (tube feedings) or parenteral nutrition and intravenous fluids as long as necessary. We continue such women on pharmacologic interventions that seem to provide some relief of their nausea and vomiting.
Nutritional status and methods of alimentation (eg, tube feedings, parenteral nutrition) should be assessed in conjunction with a nutritionist or nutrition service. The optimal timing for initiating enteral or parenteral nutrition has not been established; the decision is based upon clinical judgment. In general, enteral nutrition is begun in women who cannot maintain their weight because of vomiting and despite a trial of the interventions described below. Enteral nutrition via gastric or duodenal intubation is preferable to the parenteral route and may relieve the nausea and vomiting . Adequate protein-caloric parenteral nutrition requires a central venous access device (CVAD), which may lead to catheter-related infection or thrombosis .
The American Gastroenterological Association (AGA) technical review for parenteral nutrition , as well as other AGA guidelines, can be accessed through the AGA web site at www.gastro.org/practice/medical-position-statements.
MANAGEMENT OF STABLE AND IMPROVING PATIENTS — We continue the drug regimen that has been effective until the patient has been completely asymptomatic (no nausea or vomiting) for at least a week. At that time, we discontinue the medications and see how she does. If nausea and vomiting recurs, we resume therapy. The majority of women will have resolution of nausea and vomiting by 16 to 20 weeks of gestation and will be able to discontinue their medications. Rare patients require therapy beyond 20 weeks.
OUTCOME AND PROGNOSIS
Nausea and vomiting of pregnancy — Although the maternal course can be long and tedious , nausea and vomiting of pregnancy is typically not associated with adverse pregnancy outcomes. There is good evidence that women with nausea and vomiting in early pregnancy have a lower rate of miscarriage than women without these symptoms. In one meta-analysis, the odds of miscarriage in women with nausea and vomiting in the first 20 weeks of pregnancy was OR 0.36 (95% CI 0.2-0.42) . However, the analysis did not correlate outcome with respect to the severity of the disorder. Most of the women in these studies had mild symptoms, rather than hyperemesis.
Hyperemesis — Despite the potential severity of hyperemesis gravidarum and its attendant early weight loss, most studies have reported no difference in birth weight or gestational age at birth between affected pregnancies and those unaffected by severe vomiting, as long as prepregnancy weight was normal and there was "catch-up" weight gain later in pregnancy [102,103,106]. In contrast, women with severe vomiting who require multiple hospitalizations may not have "catch up" weight gain; an adverse effect on birthweight is more likely in these women [101,107-110]. Women who have less than 7 kg weight gain are more likely to have a preterm birth/low birth weight infant [110,111]. When studies of women with hyperemesis gravidarum were pooled without regard to prepregnancy weight or catch-up weight gain, the risks of preterm delivery, low birth weight (LBW), and birth of a small for gestational age (SGA) infant were slightly but significantly increased (preterm delivery OR 1.32, 95% CI 1.04-1.68; LBW OR 1.42, 95% CI 1.27-1.58, SGA OR 1.28, 95% CI 1.02-1.60, 17.9 versus 12.7 percent) .
An association between second trimester hyperemesis gravidarum and placental dysfunction (eg, preeclampsia, abruption, SGA) has been reported . It is unclear whether there is a small increase in risk of perinatal death [100,113].
The availability of parenteral and enteral nutrition has reduced maternal morbidity, and mortality is virtually nonexistent in women who are treated. If left untreated, there have been reports of sequelae of micronutrient deficiency (eg, most commonly Wernicke encephalopathy from deficiency of vitamin B1; possibly very rare bleeding diathesis or embryopathy from vitamin K deficiency) and adverse effects of malnutrition (immunosuppression, poor wound healing, muscle wasting) [81,107,114-118]. Esophageal tears (Mallory-Weiss), esophageal rupture, pneumothoraces, pneumomediastinum, osmotic demyelination syndrome (formerly known as central pontine myelinolysis), hepatic insufficiency, and acute tubular necrosis are other rare complications in women with persistent severe vomiting [119-121].
There is, however, psychosocial morbidity, including substantial effects on ability to work outside the home, household duties, parenting activities, and social interaction [122-125]. The burdens of caregiver-time and use of health care resources also need to be considered.
Long-term outcomes — Although long-term follow-up data are limited, nausea and vomiting of pregnancy and hyperemesis do not appear to adversely affect cognitive development of offspring [126,127], but hyperemesis has been associated with reduced insulin sensitivity in prepubertal children  and development of testicular cancer before age 40 . Larger follow-up studies are needed to determine whether nausea and vomiting of pregnancy and/or hyperemesis gravidarum have long-term effects on offspring. These studies need to use well-defined criteria for the severity of the disease, and adjust for key maternal characteristics, such as prepregnancy weight and weight gain during pregnancy. Metabolic and cardiovascular outcomes should be evaluated since small for gestational age birth has been linked to chronic disease in adult life .
There are no good data on long-term maternal outcomes.
Recurrence — The disorder is likely to recur in subsequent pregnancies [110,131-133]. Two population based series reported the risk of recurrent hyperemesis in a second pregnancy was 15  and 20  percent in women with previous hyperemesis, but only 0.7 percent in women with no such history . Another study contacted women who had one pregnancy complicated by hyperemesis gravidarum and registered on an internet site sponsored by the Hyperemesis Education and Research Foundation . Of 100 respondents, 57 had become pregnant again, of whom 46 experienced recurrent severe nausea and vomiting; 37 women reported that they did not want to get pregnant a second time because of the recurrence risk of hyperemesis gravidarum.
PREVENTION — Ideally, all women of child-bearing age should be advised to take a daily multivitamin with folic acid beginning in the preconception period; this reduces the risk of congenital anomalies, particularly neural tube defects, and may help to decrease the frequency and severity of nausea and vomiting during pregnancy [134-136]. The positive effects of multivitamins are likely due to the general optimization of nutritional status and metabolism. (See "Folic acid supplementation in pregnancy".)
In addition, heartburn and acid reflux have been associated with increased severity of nausea and vomiting of pregnancy, which suggests that managing these disorders prior to pregnancy might prevent or reduce the severity of symptoms . (See "Medical management of gastroesophageal reflux disease in adults", section on 'Pregnancy and lactation'.)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, “The Basics” and “Beyond the Basics.” The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on “patient info” and the keyword(s) of interest.)
●Basics topics (see "Patient information: Morning sickness (The Basics)" and "Patient information: Taking over-the-counter medicines during pregnancy (The Basics)" and "Patient information: Hyperemesis gravidarum (The Basics)")
●Beyond the Basics topic (see "Patient information: Nausea and vomiting of pregnancy (Beyond the Basics)")
SUMMARY AND RECOMMENDATIONS — A step-wise approach to treatment of nausea and vomiting of pregnancy is provided in the algorithm (algorithm 1). The steps are based on evidence of efficacy and safety profiles. The goal is to reduce symptoms through changes in diet/environment and by medication, correct consequences or complications of nausea and vomiting, and minimize the fetal effects of maternal nausea and vomiting and its treatment.
●Women should try to become aware of, and avoid, environmental triggers and foods which might provoke their nausea and vomiting. (See 'Initial approach' above.)
●Acupuncture and acupressure have not been proven to significantly reduce nausea and vomiting. However, given the absence of harm and the strong placebo effect, some patients may benefit from a trial of acupressure wrist bands. Patient preferences should guide therapy. (See 'Acupuncture and acupressure' above.)
●Where available, we suggest pyridoxine-doxylamine succinate combination therapy for initial pharmacologic treatment of nausea of pregnancy (Grade 2C). If this drug is not available, we suggest pyridoxine, adding doxylamine succinate if pyridoxine alone is not effective. (See 'Doxylamine succinate and pyridoxine' above and 'Pyridoxine (vitamin B6)' above.)
●If nausea and vomiting persists, we suggest adding diphenhydramine 25 to 50 mg orally every four to six hours or meclizine 25 mg orally every six hours (Grade 2C). If symptoms do not improve, we suggest adding a dopamine antagonist (prochlorperazine, metoclopramide) (Grade 2C). (See 'Dopamine antagonists' above.)
●Women who are dehydrated or have electrolyte abnormalities or acid-base disturbances should receive intravenous fluids. Thiamine supplements should be added to the intravenous solution to prevent Wernicke's encephalopathy. We suggest a short period of gut rest during hydration, followed by reintroduction of oral intake with liquids and bland, low fat foods. (See 'Fluids and nutrition' above and 'Diet' above.)
●We reserve use of glucocorticoids for treatment of refractory cases after the first trimester. (See 'Glucocorticoids' above.)
●The optimal timing for initiating enteral or parenteral nutrition has not been established; the decision is based upon clinical judgment. In general, enteral nutrition is begun in women who cannot maintain their weight because of vomiting and despite a step-wise trial of pharmacologic interventions. (See 'Enteral and parenteral nutrition' above.)
●We suggest that women of child-bearing age take a multivitamin with folic acid to help prevent nausea and vomiting during pregnancy (Grade 2C), as well as for reducing the risk of neural tube defects. (See 'Prevention' above.)
All topics are updated as new information becomes available. Our peer review process typically takes one to six weeks depending on the issue.