Evidence of the effectiveness and adverse effects of antiemetic treatments for hyperemesis gravidarum
| Drug or therapy | Dose* | Effectiveness | Safety in pregnancy | Other comments |
|---|---|---|---|---|
| Doxylamine– pyridoxine† | 10 mg doxylamine + 10 mg pyridoxine every 6 h | More effective than placebo in reducing nausea and vomiting scores. However, the clinical importance of this effect is uncertain because of its modest magnitude. (40)– (44) | Doxylamine has been extensively studied in more than 60 000 pregnancies. The studies do not reveal a higher risk of congenital abnormalities or specific abnormalities. (21) | Antihistamines can provide an additional anticholinergic benefit to those distressed by ptyalism. (45) |
| Cyclizine†‡ | 50 mg every 8 h | Has not been shown to be more effective than placebo or other treatments. (22), (37) | Almost 3500 pregnancies have been studied with cyclizine. No evidence indicates that this drug increases the risk of birth defects. (46) | Antihistamines can provide an additional anticholinergic benefit to those distressed by ptyalism. (45) |
| Promethazine† | 25 mg every 8 h | More effective than placebo in relief of nausea and vomiting. (21) | Promethazine has been extensively studied in more than 25 000 pregnancies. The studies do not reveal a higher risk of congenital abnormalities or specific abnormalities. (21) | Promethazine is a strong sedative. When used just before delivery, sedation and respiratory depression of the newborn is a theoretical risk. (21) |
| Dimenhydrinate† | 25–50 mg every 6 h | More effective than placebo in relief of nausea and vomiting. (21) | Dimenhydrinate is commonly used and safety data are generally reassuring that it is not a teratogen. (21) | If the patient is taking doxylamine, the total dose of dimenhydrinate should not exceed 200 mg/d to reduce the risk of adverse effects, or doxylamine should be stopped. |
| Metoclopramide | 10 mg every 8 h | More effective than placebo in reducing the intensity of nausea and vomiting. (21), (47) | Metoclopramide has been extensively studied in more than 90 000 pregnancies. (48) Only 1 study (49) (n = 958 pregnancies) reported a small increase of genital organ defects; all other studies showed no increase in prevalence of congenital abnormalities. Metoclopramide may therefore be safely prescribed. | Metoclopramide can give tardive dyskinesia, but this is uncommon in young patients. Early detection and discontinuation of metoclopramide is important. If metoclopramide is prescribed, clinicians should warn for this adverse effect. (21) |
| Ondansetron | 4 mg every 8 h | In a small RCT, use of ondansetron resulted in clinically significant reductions in both nausea and vomiting compared with the combination of doxylamine and pyridoxine. (50) In another RCT, it was more effective than metoclopramide for reducing vomiting but not nausea. (51) | Because the results of the studies are contradictory, whether ondansetron is associated with a slightly increased risk of cleft palate and heart defects, or whether the severity of the indication plays a role, is unclear. (21), (52), (53) | Ondansetron should only be used in the first trimester if other antiemetics are insufficiently effective. Clinicians should weigh the risk of untreated severe hyperemesis gravidarum and small increased risk of congenital abnormalities, in consultation with the patient. |
| Steroids | IV hydrocortisone 100 mg every 12 h or methylprednisolone 125 mg; switch from IV to oral as soon as clinical improvement occurs: oral prednisolone 40 mg (day 1), 20 mg (day 2–3), 10 mg (day 4–6), 5 mg (day 7–14) | More effective than placebo, the rate of hospital readmission is lower. (21), (37) | No higher risk of congenital abnormalities has been clearly shown; therefore, corticosteroids can be used during pregnancy if necessary. Preference is given to prednisolone and hydrocortisone as these corticosteroids are least likely to reach the unborn child; these should be used for as short a time and dosed as low as possible. | Long-term use of high doses of corticosteroids can slow the growth of the unborn child and inhibit the child’s adrenal cortex. As a result, the child may have low blood glucose levels, low blood pressure, and electrolyte disturbances in the first days after birth. |