Up to 80% of expectant women experience some degree of nausea and vomiting in pregnancy.1 For approximately one third of these women, nausea and vomiting in pregnancy causes a significant clinical, psychological, and economic burden because of missed work time, increased health care visits, and negative effects on family relationships.1–4 Many women with this condition report increased feelings of depression and believe it negatively affects their relationship with their partner.5 Furthermore, nearly 50% of working pregnant women believe their job efficiency was reduced and approximately 25% require time off work.6 Severe nausea and vomiting of pregnancy can result in dehydration and weight loss that require hospitalization; rare complications include Wernicke encephalopathy, splenic avulsion, esophageal rupture, pneumothorax, and acute tubular necrosis.3
Pharmacologic management has been attempted with a variety of medications with varying success. Notably, limited data exist to guide the choice of antiemetic drugs for nausea and vomiting of pregnancy. Although formal evaluation has been limited, available studies and clinical experience support the routine use of pyridoxine, metoclopramide, ondansetron, and antihistamines such as doxylamine.1–3,6–8 The combination of pyridoxine (vitamin B6) and doxylamine is recommended by the American College of Obstetricians and Gynecologists as first-line therapy for nausea and vomiting in pregnancy.9 However, despite the lack of evidence, the 5-hydroxytryptamine type 3 receptor antagonist ondansetron has become the most frequently prescribed antiemetic for the treatment of nausea and vomiting of pregnancy in the United States.10 The objective of our study was to evaluate whether ondansetron was superior to the combination of pyridoxine and doxylamine for the treatment of nausea and vomiting of pregnancy.
MATERIALS AND METHODS
We performed a double-blind, randomized, controlled trial comparing the efficacy of ondansetron with the combination of pyridoxine and doxylamine for the treatment of nausea and vomiting in pregnancy. Patients were enrolled from October 2012 through April 2013 at a tertiary care medical center that serves beneficiaries of active duty and retired military personnel. Enrollment occurred in both the emergency department and the obstetrics and gynecology clinic. This study was approved by the hospital's Institutional Review Board and was registered at clinicaltrials.gov (NCT01668069).
Eligible women were those requesting treatment for nausea with or without vomiting associated with pregnancy who were at least 18 years of age and at less than 16 weeks of gestation (by last menses or ultrasonogram). Estimated gestational age was confirmed by ultrasonogram at a later date if not available on the day of enrollment. Women were excluded from the study if their nausea or vomiting predated the pregnancy, hospitalization was required at the time of initial enrollment, they were already using anti-emetics (including metoclopramide, ondansetron, doxylamine, pyridoxine, or promethazine), they had an allergy to any of the study medications, they were unable to return for a follow-up visit in 1 week, or they were unable to obtain the medications on the day of enrollment. After written informed consent was obtained, a computer-generated program randomly allocated participants to either pyridoxine and doxylamine group or ondansetron and placebo group. Before the start of enrollment, one pharmacist prepared all medications and sealed bottles in identical numbered brown bags that corresponded to the patient number. The dispensing pharmacist, treating provider, patient, and enrolling investigator were blinded to the medication regimen. Each regimen consisted of capsules that were identical in appearance. The United States government paid for all study medications. Patients were not specifically counseled at time of enrollment or follow-up on dietary changes or nonpharmacologic methods to remedy nausea.
Before initiating treatment, each patient was asked to grade the severity of nausea and emesis experienced over the previous 7 days on two separate 100-mm visual analog scales (VAS) ranging from 0 (indicating no nausea or emesis) to 100 (indicating worst nausea or emesis imaginable). A VAS was used because it correlates well with numerical and verbal descriptors of nausea and has been used in the oncology, emergency medicine, and anesthesia literature to investigate the efficacy of anti-emetic medications.11–13 There are no well-validated tools to measure nausea or vomiting associated with pregnancy. Previous studies that span a variety of medical specialties use various criteria to define a clinically significant improvement in nausea or vomiting with regard to the VAS. We chose the most conservative of values (25 mm) to indicate a clinically significant reduction in nausea and vomiting on the VAS.
The ondansetron group was given one capsule containing ondansetron 4 mg and one placebo capsule to be taken orally every 8 hours for 5 days. The pyridoxine and doxylamine group was given one capsule containing pyridoxine 25 mg and one capsule containing doxylamine 12.5 mg to be taken orally every 8 hours for 5 days. Patients were reevaluated 5 to 7 days after initiating the drug regimen. At this time, the patients graded the severity of their nausea and emesis over the treatment period on the VAS. Patients were also asked if they experienced any adverse effects over the 5-day treatment period, including constipation, sedation, headache, an alternative previously unspecified adverse effect, or a combination of these. Age, estimated gestational age, current medications, gravidity, and parity were recorded for each patient. At follow-up, the remaining pills were collected to assess compliance.
Statistical analysis was performed using Stata 13 software. An a priori power calculation indicated that 14 patients per group (28 total) provided 90% power, at an alpha of 0.05, to detect a 25-mm difference in the mean improvement on the VAS between the groups with a standard deviation of 22 mm. To account for patient attrition, we planned to enroll 18 patients per group for a total of 36 participants. We planned to analyze data using the intention-to-treat principle, all tests were two-sided, and the level of significance was set at less than .05. Demographic characteristics and the mean difference on the VAS for nausea and emesis between each group were compared using the Wilcoxon rank-sum test. The difference in the proportion of patients who had a clinically significant improvement (25 or more VAS units) in their nausea or vomiting was assessed using the Fisher exact test. Finally, the difference in the proportion of patients who experienced side effects in each group was compared using the Fisher exact test. Five patients were lost to follow-up and the initial data for one patient were not available at the time of the final analysis. The influence of missing data on the results was assessed as follows. Missing data were estimated by multiple imputation over variables by group that had missing data. The imputation was an iterative Markov chain Monte Carlo method using 100 imputations, each of which consisted of 1,000 simulations on a multivariate Normal regression, using the variables with missing data as predictors. Descriptive statistics of variables with missing data without and with imputed values were consistent, differing only negligibly, and t-test results were far from significant. The rank-sum and Fisher exact tests were repeated with imputed data included and the results were tabulated.
Forty eligible patients were approached for enrollment and four patients declined participation. A total of 36 women were enrolled and randomized to treatment, with 30 fully completing the study (13 in the ondansetron group, 17 in the pyridoxine and doxylamine group) (Fig. 1). Five patients were lost to follow-up: one from the pyridoxine and doxylamine group and four from the ondansetron group. Additionally, the pretreatment nausea and emesis VAS were unavailable for review at the time of final analysis for one patient in the ondansetron group, so her data were excluded. Baseline characteristics for each group are reported in Table 1. No significant differences between the groups were observed with regard to demographics (gravidity, parity, or estimated gestational age) or level of nausea or vomiting before treatment. There was no difference in demographics in the patients who were lost to follow-up.
The primary outcome was the reduction in nausea on the VAS. Patients using ondansetron reported a greater reduction in nausea than those using pyridoxine and doxylamine (median 51 mm [interquartile range 37–64] compared with 20 mm [interquartile range 8–51]; P=.019) (Table 2 and Fig. 2). Additionally, those who were using ondansetron reported less vomiting on the VAS as compared with the pyridoxine and doxylamine group (median 41 [interquartile range 17–57] compared with 17 [interquartile range −4 to 38]; P=.049) (Table 2 and Fig. 2).
In the ondansetron group, 12 out of the 13 patients had a clinically significant reduction in nausea (defined as a 25-mm or greater reduction in nausea on the VAS); however, in the pyridoxine and doxylamine group, only 7 out of 17 patients had a clinically significant reduction (P=.007) (Table 3 and Fig. 3). In the ondansetron group, 10 out of the 13 patients had a reduction in emesis on the VAS; however, in the pyridoxine and doxylamine group, only 6 out of 17 patients had a reduction in emesis (P=.033) (Table 3 and Fig. 4).
This study was adequately powered to detect only differences in the primary outcome and we found no differences between the groups with respect to sedation or constipation. Four patients in the ondansetron group and seven patients in the pyridoxine and doxylamine group reported sedation (P=.707), whereas five patients in the ondansetron group and three patients in the pyridoxine and doxylamine group reported constipation (P=.412). Additionally, in the ondansetron group, one patient reported headache, one patient reported dry mouth, one patient reported pruritis, and one patient reported increased salivation. At follow-up, one patient was admitted for reasons unrelated to her nausea in pregnancy. There were no unexpected adverse events in either treatment group.
Compliance was similar between groups; both had a mean of 7 out of 30 pills remaining at follow-up (P=.906). No patient included in the final analysis returned for additional medications or treatments before follow-up.
The five patients lost to follow-up were tracked in our electronic medical record. Four patients were from the ondansetron group and one patient was from the pyridoxine and doxylamine group. Three were not prescribed any additional medication and improved on follow-up visits. One patient in ondansetron group reported improved symptoms and was given additional ondansetron 6 days after enrollment. One patient in the ondansetron group reported no change in symptoms and was given ondansetron and pyridoxine 4 days after enrollment.
We investigated the sensitivity of our results to missing data by imputing the absent data and re-running the statistical tests as described in Materials and Methods. Significance in VAS reduction decreased 1.8% in the test of difference between groups for change in nausea attributable to treatment (P increased to .037) (Fig. 5) and increased 3.9% for change in emesis (P decreased to .010) (Table 2, Fig. 6). Significance in clinical improvement decreased 0.8% on the test of difference between groups for a change in nausea attributable to treatment (P increased to .015) and decreased 1.1% for a change in emesis (P increased to .044) (Table 3). Because changes were not large and occurred in both directions, and because all tests remained significant, we concluded that the missing data did not represent a major consistent bias.
Two randomized controlled trials showed that pyridoxine alone significantly reduced the severity of nausea and vomiting in pregnancy symptoms in women with moderate or severe nausea and vomiting when compared with placebo.8,14 The combination of pyridoxine and doxylamine (an antihistamine that blocks H1 receptors), known as Bendectin in the United States, was widely prescribed to women with nausea and vomiting in pregnancy from 1958 to 1983.6 When the manufacturer voluntarily removed the drug from the market because of false allegations regarding teratogenic effects, there was a twofold to threefold increase in the rate of hospitalization because of nausea and vomiting in pregnancy.6 Importantly, the safety of the combination of doxylamine and pyridoxine since has been demonstrated in more than 200,000 first-trimester exposures.7 Meta-analyses report that use of pyridoxine and doxylamine during pregnancy is not associated with an increased incidence of birth defects.15,16 Finally, a case-control investigation from the Northern California Kaiser Permanente Birth Defects Study concluded no causal association between Bendectin use and birth defects. Both pyridoxine and doxylamine are U.S. Food and Drug Administration (FDA) Pregnancy Safety Class A, and the combination has been associated with a 70% reduction in nausea and vomiting.9 Recently, the FDA approved the combination of pyridoxine and doxylamine for the treatment of nausea and vomiting of pregnancy under the trade name Dicelgis.
Many pharmacologic therapies that are prescribed in the context of pregnancy are not specifically approved for use in a gravid population. Although ondansetron does not have FDA indication for treatment of nausea and vomiting of pregnancy, it is commonly used in practice. Nonpregnant patients have shown dramatic responses to this antiemetic, with up to 50% reduction in vomiting when compared with conventional treatments, which has led to prescribing ondansetron during pregnancy.1,17,18 A comparative study of patients with hyperemesis gravidarum suggested ondansetron and promethazine had similar efficacy; however, ondansetron was less sedating.16 Three case reports to date describe the use of ondansetron for nausea and vomiting of pregnancy because of failed conventional treatment of hyperemesis gravidarum.17–19 Although currently FDA Pregnancy Safety Class B, Pasternak et al recently reported that in a cohort of 608,385 pregnancies, receipt of ondansetron was not associated with a significantly increased risk of spontaneous abortion, stillbirth, major birth defects, preterm delivery, or delivery of a low-birth-weight neonate.11,20 Although another study using the same database reported an association between filling a prescription for ondansetron and heart defects in the offspring, these data have only been presented in abstract form and cannot be used to establish causation. The FDA recently warned about the potential for QT prolongation with the 32-mg intravenous dosing of ondansetron; however, the recommended oral dosing was not changed.21 Certainly, this must be considered before prescribing ondansetron and treatment for any patient must be individualized based on their medical history and concurrent medication profile.
The average wholesale price (base cost) of the study medications is $0.13 for 12.5 mg of doxylamine, $0.03 for 25 mg of pyridoxine, and $1.35 for 4 mg of ondansetron (generic), all in tablet form.22 Whether ondansetron would be cost-beneficial or cost-effective for the treatment of nausea and vomiting of pregnancy requires formal analysis and was beyond the scope of this study.
A limitation to this study is that there is no known standard or validated tool for quantifying nausea or vomiting in the context of pregnancy. We used the VAS that has been used in the anesthesia and emergency medicine literature, and we extrapolated its use to a pregnant population. Another possible limitation to this study is that it was conducted at a single center treating primarily military members and their beneficiaries. Military hospitals offer free care and medications, which may increase the treatment-seeking. Also, although none of the patients required hospitalization for nausea during pregnancy during the trial period, our study was not powered to detect a difference between the need for hospitalization, time out of work, or improved mood between groups.
In conclusion, our investigation showed ondansetron to be superior to the combination of pyridoxine and doxylamine in the treatment of nausea during pregnancy. Although our sample was quite small, the strength of outcome differences suggests that it might well be considered as a possible first-line therapy. Decreasing the incidence of nausea during pregnancy will likely improve the mental and physical health of patients, allowing less time away from work, resulting in less health care expenditures on urgent visits, and increasing quality of life.
1. Niebyl J. Nausea and vomiting in pregnancy. N Engl J Med 2010;363:1544–50.
2. Ebrahimi N, Maltepe C, Einarson A. Optimal management of nausea and vomiting of pregnancy. Int J Women's Health 2010;2:241–8.
3. Festin M. Nausea and vomiting in early pregnancy. Clin Evid 2009;2009:pii 1405.
4. Lombardi D, Istwan N, Rhea D, O'Brien J, Barton J. Measuring outpatient outcomes of emesis and nausea management in pregnant women. Manag Care 2004;13:48–52.
5. Mazzotta P, Stewart D, Atanackovic G, Koren G, Magee L. Psychosocial morbidity among women with nausea and vomiting of pregnancy: prevalence and association with anti-emetic therapy. J Psychosom Obstet Gynaecol 2000;21:129–36.
6. Badell M, Ramin S, Smith J. Treatment options for nausea and vomiting during pregnancy. Pharmacotherapy 2006;26:1273–87.
7. Magee L, Mazzotta P, Koren G. Evidence-based review of safety and effectiveness of pharmacologic therapy for nausea and vomiting of pregnancy (NVP). Am J Obstet Gynecol 2002;186:S256–61.
8. Matthews A, Dowswell T, Haas D, Doyle M, O'Mathúna D. Interventions for nausea and vomiting in early pregnancy. The Cochrane Database of Systematic Reviews 2010, Issue 9. Art. No.: CD007575. DOI: 10.1002/14651858.CD007575.pub2.
9. Nausea and vomiting of pregnancy. ACOG Practice Bulletin No. 52. American College of Obstetricians and Gynecologists. Obstet Gynecol 2004;103:803–16.
10. Kutcher J, Engle A, Firth J, Lamm S. Bendectin and birth defects. II: Ecological analyses. Birth Defects Res A Clin Mol Teratol 2003;67:88–97.
11. Meek R, Kelly A, Hu X. Use of the visual analog scale to rate and Monitor severity of nausea in the emergency department. Acad Emerg Med 2009;16:1304–10.
12. Tanen DA, Miller S, French T, Riffenburgh RH. Intravenous sodium valproate versus prochlorperazine for the emergency department treatment of acute migraine headaches: a prospective, randomized, double-blind trial. Ann Emerg Med 2003;41:847–53.
13. Hesketh PJ, Grunberg SM, Gralla RJ, Warr DG, Roila F, de Wit R, et al.. The oral neurokinin-1 antagonist aprepitant for the prevention of chemotherapy-induced nausea and vomiting: a multinational, randomized, double-blind, placebo-controlled trial in patients receiving high-dose cisplatin–the Aprepitant Protocol 052 Study Group. J Clin Oncol 2003;21:4112–9.
14. Vutyavanich T, Wongtra-ngan S, Ruangsri R. Pyridoxine for nausea and vomiting of pregnancy: a randomized, double-blind, placebo-controlled trial. Am J Obstet Gynecol 1995;173(3 pt 1):881–4.
15. Einarson TR, Leeder JS, Koren G. A method for meta-analysis of epidemiological studies. Drug Intell Clin Pharm 1988;22:813–24.
16. McKeigue PM, Lamm SH, Linn S, Kutcher JS. Bendectin and birth defects: I. a meta-analysis of the epidemiologic studies. Teratology 1994;50:27–37.
17. Siu S, Yip S, Cheung C, Lau T. Treatment of intractable hyperemesis gravidarum by ondansetron. Eur J Obstet Gynecol Reprod Biol 2010;105:73–4.
18. Sullivan C, Johnson C, Roach H, Martin R, Stewart D, Morrison J. A pilot study of intravenous ondansetron for hyperemesis gravidarum. Am J Obstet Gynecol 1996;174:1565–8.
19. Tincello D, Johnstone M. Treatment of hyperemesis gravidarum with the 5-HT3 antagonist ondansetron (Zofran). Postgrad Med J 1996;72:688–9.
20. Pasternak B, Svanstrom H, Hviid A. Ondansetron in pregnancy and risk of adverse fetal outcomes. N Engl J Med 2013;368:814–23.
21. U.S. Food and Drug Administration. FDA drug safety communication: new information regarding QT prolongation with ondansetron (Zofran). Available at: http://www.fda.gov/Drugs/DrugSafety/ucm310190.htm
. Retrieved May 26, 2014.
22. CardinalHealth. Ohio: pharmaceutical distribution; c2010–11. Available at: http://www.cardinalhealth.com/
. Retrieved July 9, 2013.