Prevalence and severity of nausea and vomiting of pregnancy and effect of vitamin supplementation*

Svetlana Emelianova, MD, Paolo Mazzotta, MSc, Adrienne Einarson, RN, Gideon Koren, MD, FACCT, FRCPC
The Motherisk Program, the Hospital for Sick Children, Toronto, Ontario, Canada

* Published in Clin Invest Med 1999;22(3):106-10.

Supported in part by a grant from Duchesnay Inc., Laval, Quebec


Nausea and vomiting are by far the most common medical conditions during pregnancy, with an estimated prevalence around 50-70%.1-3 However, little information exists on the severity of nausea and vomiting and the efficacy of drug therapy for this condition. The rate of its most severe form, hyperemesis gravidarum (HG), has been estimated at 0.5-2% of all pregnancies.3 However, the relation between the severity of vomiting and weight loss has not been established.

The aetiology of nausea and vomiting of pregnancy (NVP) is not known. Studies suggesting placental hormonal effects have not been conclusive.4,5

In a randomized, blinded controlled trial of prevention of neural tube defects with folic acid, Czeizel found that women who received vitamins were less likely to experience nausea and vomiting, including less severe forms, than women who received a placebo.6 Czeizel suggested an aetiologic role of deficiency of vitamin B6, which is known to have antiemetic properties and is one of the constituents of Diclectin (doxylamine succinate/pyridoxine hydrochloride), which has been widely used for NVP.7 This observation suggests that some pregnant women with nausea and vomiting may have subtherapeutic levels of vitamin B6. Alternatively, large doses of vitamin B6 may diminish nausea and vomiting even in patients with normal levels of the vitamin.

The objectives of the present study were three-fold. We wished to prospectively ascertain the prevalence of nausea and vomiting in a population-based sample of pregnant women in Canada and to characterize the distribution of its severity. Such information is important in the development of a comprehensive management program, as well as in estimating the burden of this condition on women's well-being. Secondly, because of the lack of objective clinical tools to assess the severity of nausea and vomiting and the success of its management, we wished to correlate, in a large cohort of women experiencing nausea and vomiting, the relation between the number of daily episodes of vomiting and weight loss. Finally, we wished to verify the suggested association between periconceptional use of vitamins and the severity of nausea and vomiting.


This study was conducted through the Motherisk Program, a Toronto-based information and counselling service for women, their health care professionals and their families, on the safety and risk of exposure to drugs, chemicals, radiation, and infections during pregnancy and lactation. The Program currently counsels about 150 women daily through its main telephone line. In 1996 we added an additional line (NVP Healthline) to counsel women on the pharmacological and non-pharmacological management of nausea and vomiting of pregnancy.

We recruited three cohorts of women. To characterize the severity of nausea and vomiting in pregnant women, we recruited a random sample of 193 pregnant women between May and July of 1997 (Cohort A). These women had contacted the general telephone line of the Motherisk Program during the first trimester of pregnancy seeking advice regarding drugs, chemicals, radiation, or infections (that is, they were not seeking advice specifically related to nausea and vomiting). During the counselling session they were questioned about the presence of nausea and vomiting and their manifestations, including the mean daily number of episodes of vomiting over one week, and maximal weight loss. Other causes of nausea and vomiting were ruled out through a detailed interview and follow-up calls at 20 weeks' gestation. Patients with predisposing factors (e.g., twin pregnancy or hydatid mole) or other causes of nausea and vomiting (e.g., gastroenteritis or thyrotoxicosis) were excluded.

To characterize the relation between the mean daily number of episodes of vomiting and weight loss, all 555 women who called the NVP Healthline between May and July, 1997, specifically regarding the management of nausea and vomiting, were surveyed and followed-up prospectively (Cohort B). They were asked specifically about their maximal number of daily episodes of vomiting and maximal weight loss.

To study the association between vomiting in pregnancy and periconceptional use of vitamins, we recruited at random 301 women who called the NVP Healthline between September and December of 1997 (Cohort C). We asked them about the characteristics of nausea and vomiting, whether they had started taking vitamins in relation to pregnancy, and which preparations they used.

We calculated the prevalence and distribution of severity of nausea and vomiting among the women in Cohort A. In Cohort B, we calculated the mean weight loss (and 95% confidence interval) and correlated it with the maximal number of daily episodes of vomiting using linear regression analysis. Finally, we ascertained whether the women in Cohort C were supplemented with vitamin B6 and compared the rates of vomiting according to use of vitamins among groups or subgroups using X2 analysis.


Of the 193 women in Cohort A, 130 (67.4%) reported experiencing nausea or vomiting, or both; 87 (45.1%) experienced only nausea (Table 1). Of the 43 women (22.3% of the cohort) who experienced vomiting, 18 (9.3% of the cohort) reported weight loss.

Table 1 - Characteristics of nausea and vomiting of pregnancy


In Cohort B there was a significant correlation between the maximal number of daily episodes of vomiting and maximal weight loss (r2 = 0.25, p < 0.0001). After calculating the mean weight loss for all patients reporting a given number of daily episodes of vomiting, we found a highly significant correlation between the number of daily episodes of vomiting and mean weight loss (r2 = 0.99, p = < 0.0001). Women with one episode of vomiting per day experienced on average a 1.5 kg weight loss, and those with 10 episodes per day reported a mean weight loss of 6 kg (Figure 1). The equation best describing this relationship is: weight loss (kg) = 0.5 x (number of episodes of vomiting) +1.



There was a large variation in weight loss, and of the 425 women who reported at least some vomiting, 148 (34.8%) did not experience any weight loss. Even among the 52 women who reported 10 episodes of vomiting per day at the peak of their nausea, 11 (21.2%) did not experience any weight loss.

Of the 301 women in Cohort C, 199 were supplemented with vitamins, and 80 of them (40%) had vomiting. Of the 102 women who were not supplemented, 60 (59%) had vomiting (p = 0.002). Out of the 301 women in Cohort C, 241 were able to identify the multivitamin preparation they used; 184 used Materna, a multivitamin preparation which contains 10 mg of vitamin B6. The remaining women used 12 different preparations containing between 2 and 25 mg of vitamin B6. There was no correlation between the amount of vitamin B6 taken and the risk of vomiting, possibly because more than 85% of the women took the same dosage (10 mg per day). In a further analysis, we examined whether vitamin supplementation before week 6 of gestation (i.e., before nausea and vomiting starts in most women) was associated with lower rates of vomiting. Information on the exact time of use of multivitamins was available for 194 women. Women who experienced vomiting were less likely to have used vitamins before week 6 of gestation than those who did not experience vomiting (63.9% [76/119] versus 77.3% [58/75], p = 0.05).


The rate of nausea and vomiting in our Cohort A, 67%, corresponds to the rate in previous studies1 and lends credibility to this population-based cohort. Since these women called the Motherisk Program for advice on matters other than nausea and vomiting, they are representative of the general population of pregnant women, and thus, we avoided a referral bias (i.e., that women with severe nausea and vomiting are more likely to call). It is noteworthy that most women experienced nausea only, without weight loss, and hence were probably at low risk for dehydration or electrolyte imbalance.

It is important to identify the subgroup of women at risk for weight loss, dehydration and metabolic complications so they can receive optimal treatment. It is especially important to identify them early, before morbidity is severe and necessitates aggressive therapy or even admission to hospital.6,7

To the best of our knowledge, the correlation between the maximal daily number of vomiting episodes and weight loss has not previously been reported. We found a strong linear correlation between these two variables. The wide variation in weight loss as a function of the number of episodes of vomiting probably reflects different degrees of success of rehydration and maintenance of adequate energy intake despite vomiting, which may be dictated to a large extent by the effectiveness of antinausea therapy. It was recently reported that the withdrawal of the delayed-release form of doxylamine succinate/pyridoxine hydrochloride (Bendectin, specifically indicated for NVP) from the North American market was temporally associated with a three-fold increase in the rate of hospitalization for severe forms of nausea and vomiting among pregnant women in the United States and Canada.7 The equation that we present may serve as a simple clinical marker to estimate the severity of NVP in a particular case. For example, a typical woman who reports five episodes of vomiting per day could be expected to lose on average 3.5 kg. If she presents to her physician with a weight loss of 8 kg, it may represent failure to rehydrate or replenish energy and may prompt closer follow-up and possibly intravenous rehydration. In contrast, if she has lost only 1.5 kg, she is doing better than the expected mean.

Our study confirms the preliminary observation by Czeizel that vomiting in pregnancy is associated overall with a two-fold lower probability of vitamin supplementation in early pregnancy.6 Further analysis in our study suggests that the crucial period for multivitamin supplementation is before six weeks' gestation, corresponding to the typical time of presentation of nausea and vomiting.1,2 Further study is needed to elucidate the biological basis of this observation. It is possible that low levels of vitamin B6 cause vomiting in some women.8 However, direct measurement of vitamin B6 levels is needed to confirm this hypothesis.


  1. Ornstein M, Einarson A, Koren G. Bendectin/Diclectin for morning sickness: a Canadian follow-up of an American tragedy. Reprod Toxicol 1995;9:1-6.
  2. Weigel RM, Weigel MM. Nausea and vomiting of early pregnancy and pregnancy outcome. A meta-analytical review. Br J Obstet Gynaecol 1989;96:1312-18.
  3. Bashiri A, Neumann L, Maymon E, Katz M. Hyperemesis gravidarum: epidemiologic features, complications and outcome. Eur J Obstet Gynecol 1995;63:135-8.
  4. Kauppila A, Huhtaniemi I, Ylikorkala O. Raised serum human chorionic gonadotrophin concentrations in hyperemesis gravidarum. Br Med J 1979;1:1670-1.
  5. Soules MR, Hughes CL Jr, Garcia JA, Livengood CH, Prystowsky MR, Alexander E 3d. Nausea and vomiting of pregnancy: role of human chorionic gonadotropin and 17-hydroxyprogesterone. Obstet Gynecol 1980;55:696-700.
  6. Czeizel AE. Prevention of hyperemesis gravidarum is better than treatment. Am J Obstet Gynecol 1996;174:1667.
  7. Koren G, Pastuszak A, Ito S. Drugs in pregnancy. N Engl J Med 1998;338:1128-37.
  8. Sahakian V, Rouse D, Sipes S, Rose N, Niebyl J. Vitamin B6 is effective therapy for nausea and vomiting of pregnancy: a randomized, double-blind placebo-controlled study. Obstet Gynecol 1991;78:33-6.

Back to home