International consensus on standards for studying the efficacy of pharmacological therapies for nausea and vomiting of pregnancy

Mati Berkovitch, Gerald G. Briggs, Robert K.H. Chin, Peter von Dadelszen, Neil Deuchar, Roger Gadsby, T. Murphy Goodwin, Bengt Källén, Gideon Koren, Steven H. Lamm, Catherine MacKinnon, Laura A. Magee, C. Ineke Neutel, Aurel Schofield, Anthony R. Scialli, Christopher A. Sullivan, Forrest D. Tierson, Ronald M. Weigel


Nausea and vomiting of pregnancy (NVP) afflicts 80% of pregnant women, typically during the morning hours of the first trimester of pregnancy, but often throughout the day and beyond the first trimester. In its severe forms, NVP may affect the health and well-being of the mother and her unborn child, however, at any level of severity, NVP may affect the quality of life of the woman and her family.

Due to concerns regarding fetal safety, there have been relatively few studies on the efficacy of antiemetics for NVP. The most widely tested drug, Bendectin (doxylamine succinate plus pyridoxine), was voluntarily removed from the American market by its manufacturer in 1983 despite being safe, due to prohibitive legal costs secondary to numerous litigations, all of which were eventually rejected at higher courts.

On October 30-31, 1998, an expert group of clinicians and scientists met in Toronto to review the state of knowledge and identify questions that need to be answered regarding NVP through clinical research.

A systematic review of studies examining pharmacological and non-pharmacological treatment modalities has identified issues in interpretation and comparison of these trials. The objective of this consensus statement is to foster a logical, evidence-based approach to such studies, so their results can be compared, and if needed, even combined.

Quantifying the Endpoints

A major issue in interpreting and comparing studies on the management of NVP has been the use of numerous different methods to qualify and quantify nausea, vomiting and related symptoms. Different researchers have arbitrarily defined NVP as "mild", "moderate" or "severe" without clear quantitative definitions or biological logic.

Often the assessments have been retrospective. In some cases end points were measured dichotomously as success/failure or as percentage preferring one drug over the other (in crossover studies). While these approaches may be reasonable in randomised control trials (RCTs), assuming that all arms of the studies are evaluated similarly, the clinical interpretation of arbitrary endpoints and comparison between studies becomes difficult, if not impossible.

We feel that future studies should use a standard, validated tool that will allow comparison among studies.

The Rhodes Index appears to be an optimal tool for this purpose. It separately scores, as categorical variables, the number of vomiting episodes per day, the size of the vomiting, the degree and length of nausea and retching, as well as the distress associated with the condition (Table 1). The inventory can be done once or twice a day and, in addition to an overall score, one can report separately on the frequency and changes in nausea, vomiting, retching and stress.



Whenever possible, other outcome measures, such as the need for intravenous rehydration (in or out of the hospital), hospitalization, loss of time from work, should also be collected. More efforts need to be invested to measure the quality of life of women with NVP and the ability of drugs to modify it.

Issues of Fetal Safety

A major issue in performing drug trials for treatment of NVP is the need to expose the fetus to medications during embryogenesis. Without strong evidence of fetal safety, it is difficult to justify such studies. The Consensus Group believes there are several new pieces of evidence that can help in making rational decisions in this difficult situation.

  1. As a group, H1-blockers have been shown by several meta-analyses, based on hundreds of thousands of patients, not to increase the general rate of major malformations.1-4 In the case of doxylamine (as part of Bendectin/Diclectin), the risk was shown not to increase above baseline even for specific groups of malformations.
  2. There is now evidence that NVP is associated with a protective effect against major malformations in general,4 and specifically for cardiovascular malformations.5
  3. In women suffering from NVP beyond eight weeks of gestation, the potential teratogenic risks may be nullified for specific malformations which can no longer be produced at that stage of fetal development (e.g., neural tube defects, cleft palate).
  4. To date, the many national registries and Teratogen Information Services worldwide collect a large number of reports of drug exposure during the first trimester of pregnancy by women who were not aware of their pregnancies. These sources can, and should, be utilized to generate safety data on antiemetics. For example, it is unreasonable that metoclopramide, which is the drug of choice for NVP in many European countries, has not had a single cohort safety study nor an efficacy trial.2, 6, 7

The Use of Placebo

Because there is a well described placebo effect on nausea and vomiting, similar to non-pregnant states, it is important to compare drugs against placebo.

Researchers must ensure clear cut and unambiguous measures of treatment failure that lead to discontinuation of a placebo/treatment arm, because of the potential for dehydration and weight loss in patients with NVP.

In countries where a labelled effective drug for NVP is available, it is reasonable to compare a new drug to the existing one, thus preventing women from being deprived from an active therapy. Alternatively, the placebo arm may include fewer patients than the "active" arm, e.g., in a ratio of 1:2 or 1:3.


  1. Lamm S. The epidemiological assessment of the safety and efficacy of Bendectin. The 1st International Conference on Nausea and Vomiting of Pregnancy. October 1998.
  2. Briggs GG. Droperidol-diphenhydramine for hyperemesis gravidarum. The 1st International Conference on Nausea and Vomiting of Pregnancy. October 1998.
  3. Magee LA. The safety and effectiveness of anti-emetic therapy for NVP. The 1st International Conference on Nausea and Vomiting of Pregnancy. October 1998.
  4. Källén B. Hyperemesis gravidarum during pregnancy and delivery outcome: A registry study. The 1st International Conference on Nausea and Vomiting of Pregnancy. October 1998.
  5. Boneva RS, Moore CA, Botto L, Wong L-Y, Erickson JD. Nausea of pregnancy, antinausea preparations and congenital heart defects: a population-based case-control study. The 1st International Conference on Nausea and Vomiting of Pregnancy. October 1998.
  6. Berkovitch M. The safety of metoclopramide during pregnancy. The 1st International Conference on Nausea and Vomiting of Pregnancy. October 1998.
  7. Einarson A, Koren G, Bergman U. Treatment of nausea and vomiting in pregnancy: A comparative European study. The 1st International Conference on Nausea and Vomiting of Pregnancy. October 1998. [poster]

Back to home