There is limited evidence for the efficacy of doxylamine-pyridoxine for NVP. A 2010 Cochrane review¹⁶x16.Matthews, A., Dowswell, T., Haas, D.M., and Doyle, M. O’Mathuna DP Interventions for nausea and vomiting in early pregnancy. Cochrane Database Syst Rev. 2010; 9: CD007575

PubMed | Google ScholarSee all References concluded that two older studies^17.x17.Geiger, C.J., Fahrenbach, D.M., and Healey, F.J. Bendectin in the treatment of nausea and vomiting in pregnancy. Obstet Gynecol. 1959; 14: 688–690

PubMed | Google ScholarSee all References^{, 18.}x18.McGuinness, B.W. and Binns, D.T. ‘Debendox⁷ in pregnancy sickness. J R Coll Gen Pract. 1971; 21: 500–503

PubMed | Google ScholarSee all References provided insufficient evidence supporting the efficacy of Bendectin over placebo. A 1959 double-blind study of Bendectin (n = 53) versus placebo (n = 57) found a treatment effect or symptom improvement in 50 patients given Bendectin (94%) versus 37 patients given placebo (65%) (P < 0. 001). ¹⁷x17.Geiger, C.J., Fahrenbach, D.M., and Healey, F.J. Bendectin in the treatment of nausea and vomiting in pregnancy. Obstet Gynecol. 1959; 14: 688–690

PubMed | Google ScholarSee all References The method for determining clinical response was not described. A 1971 double-blind randomized control trial of Debendox (n = 41) versus placebo (n = 40) employed a five-point symptom scale.¹⁸x18.McGuinness, B.W. and Binns, D.T. ‘Debendox⁷ in pregnancy sickness. J R Coll Gen Pract. 1971; 21: 500–503

PubMed | Google ScholarSee all References The fraction of patients who were “better” did not significantly differ between the Debendox (71%) and placebo (55%) groups after 14 days (P = 0.17). Despite this negative finding, the results were reported as positive because the response rate in the Debendox group was “significantly better than the theoretical 50% improvement that would be expected from purely random fluctuations in severity.”¹⁸x18.McGuinness, B.W. and Binns, D.T. ‘Debendox⁷ in pregnancy sickness. J R Coll Gen Pract. 1971; 21: 500–503

PubMed | Google ScholarSee all References

In a double-blind randomized trial published after the Cochrane review¹⁶x16.Matthews, A., Dowswell, T., Haas, D.M., and Doyle, M. O’Mathuna DP Interventions for nausea and vomiting in early pregnancy. Cochrane Database Syst Rev. 2010; 9: CD007575

PubMed | Google ScholarSee all References and Canadian guidelines,^3.x3.Arsenault, M.Y., Lane, C.A., and SOGC Clinical Practice Obstetrics Committee. The management of nausea and vomiting of pregnancy. SOGC Clinical Practice Guideline no. 120, October 2002. J Obstet Gynaecol Can. 2002; 24: 817–831 (quiz 832–3)

Abstract | Full Text PDF | PubMed | Google ScholarSee all References^{, 4.}x4.Einarson, A., Maltepe, C., Boskovic, R., and Koren, G. Treatment of nausea and vomiting in pregnancy: an updated algorithm. Can Fam Physician. 2007; 53: 2109–2111

PubMed | Google ScholarSee all References^{, 8.}x8.Levichek, Z., Atanackovic, G., Oepkes, D., Maltepe, C., Einarson, A., Magee, L. et al. Nausea and vomiting of pregnancy. Evidence-based treatment algorithm. Can Fam Physician. 2002; 48: 267–277

PubMed | Google ScholarSee all References 131 women were given Diclectin and 125 a placebo for 14 days.¹⁹x19.Koren, G., Clark, S., Hankins, G.D., Caritis, S.N., Miodovnik, M., Umans, J.G. et al. Effectiveness of delayed-release doxylamine and pyridoxine for nausea and vomiting of pregnancy: a randomized placebo controlled trial. Am J Obstet Gynecol. 2010; 203: 571.e1–571.e7

Abstract | Full Text | Full Text PDF | PubMed | Scopus (52) | Google ScholarSee all References Symptoms were assessed using daily ratings on the 15-point pregnancy unique quantification of emesis (PUQE) scale. A score of 3 represented no symptoms and 15 represented the most severe state of symptoms including greater than six hours of nausea, at least seven instances of retching and at least seven episodes of vomiting daily. The ANCOVA of the symptoms scores yielded a significant difference between Diclectin and control patients (P = 0.006). The difference in mean symptom score changes between women given Diclectin (−4.8 ± 2.7) and placebo (−3.9 ± 2.6) was 0.9 on the 15-point PUQE symptom scale.

Of these three studies, two showed statistically significant differences between patients receiving doxylamine-pyridoxine and placebos. In all three studies, the response magnitude in the placebo groups was three to five times larger than the difference between the active and placebo groups. The two older studies employed formulations containing dicyclomine in addition to doxylamine-pyridoxine, so no studies of the efficacy of doxylamine-pyridoxine were available at the time the Canadian guidelines were published.

The evidence supporting the efficacy of pyridoxine for NVP is similar in amount to that supporting doxylamine-pyridoxine.^20.x20.Vutyavanich, T., Wongtra-ngan, S., and Ruangsri, R. Pyridoxine for nausea and vomiting of pregnancy: a randomized, double-blind, placebo controlled trial. Am J Obstet Gynecol. 1995; 173: 881–884

Abstract | Full Text PDF | PubMed | Scopus (95) | Google ScholarSee all References^{, 21.}x21.Sahakian, V., Rouse, D., Sipes, S., Rose, N., and Niebyl, J. Vitamin B6 is effective therapy for nausea and vomiting of pregnancy: a randomized, doubleblind placebo-controlled study. Obstet Gynecol. 1991; 78: 33–36

PubMed | Google ScholarSee all References In a double-blind randomized trial,²⁰x20.Vutyavanich, T., Wongtra-ngan, S., and Ruangsri, R. Pyridoxine for nausea and vomiting of pregnancy: a randomized, double-blind, placebo controlled trial. Am J Obstet Gynecol. 1995; 173: 881–884

Abstract | Full Text PDF | PubMed | Scopus (95) | Google ScholarSee all References 169 women were given pyridoxine and 169 a placebo. The reduction in reported severity of nausea, as measured on a 10-cm visual analogue scale on five consecutive days, was significantly (P < 0. 001) greater in the pyridoxine group (2.9 ± 2.2 cm), compared with the placebo group (2.0 ± 2.7 cm). There was also a reduction of vomiting episodes as a secondary outcome, but this was not statistically significant (P = 0.055). In another double-blind randomized trial,²¹x21.Sahakian, V., Rouse, D., Sipes, S., Rose, N., and Niebyl, J. Vitamin B6 is effective therapy for nausea and vomiting of pregnancy: a randomized, doubleblind placebo-controlled study. Obstet Gynecol. 1991; 78: 33–36

PubMed | Google ScholarSee all References 31 women were given pyridoxine and 28 a placebo. In this three-day study, the reduction in reported nausea (2.9 ± 2.4) in the pyridoxine group was not significantly greater than in the placebo group (1.9 ± 2.0) on a 10-cm visual analogue scale, but there was a significant (P < 0.01) reduction in the proportion of women who reported vomiting after taking pyridoxine versus placebo. A recent Cochrane review cautioned against over-interpretation of these two small positive studies.¹⁶x16.Matthews, A., Dowswell, T., Haas, D.M., and Doyle, M. O’Mathuna DP Interventions for nausea and vomiting in early pregnancy. Cochrane Database Syst Rev. 2010; 9: CD007575

PubMed | Google ScholarSee all References

Since the effect sizes of pyridoxine and doxylamine-pyridoxine in RCTs are similar—and both are much smaller than the response to placebos—the utility of doxylamine must be questioned. The clinical significance of these effects (0.9 on a 15-point scale and 0.9 cm on a 10 cm visual analogue scale) is also debatable.

There are no high quality studies comparing the efficacy of the individual components of doxylamine-pyridoxine and/or the combination. One study compared the effects on NVP of Debendox plus an extra 10 mg of pyridoxine (for a total of 20 mg) to 10 mg of pyridoxine alone in a crossed-over fashion,²²x22.Wheatley, D. Treatment of pregnancy sickness. Br J Obstet Gynaecol. 1977; 84: 444–447

Crossref | PubMed | Scopus (20) | Google ScholarSee all References making it difficult to determine whether doxylamine or the higher dose of pyridoxine was responsible for the significantly greater symptom reduction in the former group. An unpublished 1970s RCT conducted by Merrell Dow (Bendectin’s manufacturer) seemed to show that doxylamine-pyridoxine had greater efficacy than pyridoxine alone.²³x23.Bendectin Peer Group. Bio/Basics International Peer Group Report. Merrell Dow Pharmaceuticals Inc, ; 1975: 559–637 (March 14)

Google ScholarSee all References The fraction of patients who experienced an excellent or moderate response was significantly different (P = 0.01) between pyridoxine (66%, 126/191) and doxylamine-pyridoxine (77%, 165/213). However, this study had limited internal validity because 31% of patients (709/2308) dropped out of the six-day trial or were excluded from the analysis. It is unclear how physicians classified the response to treatment (excellent, moderate, slight, or none). The results of the completed study were never published.

Three meta-analyses of safety data for Diclectin have concluded that its use during pregnancy is not associated with an overall increase in malformations.^24.x24.Einarson, T.R., Leeder, J.S., and Koren, G. A method for meta-analysis of epidemiological studies. Drug Intell Clin Pharm. 1988; 22: 813–824

PubMed | Google ScholarSee all References^{, 25.}x25.McKeigue, P.M., Lamm, S.H., Linn, S., and Kutcher, J.S. Bendectin and birth defects: I. A meta-analysis of the epidemiologic studies. Teratology. 1994; 50: 27–37

Crossref | PubMed | Scopus (96) | Google ScholarSee all References^{, 26.}x26.Seto, A., Einarson, T., and Koren, G. Pregnancy outcome following first trimester exposure to antihistamines: meta-analysis. Am J Perinatol. 1997; 14: 119–124

Crossref | PubMed | Scopus (139) | Google ScholarSee all References Two of these meta-analyses each included more than 100 000 pregnancies and reported summary odds ratios for overall malformation rates near unity.^24.x24.Einarson, T.R., Leeder, J.S., and Koren, G. A method for meta-analysis of epidemiological studies. Drug Intell Clin Pharm. 1988; 22: 813–824

PubMed | Google ScholarSee all References^{, 25.}x25.McKeigue, P.M., Lamm, S.H., Linn, S., and Kutcher, J.S. Bendectin and birth defects: I. A meta-analysis of the epidemiologic studies. Teratology. 1994; 50: 27–37

Crossref | PubMed | Scopus (96) | Google ScholarSee all References The third meta-analysis²⁶x26.Seto, A., Einarson, T., and Koren, G. Pregnancy outcome following first trimester exposure to antihistamines: meta-analysis. Am J Perinatol. 1997; 14: 119–124

Crossref | PubMed | Scopus (139) | Google ScholarSee all References reportedly included more than 200 000 pregnancies and indicated a summary odds ratio significantly below unity (indicating that doxylamine-pyridoxine use is associated with a lower risk of malformations), but these conclusions have been questioned.²⁷x27.Chin, J.W., Gregor, S., and Persaud, N. Re-analysis of safety data supporting doxylamine use for nausea and vomiting of pregnancy. Am J Perinatol. 2013; (Dec 9. [Epub ahead of print])

PubMed | Google ScholarSee all References A recent re-analysis of the data in the third meta-analysis showed that the number of women included was lower than reported and that the odds ratio for malformation was higher than reported.²⁷x27.Chin, J.W., Gregor, S., and Persaud, N. Re-analysis of safety data supporting doxylamine use for nausea and vomiting of pregnancy. Am J Perinatol. 2013; (Dec 9. [Epub ahead of print])

PubMed | Google ScholarSee all References Following the publication of this article, the Public Health Agency of Canada took down its statement about the safety of Diclectin.

One potential limitation of the included studies is that some women had taken other teratogenic medications such as tetracycline.²⁸x28.Heinonen, O.P., Slone, D., and Shapiro, S. Birth defects and drugs in pregnancy. PSG Publishing, Littleton MA; 1977

Google ScholarSee all References Exposure to other medications was not reported in all studies. Some older studies employed questionable methodologies. In the General Practitioner Research Group trial (1963),²⁹x29.General Practitioner Research Group. General Practitioner Clinical Trials: drugs in pregnancy survey. Practitioner. 1963; 191: 775–780

PubMed | Google ScholarSee all References physicians were invited to provide information about patients seen during a nine-month period but they were also given the option of including patients seen during the previous six months. The number of patients recruited in each period is not reported so the effect of any selection bias in the retrospective group is unknown. In a study reported in 1963 by Bunde and Bowles,³⁰x30.Bunde, C.A. and Bowles, D.M. A technique for controlled survey of case records. Curr Ther Res Clin Exp. 1963; 5: 245–248

PubMed | Google ScholarSee all References 21 physicians who frequently prescribed Bendectin were asked by the manufacturer to complete questionnaires about recent patients. This study’s odds ratio for Bendectin exposure and birth defects, 0.52 (95% CI 0.25 to 1.1), is substantially lower than those in other studies.

There are conflicting findings regarding an association between doxylamine-pyridoxine exposure and pyloric stenosis. In a cohort study of 13 346 children, 3835 of whom were exposed prenatally to doxylamine-pyridoxine, a pyloric stenosis risk ratio of 2.5 (95% CI 1.2 to 5.2) was found, and there was a relative risk of 7.59 (95% CI 4.95 to 11.64; P = 0.01) when more than five prescriptions were filled.³¹x31.Aselton, P., Jick, H., Chentow, S.J., Perera, D.R., Hunter, J.R., and Rothman, K.J. Pyloric stenosis and maternal Bendectin exposure. Am J Epidemiol. 1984; 120: 251–256

Crossref | PubMed | Scopus (27) | Google ScholarSee all References A case–control study of 1427 cases and 3001 control subjects found the odds of exposure to doxylamine- pyridoxine for any malformation were not significantly increased (OR 1.4; 95% CI 0.95 to 2.0), but subanalyses identified a statistically significant association for pyloric stenosis (OR 4.3; 95% CI 1.75 to 10.75).³²x32.Eskenazi, B. and Bracken, M.B. Bendectin (Debendox) as a risk factor for pyloric stenosis. Am J Obstet Gynecol. 1982; 144: 919–924

Abstract | Full Text PDF | PubMed | Scopus (45) | Google ScholarSee all References Another case–control study that used two control groups consisting of different malformations found no increased exposure to Bendectin in 325 cases with pyloric stenosis (OR 0.9; 95% CI 0.6 to 1.2 and OR 1.0; 95% CI 0.7 to 1.4).³³x33.Mitchell, A.A., Schwingl, P.J., Rosenberg, L. et al. Birth defects in relation to Bendectin use in pregnancy. II. Pyloric stenosis. Am J Obstet Gynecol. 1983; 147: 737–742

Abstract | Full Text PDF | PubMed | Scopus (35) | Google ScholarSee all References A meta-analysis of five studies found no increased risk (OR 1.04; 95% CI 0.85 to 1.3).²⁵x25.McKeigue, P.M., Lamm, S.H., Linn, S., and Kutcher, J.S. Bendectin and birth defects: I. A meta-analysis of the epidemiologic studies. Teratology. 1994; 50: 27–37

Crossref | PubMed | Scopus (96) | Google ScholarSee all References Subanalyses of two large cohort studies revealed (in the first) a relative risk for pyloric stenosis of 3.3 (95% CI 1.6 to 7.2) with seven cases in the exposed and 88 in the unexposed groups²⁷x27.Chin, J.W., Gregor, S., and Persaud, N. Re-analysis of safety data supporting doxylamine use for nausea and vomiting of pregnancy. Am J Perinatol. 2013; (Dec 9. [Epub ahead of print])

PubMed | Google ScholarSee all References and (in the second) a relative risk of 1.7 (95% CI 0.21 to 14) with one case in the exposed and six in the control groups.³⁴x34.Shiono, P.H. and Klebanoff, M.A. Bendectin and human congenital malformations. Teratology. 1989; 40: 151–155 (Erratum in: Teratology 1990;41:250–1)

Crossref | PubMed | Scopus (14) | Google ScholarSee all References Another large cohort study did not report any cases of pyloric stenosis in either group.³⁵x35.Fleming, D.M., Knox, J.D., and Crombie, D.L. Debendox in early pregnancy and fetal malformation. Br Med J (Clin Res Ed). 1981; 283: 99–101

Crossref | PubMed | Scopus (30) | Google ScholarSee all References

A paired case–control study of 204 mothers of children with acute non-lymphoblastic leukemia found a non-significant odds ratio of 1.75 (95% CI 0.98 to 3.20) for use of Bendectin, but there was a significant dose–response relationship.³⁶x36.Robison, L.L., Buckley, J.D., Daigle, A.E., Wells, R., Benjamin, D., Arthur, D.C. et al. Maternal drug use and risk of childhood nonlymphoblastic leukemia among offspring. An epidemiologic investigation implicating marijuana (a report from the Childrens Cancer Study Group). Cancer. 1989; 63: 1904–1911

PubMed | Google ScholarSee all References Another case—control study of 555 patients with childhood malignancies (including solid tumours) and 1110 control subjects found an odds ratio of 0.99 for prenatal exposure to Bendectin/Debendox.³⁷x37.McKinney, P.A., Cartwright, R.A., Stiller, C.A., Hopton, P.A., Mann, J.R., Birch, J.M. et al. Inter-regional epidemiological study of childhood cancer (IRESCC): childhood cancer and the consumption of Debendox and related drugs in pregnancy. Br J Cancer. 1985; 52: 923–929

Crossref | PubMed | Scopus (14) | Google ScholarSee all References A subanalysis in that study showed an odds ratio of 10.04 (95% CI 1.76 to 57.37) of malignancy if Debendox was taken for one to two months, but there was no significant relationship for longer durations of use. The Diclectin product monograph mentions the negative finding in this latter study.

If evidence for the safety of the combination of doxylamine and pyridoxine is also evidence for the safety of pyridoxine alone, there is not greater evidence for the safety of doxylamine-pyridoxine than for pyridoxine alone (as is suggested by Canadian guidelines).^4.x4.Einarson, A., Maltepe, C., Boskovic, R., and Koren, G. Treatment of nausea and vomiting in pregnancy: an updated algorithm. Can Fam Physician. 2007; 53: 2109–2111

PubMed | Google ScholarSee all References^{, 8.}x8.Levichek, Z., Atanackovic, G., Oepkes, D., Maltepe, C., Einarson, A., Magee, L. et al. Nausea and vomiting of pregnancy. Evidence-based treatment algorithm. Can Fam Physician. 2002; 48: 267–277

PubMed | Google ScholarSee all References In addition, there is separate evidence that pyridoxine monotherapy is not associated with adverse outcomes. A case–control study of 458 children with malformations and 911 control subjects found an odds ratio for pyridoxine exposure of 0.95 (95% CI 0.57 to 1.6).³⁸x38.Nelson, M.M. and Forfar, J.O. Associations between drugs administered during pregnancy and congenital abnormalities of the fetus. Br Med J. 1971; 1: 523–527

Crossref | PubMed | Scopus (254) | Google ScholarSee all References