Clinical Opinion
ajog.org
OBSTETRICS
Treating morning sickness in the United States—changes in prescribing are needed Gideon Koren, MD, FRCPC, FACMT, FAACT
Presently, 97.7% of prescriptions for the treatment of nausea and vomiting in pregnancy in the United States are with medications not labeled for use in pregnancy, not indicated for nausea and vomiting in pregnancy, and not classified as safe in pregnancy by the Food and Drug Administration. The use of ondansetron for nausea and vomiting in pregnancy has increased from 50,000 monthly prescriptions in 2008 to 110,000 at the end of 2013, despite unresolved issues regarding fetal safety and Food and Drug Administration warnings about serious dysrhythmias. In April 2013, the Food and Drug Administration approved the combination of doxylamine and pyridoxine, specifically for nausea and vomiting in pregnancy symptoms. Now that a safe and effective drug is available in the United States, there is no reason for women to be exposed to a drug of unproven maternal and fetal safety. Key words: nausea and vomiting, ondansetron, pregnancy
N
ausea and vomiting of pregnancy (NVP) is the most common medical condition in pregnancy, affecting an estimated 80% of pregnant women.1 Because the withdrawal of Bendectin (delayed release doxylamine 10 mgpyridoxine 10 mg; WMS Merrell Dow, Cincinnati, OH) from the US market in 1983, American women did not have access to a Food and Drug Administration (FDA)-approved drug for NVP for 30 years, until the same drug combination (under the trade name Diclegis; Duchesnay Inc., Blainville, Canada) was approved by the FDA in April 2013.2 The withdrawal of Bendectin from the US left American women without an FDA-approved drug for NVP and was associated with a 3-fold increased risk From The Motherisk Program, The Hospital for Sick Children and University of Toronto, Toronto, ON. Received June 17, 2014; revised Aug. 15, 2014; accepted Aug. 18, 2014. Gideon Koren has served as a paid consultant for Duchesnay Inc. Blainville, Quebec, manufacturer of Diclegis. Duchesnay Inc. has supported research studies by the Motherisk Program in Toronto. 0002-9378/$36.00 ª 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ajog.2014.08.017
of hospitalization of women for the severe forms of this condition.3 Presently, 97.7% of prescriptions for the treatment of NVP in the US are with medications not labeled for use in pregnancy, not indicated for NVP, and not classified as safe in pregnancy (FDA category A). The use of ondansetron for the treatment of NVP has steadily increased from 50,000 prescriptions per month in 2008 to 110,000 at the end of 2013 (Figure). This means that around 1 million pregnant American women are exposed to ondansetron out of 4 million pregnancies a year. Ondansetron (GlaxoSmithKlein Inc, Philadelphia, PA) is a serotonin 5-HT3 receptor antagonist, originally introduced to prevent nausea and vomiting induced by cancer chemotherapy, radiation therapy, and surgery. The fact that ondansetron became generic in 2007, and hence its price dropped, might have been an important cause for this increase, with easier access to Medicaid and health maintenance organizations. Prescribing ondansetron as a first line option is not consistent with American Professors in Gynecology and Obstetrics and American College of Obstetricians and Gynecologists evidence-based recommendations for the management of NVP.4,5 It should be remembered that most drugs used in pregnancy, including steroids
602 American Journal of Obstetrics & Gynecology DECEMBER 2014
for the prevention of respiratory distress syndrome, all tocolytic agents, and magnesium sulfate for the prevention of cerebral palsy, to mention a few, have not been approved by the FDA. Yet, they are standard of care. In contrast, in the case of ondansetron there are unresolved issues surrounding the fetal and maternal safety, including recent warnings by the FDA on its potential to cause serious dysrhythmias.6
Fetal safety concerns regarding ondansteron The fetal safety of the ondansetron was first investigated in humans by Einarson et al7 in 2004 through a prospective controlled cohort study of 176 women, in whom we could not detect an increased teratogenic risk. However, this sample size had the statistical power to rule out only a 5-fold increased risk of major malformations, and not any specific malformation. In February 2013, Pasternak et al8 reported that ondansetron was not associated with increased malformation rates when used for morning sickness. This was based on retrospective analysis of data from the Danish Birth Registry, collected between 2004 and 2011 and linked to the National Prescription Register. Each of the 1970 women exposed to ondansetron was matched to 4 unexposed controls. Of note, the mean age at exposure was 10 gestational weeks, which means that half of the cases were exposed to ondansetron at later than 10 gestational weeks, when the morphologic malformations sought in this study could not be produced any more. This can cause a bias toward the null, diluting an existing risk because of inclusion of cases that were not exposed during embryogenesis. However, in August of 2013, Andersen et al9 from Denmark presented a second study using the same Danish registries covering more years (1997-2010) and more pregnant women (897,018 vs
Obstetrics
ajog.org 608, 835). In contrast to Pasternak et al,8 Andersen’s study detected a 2-fold increased risk of cardiac malformations with ondansetron (odds ratio [OR], 2.0; 95% confidence interval [CI], 1.3e3.1), leading to an overall 30% increased risk of major congenital malformations. To rule out confounding by indication, Andersen et al9 also examined metoclopramide taken for morning sickness, detecting no increase in teratogenic risk. The fact that the same large registry can be investigated to yield such opposing results is concerning. There is an exponential rise in use of prescription database linkage to birth registries. None of these were designed specifically to address fetal drug safety, and there may be flaws in the quality and completeness of the available data. Of potential importance, a recent large case control study by the Sloan epidemiology unit and the Centers of Disease Control and Prevention, has reported a 2-fold increased risk for cleft palate associated with ondansetron taken for NVP in the first trimester of pregnancy (OR, 2.37; 95% CI, 1.28e4.76).10
Maternal safety concerns regarding ondansetron The maternal safety of ondansetron has been challenged in June 2012, when the FDA issued a warning of possible serious cardiac output (QT) prolongation and Torsade the Pointe among people receiving ondansetron.11 As a result, the FDA requires strict workup of patients receiving ondansetron, to rule out long QT, electrolyte imbalance, congestive heart failure or taking concomitant medications that prolong the QT interval.12 Because this drug is not approved by the FDA for pregnant women, the FDA did not specifically address precautions in pregnancy. However, in the context of NVP, women with severe NVP often exhibit electrolyte abnormalities (hypokalenia or hypomagnesemia). Presently, counseling of women who receive ondansetron for morning sickness suggests that these FDA precautions are not being followed. Serotonin syndrome is a lifethreatening disorder of excessive serotonergic activity, typically occurring when 2 or more serotonin-modifying agents are used simultaneously, although it may
Clinical Opinion
FIGURE
Sales of antiemetics for NVP in the US, 2008-2014
IMS National Prescription Audit from 2008 to 2014. Dramamine; McNeil Consumer Healthccare, Washington, PA. Ondansetron; GlaxoSmithKlein Inc, Philadelphia, PA. Diclegis; Duchesnay Inc., Blainville, Canada. IMS, IMS Health Inc, Danbury, CT; NVP, nausea and vomiting of pregnancy. Koren. Nausea and vomiting in pregnancy. Am J Obstet Gynecol 2014.
also occur with a single agent.12 From Jan. 1, 1998, to Dec. 30, 2002, Health Canada received 53 reports of suspected serotonin syndrome, most often reported with the use of selective serotonin reuptake inhibitors (SSRIs), monoamine oxidase inhibitors and selective serotoninnorepinephrine reuptake inhibitors. The clinical presentation of seretonergic syndrome is characterized by the triad of cognitive or behavioral changes (confusion, agitation, lethargy, coma), autonomic instability (hyperthermia, tachycardia, diaphoresis, nausea and vomiting, diarrhea and dilated pupils) and neuromuscular changes (myoclonus, hyperreflexia and tremor).12 Critically, serotonin syndrome has also been reported with the concomitant use of 5-HT3 receptor antagonists (eg, ondansetron, dolasetron, granisetron).13 Because large numbers of pregnant women suffering from depression are prescribed SSRIs, and up to 80% experience morning sickness a possible interaction between SSRIs and ondansetron, leading to serotonin syndrome, must be considered. Because the paramount challenge of treating pregnant women with medications surrounds fetal and maternal safety, ondansetron should be used cautiously only after drugs with a better
safety record, which have been labeled to use in pregnancy (eg, doxylaminepyridoxine) have been tried.
The fetal safety of pyridoxinedoxylamine In contrast to ondansetron, the fetal safety of the pyridoxine-doxylamine combination has been proven in numerous studies and by several metaanalyses, making it one of only few molecules receiving a Pregnancy Category A classification by the FDA. Bendectin was the most frequently prescribed antiemetic for the treatment of nausea and vomiting between 1956 and 1983 with an estimated 33 million exposures. Originally, it was formulated as a delayed-release combination of 10 mg doxylamine succinate, 10 mg pyridoxine and 10 mg dicyclomine hydrochloride. However, in 1976, an 8-way study of doxylamine, pyridoxine HCl, and dicyclomine showed that dicyclomine had no independent antiemetic effect, and subsequently, bendectin was reformulated excluding dicyclomine.14-16 To address the question of potential teratogenicity of the pyridoxinedoxylamine combination in humans, several metaanalyses were conducted, which combined all controlled studies of pregnancy outcome following the use
DECEMBER 2014 American Journal of Obstetrics & Gynecology
603
81 (41-Debendox; 40-placebo)
56
Debendox
Debendox þ10 mg Pyridoxine/Placeboþ10 mg Pyridoxine
Bendectin
Bendectin
Diclectin
Diclectin
Diclectin
Bendectin
Bendectin
Guinness and Binns, 1977
Whealtley et al, 2000
DESI, 1975
DESI, 1975
604 American Journal of Obstetrics & Gynecology DECEMBER 2014
Bishai et al, 2000
Maltepe et al, 2013
Koren et al, 2010
Kutcher et al, 2003
Neutel and Johansen, 1995
Koren. Nausea and vomiting in pregnancy. Am J Obstet Gynecol 2014.
d/c, discontinuation; Gw, gestational week; NVP, nausea and vomiting of pregnancy.
Bendectin; Merrell Dow, Cincinnati, OH. Debendox, Merrell Dow. Diclectin; Duchesnay Inc., Blainville, Canada.
Epidemiologic analysis
Epidemiologic analyses
256 (131-Diclectin, 125-placebo)
31 preemptive Diclectin; 29 only after symptoms started
149 (long-term effectiveness)
2300 (doxylamine/dicyclomine/ pyridoxine)
716 (doxylamine/dicyclomine)
109 (52 Bendectin, 57 placebo)
Bendectin
Geiger et al, 1958
Subjects, n
Drug studied
Author and year
Epidemiologic data
Epidemiologic data
Randomized double blind, multicenter, placebo-controlled trial
Randomized, controlled
Observational prospective
Double blind placebo controlled (various combinations of the ingredients with placebo)
Double-blind placebo controlled
Crossover, double blind placebo controlled
Double-blind placebo controlled
Double blind, placebo controlled
Study type
Studies on the effectiveness of the delayed release combination of doxylamine/pyridoxine
TABLE
2-3 fold increase of hospitalization rate after removal of Bendectin
No teratogenic effect of Bendectin, 2-fold increase in hospitalization, after d/c
Improvement of symptoms in Diclectin group (P < .006)
Preemptive use of Diclectin more effective than when symptoms started
71% effectiveness in first 6-10 gw, 84% in 20 gw
Doxylamine most effective, Dicyclomine no effect, pyridoxine effective for nausea but not vomiting
Doxylanie and dicyclomine more effective than placebo
Improvement in nausea (P < .001), inseverity of retching (P < .05) and vomiting (P < .02)
Improvement of NVP in 70.7% in exposed group/55% in placebo (P < .05)
Improvement of NVP in 94% in exposed group/65% placebo (P < .001)
Results
3
30
29
28
27
16
16
26
25
24
Reference
Clinical Opinion Obstetrics ajog.org
ajog.org of this product during the first trimester of pregnancy. All of these analyses failed to show an overall increase in malformation rates, or in specific malformations. A systematic review of 12 cohort and 5 case-control studies totaling 200,000 patients, calculated an overall summary OR of 1.01, with a 95% CI of 0.66e1.55. When the 2 types of studies were separated according to their design, the summary OR was 0.95 (95% CI, 0.62e1.45) for cohort studies, and 1.27 (95% CI, 0.83e1.94) for case-control studies.17 A second metaanalysis synthesized 16 cohort and 11 case-control studies. The relative risk for any malformation at birth in association with exposure to Bendectin in the first trimester was 0.95 (95% CI, 0.88e1.04). Separate analyses for cardiac defects, limb defects, oral clefts and genital tract malformations yielded pooled estimates of relative risk ranging from 0.81 for oral clefts to 1.11 for limb defects, with no differences in malformation rates between the pyridoxine-doxylamine combination and the controls. As a group, these studies have shown no differences in the risk of birth defects between those infants whose mothers had taken Bendectin during the first trimester of pregnancy and those who had not.18 An ecologic proof of the fetal safety of the pyridoxine-doxylamine combination was published, showing that the withdrawal of the drug from the US market was not associated with decreased rates of major congenital malformations in general, or of any specific malformation.19 In addition, the pyridoxine-doxylamine combination is one of very few drugs that have safety information on the neurodevelopment of children exposed in utero. A prospective controlled cohort study of mother-child pairs was conducted to determine the effects of NVP and its treatment with the pyridoxine-doxylamine combination on child neurodevelopment. Three groups of children were studied at 3-7 years of age: 45 born to mothers who had NVP and were exposed to the pyridoxinedoxylamine combination, 47 with mothers who had NVP but no pyridoxine- doxylamine was used, and 29 born
Obstetrics to mothers not experiencing NVP, and mothers were assessed for IQ and socioeconomic status. The results showed that the pyridoxine-doxylamine combination does not appear to adversely affect fetal brain development and can safely be used to treat NVP.20 In 1989, a report on the safety of the pyridoxine/doxylamine combination for use in the management of NVP was prepared by a panel of Canadian and American experts for the Special Advisory Committee on Reproductive Physiology to the Health Protection Branch of Health Canada (currently called the Health Products and Food Branch). They concluded that “numerous studies in animals and in humans that have been reported in the scientific and medical literature demonstrate that Bendectin is not a teratogen.The safety of the pyridoxine-doxylamine combination in the management of nausea and vomiting of pregnancy has been established by its use in many thousands of pregnant women.”21 These conclusions are similar to those leading the FDA to approve this combination in 2013.2 Similarly, reputable teratogen reference guides concluded that the pyridoxine-doxylamine combination is not associated with an increased risk for adverse pregnancy outcomes.22,23 Because of the extensive fetal safety data that exist, the pyridoxine-doxylamine combination received a FDA Pregnancy Category A classification, indicating that adequate and well-controlled studies have failed to demonstrate a risk to the fetus in the first trimester of pregnancy and there is no evidence of risk in later trimesters.2
The effectiveness of doxylaminepyridoxine The clinical effectiveness of the delayedrelease combination of doxylamine and pyridoxine has been documented over a span of 50 years by several randomized, controlled trials as well as in open postmarketing studies. In addition, several placebo-controlled clinical trials have been published, the results of which have confirmed the effectiveness of this combined agent (Table).
Clinical Opinion
Compelling evidence supporting the effectiveness of this delayed-release combination was provided by populationbased studies showing that the withdrawal of the pyridoxine-doxylamine combination from the US in 1983 was temporally related to a 2- to 3-fold increase in the rates of hospitalization of women for NVP.3 These data suggest that the doxylamine-pyridoxine combination is not only capable of eradicating mild and moderate forms of NVP, but also of preventing severe cases. Data presented by Neutel reiterate these findings: during the 1990s the increased use of the pyridoxine-doxylamine combination by Canadian women has been associated with a reduction in the hospitalization rates for severe NVP. In conclusion, with the availability of a safe and effective FDA-approved drug for NVP, there is no reason for women to be exposed to a drug of unproven maternal and fetal safety, which has not been labeled for NVP. REFERENCES 1. Clark SM, Costantine MM, Hankins GD. Review of NVP and HG and early pharmacotherapeutic intervention. Obstet Gynecol Int 2012;2012:252676. 2. Slaughter SR, Hearns-Stokes R, van der Vlugt T, Joffe HV. FDA approval of doxylaminepyridoxine therapy for use in pregnancy. N Engl J Med 2014;370:1081-3. 3. Neutel CI, Johansen HL. Measuring drug effectiveness by default: the case of Bendectin. Can J Public Health 1995;86:66-70. 4. 4. American College of Obstetricians and Gynecologists. ACOG Practice Bulletin: nausea and vomiting of pregnancy. Obstet Gynecol 2004;103:803-14. 5. American Professors in Gynecology and Obstetrics (APGO) Educational Series on Women’s Health. Nausea and vomiting of pregnancy. Crofton, MD: APGO. 2013. Available at: www.apgo.org. Accessed June 20, 2014. 6. Koren G. Scary science: ondansetron safety in pregnancy-two opposing results from the same Danish registry. Ther Drug Monit 2014 Jan 9 [Epub ahead of print]. 7. Einarson A, Maltepe C, Navioz Y, Kennedy D, Tan MP, Koren G. The safety of ondansetron for nausea and vomiting of pregnancy: a prospective comparative study. BJOG 2004;111: 940-3. 8. Pasternak B, Svanström H, Hviid A. Ondansetron in pregnancy and risk of adverse fetal outcomes. N Engl J Med 2013;368:814-23. 9. Andersen JT, Jimmenez-Solem E, Andersen NL. Ondansetron use in early
DECEMBER 2014 American Journal of Obstetrics & Gynecology
605
Clinical Opinion
Obstetrics
pregnancy and the risk of congenital malformations. Int Soc Pharmacoepidemiol 2013. Abstract 25, Pregnancy session 1. Available at: http:// www.motherisk.org/videos/index.jsp. Accessed Sept. 9, 2014. 10. Anderka M, Mitchell AA, Louik C, Werler MM, Hernández-Diaz S, Rasmussen SA. National Birth Defects Prevention Study. Medications used to treat nausea and vomiting of pregnancy and the risk of selected birth defects. Birth Defects Res A Clin Mol Teratol 2012;94: 22-30. 11. Food and Drug Administration. FDA Drug safety communication. Available at: http://www. fda.gov/Drug/Safety/ucm271913.htm. Accessed Sept. 9, 2014. 12. Mason PJ, Morris VA, Balcezak TJ. Serotonin syndrome. Presentation of 2 cases and review of the literature. Medicine (Baltimore) 2000;79:201-9. 13. Turkel SB, Nadala JG, Wincor MZ. Possible serotonin syndrome in association with 5-HT(3) antagonist agents. Psychosomatics 2001;42: 258-60. 14. Brent R. Bendectin and birth defects: hopefully, the final chapter. Birth Defects Res A Clin Mol Teratol 2003;67:79-87. 15. Kutcher JS, Engle A, Firth J, Lamm SH. Bendectin and birth defects. II: ecologic
analyses. Birth Defects Res A Clin Mol Teratol 2003;67:88-97. 16. Bendectin Peer Review Report 1975. Overall Summary of 8-way Bendectin Study (unpublished study from the FDA databan DESI-10598). FDA1975 Contract no.: DESI 10598. 17. Einarson TR, Leeder JS, Koren G. A method for meta-analysis of epidemiological studies. Drug Intell Clin Pharm 1988;22:813-24. 18. 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. 19. Kutcher JS, Engle A, Firth J, Lamm SH. Analyses. Birth Defects Res A Clin Mol Teratol 2003;67:88-97. 20. Nulman I, Rovet J, Barrera M, KnittelKeren D, Feldman BM, Koren G. Long-term neurodevelopment of children exposed to maternal nausea and vomiting of pregnancy and diclectin. J Pediatr 2009;155:45-50. 21. 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. 22. Briggs GG, Roger K. Freemen, Yaffe SJ. Drugs in pregnancy and lactation, 9th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2011.
606 American Journal of Obstetrics & Gynecology DECEMBER 2014
ajog.org 23. REPROTOX database [electronic version]. Available at: https://www.remotox.org. Accessed Sept. 9, 2014. 24. Geiger CJ, Fahrenbach DM, Healey FJ. Bendectin in the treatment of nausea and vomiting in pregnancy. Obstet Gynecol 1959;14:688-90. 25. McGuinness BW, Binns DT. ‘Debendox’ in pregnancy sickness. J R Coll Gen Pract 1971;21:500-3. 26. Wheatley D. Treatment of pregnancy sickness. Br J Obstet Gynaecol 1977;84:444-7. 27. Bishai R, Mazzotta P, Atanackovic G, et al. Critical appraisal of drug therapy for nausea and vomiting of pregnancy: II. Efficacy and safety of diclectin (doxylamine-B6). Can J Clin Pharmacol 2000;7:138-43. 28. Maltepe C, Koren G. Preemptive treatment of nausea and vomiting of pregnancy: results of a randomized controlled trial. Obstet Gynecol Int 2013;2013:809787. 29. Koren G, Clark S, Hankins GD, 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;203:16. 30. Kutcher JS, Engle A, Firth J, Lamm SH. Bendectin and birth defects. II: ecological analyses. Birth Defects Res A Clin Mol Teratol 2003;67:88-97.
ajog.org
OBSTETRICS
Treating morning sickness in the United States—changes in prescribing are needed Gideon Koren, MD, FRCPC, FACMT, FAACT
Presently, 97.7% of prescriptions for the treatment of nausea and vomiting in pregnancy in the United States are with medications not labeled for use in pregnancy, not indicated for nausea and vomiting in pregnancy, and not classified as safe in pregnancy by the Food and Drug Administration. The use of ondansetron for nausea and vomiting in pregnancy has increased from 50,000 monthly prescriptions in 2008 to 110,000 at the end of 2013, despite unresolved issues regarding fetal safety and Food and Drug Administration warnings about serious dysrhythmias. In April 2013, the Food and Drug Administration approved the combination of doxylamine and pyridoxine, specifically for nausea and vomiting in pregnancy symptoms. Now that a safe and effective drug is available in the United States, there is no reason for women to be exposed to a drug of unproven maternal and fetal safety. Key words: nausea and vomiting, ondansetron, pregnancy
N
ausea and vomiting of pregnancy (NVP) is the most common medical condition in pregnancy, affecting an estimated 80% of pregnant women.1 Because the withdrawal of Bendectin (delayed release doxylamine 10 mgpyridoxine 10 mg; WMS Merrell Dow, Cincinnati, OH) from the US market in 1983, American women did not have access to a Food and Drug Administration (FDA)-approved drug for NVP for 30 years, until the same drug combination (under the trade name Diclegis; Duchesnay Inc., Blainville, Canada) was approved by the FDA in April 2013.2 The withdrawal of Bendectin from the US left American women without an FDA-approved drug for NVP and was associated with a 3-fold increased risk From The Motherisk Program, The Hospital for Sick Children and University of Toronto, Toronto, ON. Received June 17, 2014; revised Aug. 15, 2014; accepted Aug. 18, 2014. Gideon Koren has served as a paid consultant for Duchesnay Inc. Blainville, Quebec, manufacturer of Diclegis. Duchesnay Inc. has supported research studies by the Motherisk Program in Toronto. 0002-9378/$36.00 ª 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ajog.2014.08.017
of hospitalization of women for the severe forms of this condition.3 Presently, 97.7% of prescriptions for the treatment of NVP in the US are with medications not labeled for use in pregnancy, not indicated for NVP, and not classified as safe in pregnancy (FDA category A). The use of ondansetron for the treatment of NVP has steadily increased from 50,000 prescriptions per month in 2008 to 110,000 at the end of 2013 (Figure). This means that around 1 million pregnant American women are exposed to ondansetron out of 4 million pregnancies a year. Ondansetron (GlaxoSmithKlein Inc, Philadelphia, PA) is a serotonin 5-HT3 receptor antagonist, originally introduced to prevent nausea and vomiting induced by cancer chemotherapy, radiation therapy, and surgery. The fact that ondansetron became generic in 2007, and hence its price dropped, might have been an important cause for this increase, with easier access to Medicaid and health maintenance organizations. Prescribing ondansetron as a first line option is not consistent with American Professors in Gynecology and Obstetrics and American College of Obstetricians and Gynecologists evidence-based recommendations for the management of NVP.4,5 It should be remembered that most drugs used in pregnancy, including steroids
602 American Journal of Obstetrics & Gynecology DECEMBER 2014
for the prevention of respiratory distress syndrome, all tocolytic agents, and magnesium sulfate for the prevention of cerebral palsy, to mention a few, have not been approved by the FDA. Yet, they are standard of care. In contrast, in the case of ondansetron there are unresolved issues surrounding the fetal and maternal safety, including recent warnings by the FDA on its potential to cause serious dysrhythmias.6
Fetal safety concerns regarding ondansteron The fetal safety of the ondansetron was first investigated in humans by Einarson et al7 in 2004 through a prospective controlled cohort study of 176 women, in whom we could not detect an increased teratogenic risk. However, this sample size had the statistical power to rule out only a 5-fold increased risk of major malformations, and not any specific malformation. In February 2013, Pasternak et al8 reported that ondansetron was not associated with increased malformation rates when used for morning sickness. This was based on retrospective analysis of data from the Danish Birth Registry, collected between 2004 and 2011 and linked to the National Prescription Register. Each of the 1970 women exposed to ondansetron was matched to 4 unexposed controls. Of note, the mean age at exposure was 10 gestational weeks, which means that half of the cases were exposed to ondansetron at later than 10 gestational weeks, when the morphologic malformations sought in this study could not be produced any more. This can cause a bias toward the null, diluting an existing risk because of inclusion of cases that were not exposed during embryogenesis. However, in August of 2013, Andersen et al9 from Denmark presented a second study using the same Danish registries covering more years (1997-2010) and more pregnant women (897,018 vs
Obstetrics
ajog.org 608, 835). In contrast to Pasternak et al,8 Andersen’s study detected a 2-fold increased risk of cardiac malformations with ondansetron (odds ratio [OR], 2.0; 95% confidence interval [CI], 1.3e3.1), leading to an overall 30% increased risk of major congenital malformations. To rule out confounding by indication, Andersen et al9 also examined metoclopramide taken for morning sickness, detecting no increase in teratogenic risk. The fact that the same large registry can be investigated to yield such opposing results is concerning. There is an exponential rise in use of prescription database linkage to birth registries. None of these were designed specifically to address fetal drug safety, and there may be flaws in the quality and completeness of the available data. Of potential importance, a recent large case control study by the Sloan epidemiology unit and the Centers of Disease Control and Prevention, has reported a 2-fold increased risk for cleft palate associated with ondansetron taken for NVP in the first trimester of pregnancy (OR, 2.37; 95% CI, 1.28e4.76).10
Maternal safety concerns regarding ondansetron The maternal safety of ondansetron has been challenged in June 2012, when the FDA issued a warning of possible serious cardiac output (QT) prolongation and Torsade the Pointe among people receiving ondansetron.11 As a result, the FDA requires strict workup of patients receiving ondansetron, to rule out long QT, electrolyte imbalance, congestive heart failure or taking concomitant medications that prolong the QT interval.12 Because this drug is not approved by the FDA for pregnant women, the FDA did not specifically address precautions in pregnancy. However, in the context of NVP, women with severe NVP often exhibit electrolyte abnormalities (hypokalenia or hypomagnesemia). Presently, counseling of women who receive ondansetron for morning sickness suggests that these FDA precautions are not being followed. Serotonin syndrome is a lifethreatening disorder of excessive serotonergic activity, typically occurring when 2 or more serotonin-modifying agents are used simultaneously, although it may
Clinical Opinion
FIGURE
Sales of antiemetics for NVP in the US, 2008-2014
IMS National Prescription Audit from 2008 to 2014. Dramamine; McNeil Consumer Healthccare, Washington, PA. Ondansetron; GlaxoSmithKlein Inc, Philadelphia, PA. Diclegis; Duchesnay Inc., Blainville, Canada. IMS, IMS Health Inc, Danbury, CT; NVP, nausea and vomiting of pregnancy. Koren. Nausea and vomiting in pregnancy. Am J Obstet Gynecol 2014.
also occur with a single agent.12 From Jan. 1, 1998, to Dec. 30, 2002, Health Canada received 53 reports of suspected serotonin syndrome, most often reported with the use of selective serotonin reuptake inhibitors (SSRIs), monoamine oxidase inhibitors and selective serotoninnorepinephrine reuptake inhibitors. The clinical presentation of seretonergic syndrome is characterized by the triad of cognitive or behavioral changes (confusion, agitation, lethargy, coma), autonomic instability (hyperthermia, tachycardia, diaphoresis, nausea and vomiting, diarrhea and dilated pupils) and neuromuscular changes (myoclonus, hyperreflexia and tremor).12 Critically, serotonin syndrome has also been reported with the concomitant use of 5-HT3 receptor antagonists (eg, ondansetron, dolasetron, granisetron).13 Because large numbers of pregnant women suffering from depression are prescribed SSRIs, and up to 80% experience morning sickness a possible interaction between SSRIs and ondansetron, leading to serotonin syndrome, must be considered. Because the paramount challenge of treating pregnant women with medications surrounds fetal and maternal safety, ondansetron should be used cautiously only after drugs with a better
safety record, which have been labeled to use in pregnancy (eg, doxylaminepyridoxine) have been tried.
The fetal safety of pyridoxinedoxylamine In contrast to ondansetron, the fetal safety of the pyridoxine-doxylamine combination has been proven in numerous studies and by several metaanalyses, making it one of only few molecules receiving a Pregnancy Category A classification by the FDA. Bendectin was the most frequently prescribed antiemetic for the treatment of nausea and vomiting between 1956 and 1983 with an estimated 33 million exposures. Originally, it was formulated as a delayed-release combination of 10 mg doxylamine succinate, 10 mg pyridoxine and 10 mg dicyclomine hydrochloride. However, in 1976, an 8-way study of doxylamine, pyridoxine HCl, and dicyclomine showed that dicyclomine had no independent antiemetic effect, and subsequently, bendectin was reformulated excluding dicyclomine.14-16 To address the question of potential teratogenicity of the pyridoxinedoxylamine combination in humans, several metaanalyses were conducted, which combined all controlled studies of pregnancy outcome following the use
DECEMBER 2014 American Journal of Obstetrics & Gynecology
603
81 (41-Debendox; 40-placebo)
56
Debendox
Debendox þ10 mg Pyridoxine/Placeboþ10 mg Pyridoxine
Bendectin
Bendectin
Diclectin
Diclectin
Diclectin
Bendectin
Bendectin
Guinness and Binns, 1977
Whealtley et al, 2000
DESI, 1975
DESI, 1975
604 American Journal of Obstetrics & Gynecology DECEMBER 2014
Bishai et al, 2000
Maltepe et al, 2013
Koren et al, 2010
Kutcher et al, 2003
Neutel and Johansen, 1995
Koren. Nausea and vomiting in pregnancy. Am J Obstet Gynecol 2014.
d/c, discontinuation; Gw, gestational week; NVP, nausea and vomiting of pregnancy.
Bendectin; Merrell Dow, Cincinnati, OH. Debendox, Merrell Dow. Diclectin; Duchesnay Inc., Blainville, Canada.
Epidemiologic analysis
Epidemiologic analyses
256 (131-Diclectin, 125-placebo)
31 preemptive Diclectin; 29 only after symptoms started
149 (long-term effectiveness)
2300 (doxylamine/dicyclomine/ pyridoxine)
716 (doxylamine/dicyclomine)
109 (52 Bendectin, 57 placebo)
Bendectin
Geiger et al, 1958
Subjects, n
Drug studied
Author and year
Epidemiologic data
Epidemiologic data
Randomized double blind, multicenter, placebo-controlled trial
Randomized, controlled
Observational prospective
Double blind placebo controlled (various combinations of the ingredients with placebo)
Double-blind placebo controlled
Crossover, double blind placebo controlled
Double-blind placebo controlled
Double blind, placebo controlled
Study type
Studies on the effectiveness of the delayed release combination of doxylamine/pyridoxine
TABLE
2-3 fold increase of hospitalization rate after removal of Bendectin
No teratogenic effect of Bendectin, 2-fold increase in hospitalization, after d/c
Improvement of symptoms in Diclectin group (P < .006)
Preemptive use of Diclectin more effective than when symptoms started
71% effectiveness in first 6-10 gw, 84% in 20 gw
Doxylamine most effective, Dicyclomine no effect, pyridoxine effective for nausea but not vomiting
Doxylanie and dicyclomine more effective than placebo
Improvement in nausea (P < .001), inseverity of retching (P < .05) and vomiting (P < .02)
Improvement of NVP in 70.7% in exposed group/55% in placebo (P < .05)
Improvement of NVP in 94% in exposed group/65% placebo (P < .001)
Results
3
30
29
28
27
16
16
26
25
24
Reference
Clinical Opinion Obstetrics ajog.org
ajog.org of this product during the first trimester of pregnancy. All of these analyses failed to show an overall increase in malformation rates, or in specific malformations. A systematic review of 12 cohort and 5 case-control studies totaling 200,000 patients, calculated an overall summary OR of 1.01, with a 95% CI of 0.66e1.55. When the 2 types of studies were separated according to their design, the summary OR was 0.95 (95% CI, 0.62e1.45) for cohort studies, and 1.27 (95% CI, 0.83e1.94) for case-control studies.17 A second metaanalysis synthesized 16 cohort and 11 case-control studies. The relative risk for any malformation at birth in association with exposure to Bendectin in the first trimester was 0.95 (95% CI, 0.88e1.04). Separate analyses for cardiac defects, limb defects, oral clefts and genital tract malformations yielded pooled estimates of relative risk ranging from 0.81 for oral clefts to 1.11 for limb defects, with no differences in malformation rates between the pyridoxine-doxylamine combination and the controls. As a group, these studies have shown no differences in the risk of birth defects between those infants whose mothers had taken Bendectin during the first trimester of pregnancy and those who had not.18 An ecologic proof of the fetal safety of the pyridoxine-doxylamine combination was published, showing that the withdrawal of the drug from the US market was not associated with decreased rates of major congenital malformations in general, or of any specific malformation.19 In addition, the pyridoxine-doxylamine combination is one of very few drugs that have safety information on the neurodevelopment of children exposed in utero. A prospective controlled cohort study of mother-child pairs was conducted to determine the effects of NVP and its treatment with the pyridoxine-doxylamine combination on child neurodevelopment. Three groups of children were studied at 3-7 years of age: 45 born to mothers who had NVP and were exposed to the pyridoxinedoxylamine combination, 47 with mothers who had NVP but no pyridoxine- doxylamine was used, and 29 born
Obstetrics to mothers not experiencing NVP, and mothers were assessed for IQ and socioeconomic status. The results showed that the pyridoxine-doxylamine combination does not appear to adversely affect fetal brain development and can safely be used to treat NVP.20 In 1989, a report on the safety of the pyridoxine/doxylamine combination for use in the management of NVP was prepared by a panel of Canadian and American experts for the Special Advisory Committee on Reproductive Physiology to the Health Protection Branch of Health Canada (currently called the Health Products and Food Branch). They concluded that “numerous studies in animals and in humans that have been reported in the scientific and medical literature demonstrate that Bendectin is not a teratogen.The safety of the pyridoxine-doxylamine combination in the management of nausea and vomiting of pregnancy has been established by its use in many thousands of pregnant women.”21 These conclusions are similar to those leading the FDA to approve this combination in 2013.2 Similarly, reputable teratogen reference guides concluded that the pyridoxine-doxylamine combination is not associated with an increased risk for adverse pregnancy outcomes.22,23 Because of the extensive fetal safety data that exist, the pyridoxine-doxylamine combination received a FDA Pregnancy Category A classification, indicating that adequate and well-controlled studies have failed to demonstrate a risk to the fetus in the first trimester of pregnancy and there is no evidence of risk in later trimesters.2
The effectiveness of doxylaminepyridoxine The clinical effectiveness of the delayedrelease combination of doxylamine and pyridoxine has been documented over a span of 50 years by several randomized, controlled trials as well as in open postmarketing studies. In addition, several placebo-controlled clinical trials have been published, the results of which have confirmed the effectiveness of this combined agent (Table).
Clinical Opinion
Compelling evidence supporting the effectiveness of this delayed-release combination was provided by populationbased studies showing that the withdrawal of the pyridoxine-doxylamine combination from the US in 1983 was temporally related to a 2- to 3-fold increase in the rates of hospitalization of women for NVP.3 These data suggest that the doxylamine-pyridoxine combination is not only capable of eradicating mild and moderate forms of NVP, but also of preventing severe cases. Data presented by Neutel reiterate these findings: during the 1990s the increased use of the pyridoxine-doxylamine combination by Canadian women has been associated with a reduction in the hospitalization rates for severe NVP. In conclusion, with the availability of a safe and effective FDA-approved drug for NVP, there is no reason for women to be exposed to a drug of unproven maternal and fetal safety, which has not been labeled for NVP. REFERENCES 1. Clark SM, Costantine MM, Hankins GD. Review of NVP and HG and early pharmacotherapeutic intervention. Obstet Gynecol Int 2012;2012:252676. 2. Slaughter SR, Hearns-Stokes R, van der Vlugt T, Joffe HV. FDA approval of doxylaminepyridoxine therapy for use in pregnancy. N Engl J Med 2014;370:1081-3. 3. Neutel CI, Johansen HL. Measuring drug effectiveness by default: the case of Bendectin. Can J Public Health 1995;86:66-70. 4. 4. American College of Obstetricians and Gynecologists. ACOG Practice Bulletin: nausea and vomiting of pregnancy. Obstet Gynecol 2004;103:803-14. 5. American Professors in Gynecology and Obstetrics (APGO) Educational Series on Women’s Health. Nausea and vomiting of pregnancy. Crofton, MD: APGO. 2013. Available at: www.apgo.org. Accessed June 20, 2014. 6. Koren G. Scary science: ondansetron safety in pregnancy-two opposing results from the same Danish registry. Ther Drug Monit 2014 Jan 9 [Epub ahead of print]. 7. Einarson A, Maltepe C, Navioz Y, Kennedy D, Tan MP, Koren G. The safety of ondansetron for nausea and vomiting of pregnancy: a prospective comparative study. BJOG 2004;111: 940-3. 8. Pasternak B, Svanström H, Hviid A. Ondansetron in pregnancy and risk of adverse fetal outcomes. N Engl J Med 2013;368:814-23. 9. Andersen JT, Jimmenez-Solem E, Andersen NL. Ondansetron use in early
DECEMBER 2014 American Journal of Obstetrics & Gynecology
605
Clinical Opinion
Obstetrics
pregnancy and the risk of congenital malformations. Int Soc Pharmacoepidemiol 2013. Abstract 25, Pregnancy session 1. Available at: http:// www.motherisk.org/videos/index.jsp. Accessed Sept. 9, 2014. 10. Anderka M, Mitchell AA, Louik C, Werler MM, Hernández-Diaz S, Rasmussen SA. National Birth Defects Prevention Study. Medications used to treat nausea and vomiting of pregnancy and the risk of selected birth defects. Birth Defects Res A Clin Mol Teratol 2012;94: 22-30. 11. Food and Drug Administration. FDA Drug safety communication. Available at: http://www. fda.gov/Drug/Safety/ucm271913.htm. Accessed Sept. 9, 2014. 12. Mason PJ, Morris VA, Balcezak TJ. Serotonin syndrome. Presentation of 2 cases and review of the literature. Medicine (Baltimore) 2000;79:201-9. 13. Turkel SB, Nadala JG, Wincor MZ. Possible serotonin syndrome in association with 5-HT(3) antagonist agents. Psychosomatics 2001;42: 258-60. 14. Brent R. Bendectin and birth defects: hopefully, the final chapter. Birth Defects Res A Clin Mol Teratol 2003;67:79-87. 15. Kutcher JS, Engle A, Firth J, Lamm SH. Bendectin and birth defects. II: ecologic
analyses. Birth Defects Res A Clin Mol Teratol 2003;67:88-97. 16. Bendectin Peer Review Report 1975. Overall Summary of 8-way Bendectin Study (unpublished study from the FDA databan DESI-10598). FDA1975 Contract no.: DESI 10598. 17. Einarson TR, Leeder JS, Koren G. A method for meta-analysis of epidemiological studies. Drug Intell Clin Pharm 1988;22:813-24. 18. 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. 19. Kutcher JS, Engle A, Firth J, Lamm SH. Analyses. Birth Defects Res A Clin Mol Teratol 2003;67:88-97. 20. Nulman I, Rovet J, Barrera M, KnittelKeren D, Feldman BM, Koren G. Long-term neurodevelopment of children exposed to maternal nausea and vomiting of pregnancy and diclectin. J Pediatr 2009;155:45-50. 21. 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. 22. Briggs GG, Roger K. Freemen, Yaffe SJ. Drugs in pregnancy and lactation, 9th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2011.
606 American Journal of Obstetrics & Gynecology DECEMBER 2014
ajog.org 23. REPROTOX database [electronic version]. Available at: https://www.remotox.org. Accessed Sept. 9, 2014. 24. Geiger CJ, Fahrenbach DM, Healey FJ. Bendectin in the treatment of nausea and vomiting in pregnancy. Obstet Gynecol 1959;14:688-90. 25. McGuinness BW, Binns DT. ‘Debendox’ in pregnancy sickness. J R Coll Gen Pract 1971;21:500-3. 26. Wheatley D. Treatment of pregnancy sickness. Br J Obstet Gynaecol 1977;84:444-7. 27. Bishai R, Mazzotta P, Atanackovic G, et al. Critical appraisal of drug therapy for nausea and vomiting of pregnancy: II. Efficacy and safety of diclectin (doxylamine-B6). Can J Clin Pharmacol 2000;7:138-43. 28. Maltepe C, Koren G. Preemptive treatment of nausea and vomiting of pregnancy: results of a randomized controlled trial. Obstet Gynecol Int 2013;2013:809787. 29. Koren G, Clark S, Hankins GD, 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;203:16. 30. Kutcher JS, Engle A, Firth J, Lamm SH. Bendectin and birth defects. II: ecological analyses. Birth Defects Res A Clin Mol Teratol 2003;67:88-97.
