Sara A. Ephross

Monitoring pregnancy outcomes after prenatal drug exposure through prospective pregnancy registries: A pharmaceutical company commitment

OBJECTIVE Glaxo Wellcome becomes aware of prenatal exposures to its medications as early as the clinical trial phase of development. An international process for monitoring prenatal exposure to all Glaxo Wellcome medicines has been developed. For specific products there are prospective pregnancy registries. STUDY DESIGN The registries are observational, case-registration, and follow-up studies designed to detect evidence of teratogenicity associated with specific medications. After prenatal exposure to the registry medication, pregnancies are registered prospectively, through voluntary reports by health care providers. An advisory committee of independent scientists for each registry reviews data and advises in dissemination of information. Risk of birth defects, as defined by the Centers for Disease Control and Prevention, is compared with published risks both in women in the general population and in women with the underlying condition being treated, if available. RESULTS The following data show results from the prospective first-trimester exposures registered since establishment of each registry. The published risk of birth defects in the general population range is 3% to 5%, and the risk in women with epilepsy is 6% to 9%. The proportions of outcomes with birth defects are as follows: in the Acyclovir (antiviral medication) Pregnancy Registry (1984-1998) (19/581), 3.3% (95% confidence interval, 2.0%-5.2%); in the Lamotrigine (monotherapy and polytherapy antiepileptic medication) Pregnancy Registry (1992-September 1998) (8/123), 6.5% (95% confidence interval, 3.1%-12.8%); in the Sumatriptan (migraine medication) Pregnancy Registry (1996-October 1998) (7/183), 3.8% (95% confidence interval, 1.7%-8.0%). The Valacyclovir, Bupropion, and Naratriptan registries have insufficient data for analysis. CONCLUSION None of the registries has provided a risk estimate exceeding that expected in the disorder treated, and no pattern of defects has been observed. Whereas information from the larger registries is reassuring regarding risk, these studies cannot rule out possible small excess risks from use of these drugs in pregnancy. Data obtained through these registries are shared with the medical community as a supplement to animal toxicology studies to assist in weighing potential risks and benefits of treatment for individual patients. The success of the registries depends on the continued willingness of the obstetrics and gynecology community to notify the registries of prenatal exposures.

First trimester paroxetine use and the prevalence of congenital, specifically cardiac, defects: A meta‐analysis of epidemiological studies

BACKGROUND Several studies have evaluated maternal first trimester paroxetine use and the prevalence of congenital defects, particularly cardiac defects. To synthesize current epidemiologic information, a meta-analysis was conducted. METHODS A systematic literature search was conducted for original research published from January 1, 1992, through September 30, 2008. Results were extracted using predefined criteria, and authors were contacted for additional information when necessary. Compiled results were evaluated for funnel plot asymmetry, heterogeneity, and study characteristic associations. Where appropriate, fixed-effect summary estimates were calculated and sensitivity analyses performed. RESULTS Twenty reports (11 including results for aggregated congenital and combined cardiac defects, six for aggregated congenital defects only, and three for combined cardiac defects only) met prespecified inclusion criteria. There was little evidence of funnel plot asymmetry or overall heterogeneity. Summary estimates were produced for combined cardiac defects (prevalence odds ratio [POR], 1.46; 95% confidence interval [CI], 1.17-1.82) and aggregated congenital defects (POR, 1.24; 95% CI, 1.08-1.43) and first trimester paroxetine use. Some study characteristics may be associated with differential POR estimates for paroxetine and either combined cardiac or aggregated congenital defects. CONCLUSIONS This meta-analysis found little evidence of publication bias or overall statistical heterogeneity and only weak evidence of associations with some study characteristics. Although subject to limitations, the summary estimate indicates an increased prevalence of combined cardiac defects with first trimester paroxetine use. The summary estimate also indicates an increased prevalence of aggregated congenital defects with paroxetine; however, this association may be explained, in part, by the increased prevalence of combined cardiac defects.

Menstrual and reproductive risk factors for ischemic heart disease

The role of hormones in ischemic heart disease is of considerable interest, but limited data are available pertaining to risk factors associated with endogenous hormones. We examined the association between menstrual and reproductive factors and ischemic heart disease in a cohort of 867 white, college-educated women who prospectively recorded menstrual cycle data for at least 5 years from their early 20s through their menopause. Ischemic heart disease history was obtained from a self-administered (N = 714) or proxy-administered (N = 153) questionnaire completed at a mean age of 73 years. The analysis included 44,899 person-years of follow-up and 45 cases of myocardial infarction, angioplasty, heart bypass surgery, or ischemic heart disease-related mortality. Ischemic heart disease risk decreased with increasing age at menarche (age-adjusted RR 0.76 per year, 95% CI = 0.60-0.95). Considering menstrual cycle characteristics ages 28-32, there was little overall association with length, variability, or bleeding duration. Ischemic heart disease risk increased with later age at first birth (age-adjusted RR 2.90 for ages 33-43 compared with 25-29) and later age at last birth (age-adjusted RR 3.79 for ages > or =40 compared with 35-39), but there was little association with high parity.

Pregnancy and perinatal outcomes in migraineurs using sumatriptan: a prospective study

Background: Sumatriptan is an acute treatment for migraine which is often used by women in their child-bearing years, and who become unexpectedly pregnant. Within the context of the post-marketing use of sumatriptan injection for the acute treatment of migraine, and in compliance with approved labeling, we wished to compare perinatal pregnancy outcomes in women who did and did not use the drug after conception. Methods: Open-label, prospective study conducted in 12,339 migraineurs (including 9,861 women) whose demography and consumption pattern of sumatriptan injections were typical, and were predicted to include 150 pregnancies. Outcome of pregnancy was the end-point. Results: There were 168 of 173 pregnancies that were well-documented. Sumatriptan was only used prior to conception in 92 cases. There were 76 first trimester exposures to sumatriptan. There were no differences in pregnancy outcome between the two groups. Conclusions: Perinatal and pregnancy outcome did not differ between patients who had and had not used sumatriptan after conception, at the resolution of these sample sizes. This study design complements the ongoing pregnancy registry, which is now widened to patients exposed to all formulations of sumatriptan.

Menstrual patterns and risk of adult-onset diabetes mellitus

We examined the association between menstrual patterns and risk of developing adult-onset diabetes in a prospective study of 668 white, college-educated women who completed menstrual diaries throughout their reproductive years. We calculated summary measures of cycle length and variability and bleeding duration for ages < or = 22, 23-27, 28-32, and 33-37 years. The analysis included 35,418 person-years of follow-up and 49 self-reported cases of diabetes (median age at diagnosis, 63 years). There was no association between diabetes risk and age at menarche, mean cycle length, cycle variability, or frequency of long cycles (> 42 days). Longer bleeding periods in the mid- and late reproductive years were somewhat associated with an increased risk of diabetes (adjusted rate ratio 1.4, 95% confidence interval 1.0-1.8 per day increase in bleeding duration for menses during ages 28-32). These results do not support the association of long or irregular menstrual cycles with post-menopausal diabetes incidence, but do suggest a possible association of longer bleeding duration with subsequent onset of diabetes.

Final Results From the 16-Year Sumatriptan, Naratriptan, and Treximet Pregnancy Registry

To monitor for a signal of major teratogenicity by determining the risk of all birth major defects following in utero exposure to sumatriptan, naratriptan, and the sumatriptan/naproxen sodium combination product (tablets marketed in the United States as Treximet [GlaxoSmithKline, Research Triangle Park, NC, USA]), and to monitor for unusual patterns of defects that might suggest teratogenicity.

Bupropion therapy in pregnancy and the occurrence of cardiovascular malformations in infants

Bupropion is an aminoketone marketed as an oral antidepressant and non-nicotine smoking cessation aid.1,2 It is listed by the United States Food and Drug Administration as a pregnancy category C medication. A retrospective cohort study of maternal first trimester bupropion use and the occurrence of congenital malformations in infants previously published by Cole et al. (2007)3 showed that the prevalence of all congenital and all cardiovascular malformations among infants born to women exposed to first trimester bupropion therapy, regardless of concurrent exposure to other antidepressants, was comparable with infants born to women exposed to all other antidepressants in the first trimester or to bupropion outside of the first trimester. A subsequent article by Alwan et al. (2010)4 described the association between maternal bupropion treatment in early pregnancy and congenital heart defects in infants from National Birth Defects Prevention Study (NBDPS) data. They reported an approximately two and a half-fold increase in left ventricular outflow tract obstruction (LVOTO) defects with bupropion use in early pregnancy (crude odds ratio [OR]: 2.2, 95% confidence interval [CI]: 1.0–4.8; adjusted OR: 2.6, 95% CI: 1.2–5.7). Alwan et al. (2010)4 also calculated a crude risk ratio (RR) using the Cole et al. (2007)3 published data and found an approximately threefold increase in LVOTO with first trimester bupropion exposure compared with other antidepressant use (RR: 3.2; 95% CI: 1.0–10.6).4 Given the findings of Alwan et al. (2010)4 using both NBDPS and the published Cole et al. (2007)3 data, we refined the original analysis of the Cole et al. (2007)3 data to examine specific cardiovascular malformations, in particular LVOTO. To validate the findings of Alwan et al. (2010)4 using the Cole et al. (2007)3 data, we refined the exposure categories to include monotherapy cohorts and clinically categorized the specific cardiovascular malformations using the same system employed by Alwan et al. (2010)4 and described in detail by Botto et al. (2007).5 This classification system, based on developmental and epidemiologic considerations, is designed to facilitate the assessment of isolated cardiac defects. We utilized the same study population identified from an administrative claims database for the Cole et al. (2007)3 study of bupropion use during pregnancy, which consisted of women aged 12–49 years at the time of delivery with at least one dispensing of bupropion or another antidepressant from pharmacy claims between 1 January 1995 and 30 September 2004. Methods for dataset development are described in detail elsewhere.3 The three following exposure cohorts originally created in Cole et al. (2007)3 included: maternal first trimester exposure to bupropion, regardless of other antidepressant exposure (‘bupropion first trimester monotherapy or polytherapy’), maternal first trimester exposure to antidepressants other than bupropion (‘other antidepressants first trimester monotherapy or polytherapy’), and maternal exposure to bupropion outside the first trimester, regardless of other antidepressant exposure (‘bupropion outside first trimester monotherapy or polytherapy’). For the current analysis, the following monotherapy cohorts were also subsetted from the original cohorts: maternal first trimester exposure to only bupropion (‘bupropion first trimester monotherapy’), maternal first trimester exposure to only one type of antidepressant other than bupropion, (‘other antidepressants first trimester monotherapy’), and maternal exposure to only bupropion outside the first trimester (‘bupropion outside first trimester monotherapy’). For this analysis, a pediatric cardiologist reviewed previously abstracted medical records from Cole et al. (2007)3 for specific cardiovascular malformations. The malformations were classified using the same categorizations used in Alwan et al. (2010)4 and described in detail by Botto et al. (2007).5 To

Monitoring birth outcomes in the Sumatriptan Pregnancy Registry

Background: Health professionals often need to make medical management decisions by weighing potential risk versus benefit. Because 50% of pregnancies in the United States are unplanned, the likelihood for unintentional first-trimester exposure to medications is high and increases when women of childbearing potential (ages 15-44) are taking medications used to treat chronic or recurrent conditions, such as migraine. Prospective collection of pregnancy exposure and outcome data is one method for providing data to the medical community as a tool for making risk assessments.Methods: The Sumatriptan Pregnancy Registry is an international, exposure-registration and follow-up study to prospectively collect prenatal sumatriptan exposure and outcome data. An advisory committee reviews data and assists in disseminating information. Committee members include independent scientists with expertise in obstetrics, pediatrics, neurology, internal medicine, epidemiology, clinical genetics, and teratology.Results: As of April 30, 1997, 181 outcomes were prospectively reported from pregnancies involving use of sumatriptan. Of the 171 outcomes where earliest exposure was in the first trimester there were 143 infants without birth defects, 4 live infants with birth defects, 1 stillbirth with birth defects [12 spontaneous pregnancy losses, 2 stillbirths, and 9 induced abortions without birth defects]. The proportion of outcomes involving birth defects, 5/148, 3.4% (95% CI 1.3%, 8.1%), does not differ from the expected proportion in the general population.Conclusions: To date, no pattern has emerged in either the prospectively or retrospectively reported defects. Registry data are disseminated to interested health care providers who are then in a more informed position when making medical management decisions. However, more data are needed before definitive conclusions can be drawn about the safety of sumatriptan use in pregnancy.

Monitoring pregnancy outcomes following prenatal drug exposure through prospective pregnancy registries and passive surveillance: a pharmaceutical company commitment.

Objectives: Glaxo Wellcome (G.W.) becomes aware of prenatal exposures to its medications from as early as the clinical trial phase of development. An international process for monitoring prenatal exposure to all G.W. medicines has been developed utilizing a passive surveillance system and, for specific products, pregnancy registries. Additionally, G.W. jointly sponsors the multi-company Antiretroviral and North American Antiepileptic Drug (AED) Pregnancy Registries.Study Design: The registries are observational, case-registration and follow-up studies designed to detect evidence of teratogenicity associated with specific medications. Pregnancies are registered prospectively following prenatal exposure to the registry medication. An advisory committee for each registry reviews data and assists in dissemination of information. Committee members include independent scientists with expertise in fields such as obstetrics, teratology, epidemiology, pediatrics, and the relevant therapeutic areas.Results: The following data are from the prospective first-trimester exposures in each registry. Through December 1996, the proportion of outcomes in the Acyclovir Pregnancy Registry with birth defects (n = 17/505) is 3.4% (95% CI 2.0%, 5.4%). Through March 1997, the proportion of outcomes in the Lamotrigine Pregnancy Registry with birth defects (n = 4/76) is 5.3% (95% CI 1.7%, 13.6%). Through April 1997, the proportion of outcomes in the Sumatriptan Pregnancy Registry with birth defects (n = 5/148) is 3.4% (95% CI 1.3%, 8.1%). The newer Valacyclovir and Bupropion Pregnancy Registries have insufficient data for analysis. None of the registries have provided a risk estimate exceeding that expected in the general population, and no pattern of defects has been observed.Conclusions: The outcomes accumulated to date represent a sample of insufficient size for reaching conclusions regarding the possible teratogenic risk of using these drugs in pregnancy. Data obtained through these registries are shared with the medical community as a supplement to animal toxicology studies and to assist in weighing potential risks and benefits of treatment for individual patients. The success of the registries depends on the continued willingness of the OB/GYN community to notify the registries of prenatal exposures.