Elsevier

Reproductive Toxicology

Volume 17, Issue 3, May–June 2003, Pages 255-261
Reproductive Toxicology

Maternal drug use in early pregnancy and infant cardiovascular defect

https://doi.org/10.1016/S0890-6238(03)00012-1Get rights and content

Abstract

The purpose of the paper is to identify maternal drug use that may be associated with an increased risk for cardiac defects in the offspring. A case-control study was performed with cases (cardiovascular defects without known chromosome anomalies) being identified from three Swedish health registers (n=5015) and controls being all infants born in Sweden during the period 1 July 1995–2001 (n=577,730). Information on drug exposure was obtained by interview in early pregnancy. Associations between maternal drug use and infant cardiovascular defect were identified for insulin, antihypertensives, fertility drugs, erythromycin, naproxen, anticonvulsants, nitrofurantoin, clomipramine, and budesonide in nasal preparations. Some of these associations are probably due to confounding from underlying disease or complaint, some may be due to multiple testing, some may be true drug effects. Further studies are needed to verify or reject these associations.

Introduction

The association between maternal use of anticonvulsants and infant cardiac defects is well known [1] and the possible teratogenic effect of maternal lithium use has been much discussed [2]. Cardiac defects are components of retinoic acid [3] and thalidomide embryopathy. Studies on the importance of other drugs are few and mainly retrospective.

The most extensive study of risk factors for congenital cardiovascular diseases is the Baltimore-Washington study [4], which included 4390 cardiovascular defect cases. Interviews were performed with the parents of 90% of the cases (n=3377) and with 3572 general population controls. In the analyses of all cardiovascular defects, only three drugs appeared as risk factors: diazepam, metronidazole, and ibuprofen. By breaking up the material into specific cardiovascular defects, further associations (antitussives, clomiphene, corticosteroids, diuretics, and nitrofurantoin) were observed but the number of exposed cases was often small.

The present study analyses a material of roughly the same size as that of the Baltimore-Washington study [4] but using another methodology. Case ascertainment was made by using Swedish country-wide health registers, and information on drug use during early pregnancy was obtained prospectively by midwife interviews towards the end of the first trimester.

Section snippets

Material and methods

Data were obtained mainly from the Swedish Medical Birth Registry [5], based on copies of medical records from antenatal care clinics, delivery units, and the paediatric examination of the new-born. Women who gave birth after 1 July 1995 have information on drug use during pregnancy recorded in the register. The study was, therefore, conducted for the period 1 July 1995 to the end of 2001. At the first antenatal visit (usually week 10–12), a midwife interviewed the woman, among other things on

Infants identified with a cardiovascular defect

Among records on 5565 infants identified with a cardiovascular defect (with the exception of patent ductus arteriosus in a preterm infant and of single umbilical artery), 98% (n=5427) could be linked with the Medical Birth Registry. Among them, 412 (8%) had a known chromosome anomaly and among the remaining 5015, 1094 had an associated non-cardiovascular defect (22%). Among infants without a known chromosome anomaly, 729 had one severe cardiovascular defect, and 308 infants had more than one

Discussion

The total rate of cardiovascular defects identified among infants registered in the Medical Birth Registry was 9.2 per 1000 births which is a rate similar to that described in the literature. Exposure information was obtained prospectively. Little information on the exact time of exposure or amount of drugs used exists and some exposures may have occurred before or after the organo-genetic period of the heart. This error will bias the odds ratio estimates towards unity.

This study concerned only

Acknowledgments

The study was supported by a grant from K.A. Wallenberg Foundation to B.K.

References (15)

There are more references available in the full text version of this article.

Cited by (0)

View full text