The purpose of this thesis is two-fold: (1) to refine understanding of the relationship between anticonvulsant therapy during the first trimester of pregnancy in women with epilepsy and the risk of congenital malformation among their offspring; and (2) to assess the utility of the Saskatchewan Prescription Drug and Hospital Services databases for studies of maternal drug use and certain reproductive outcomes. / In the first meta-analysis the malformation risks associated with the use of anticonvulsants in general by women with epilepsy were quantified and clarified. Comparison of the congenital malformation risk among offspring of mothers with epilepsy with first trimester anticonvulsant exposure ("exposed") relative to offspring of non-epileptic parents yielded a summary estimate of relative risk (RR) of 2.6 (95% confidence interval (CI) 2.1-3.2). (All RR's in this abstract are study-stratified Mantel-Haenszel summary estimates.) Congenital malformation risk among the offspring of exposed women with epilepsy compared to unexposed women with epilepsy yielded a summary RR of 2.9 (CI = 2.0-4.2). No evidence of increased risk to unexposed women with epilepsy compared to non-epileptic women was evident (RR = 0.9, CI = 0.5-1.6). / In the second meta-analysis the risks associated with specific types of anticonvulsant therapy were qualitatively synthesized. The analysis demonstrated the inadequacies of many study reports--vague descriptions of methods often restricted assessment of study quality and incomplete reporting of results as largely responsible for restricting the analysis to 31 studies. Women with epilepsy treated with anticonvulsant monotherapy experienced increased risk of congenitally malformed children relative to both unexposed women with epilepsy (RR = 1.8, CI = 0.8-4.8), and unexposed non-epileptic women (RR = 2.5, CI = 1.8-4.0). Insufficient data were available to demonstrate statistically significant differences in malformation risk among specific commonly-used anticonvulsant monotherapies, although phenobarbital and carbamazepine appeared to have the lowest risks. Two-drug therapy was associated with a 20% increase in risk relative to monotherapy, but three-drug therapy was associated with more than twice the risk of one-drug therapy (RR = 2.2, CI = 1.3-3.7). Although the potential role of confounding by type and severity of epilepsy could not be evaluated, the analysis suggests that avoiding therapy with three or more anticonvulsants during the first trimester would be prudent. / The second component of the thesis was a large record linkage study utilizing information from the databases of Saskatchewan Health. An essentially population-based database of maternal drug use and reproductive outcomes was created which included 104,534 livebirths and 13,685 non-livebirth outcomes occurring between April 1977 and March 1984 linked to 299,152 prescriptions dispensed to the mothers in the year preceding the pregnancy outcome. A study of anticonvulsant use during pregnancy and birth outcome was completed using the created database. The study yielded results with respect to congenital malformation risk generally consistent with the conclusions or the meta-analyses. / Evaluation of the database of maternal drug use and reproductive outcomes raised questions about the utility of Saskatchewan Health's databases for pharmacoepidemiologic research into congenital malformations. (Abstract shortened by UMI.)
| Identifer | oai:union.ndltd.org:LACETR/oai:collectionscanada.gc.ca:QMM.39555 |
| Date | January 1992 |
| Creators | Johnson, Kenneth Clark |
| Contributors | Hanley, J. A. (advisor) |
| Publisher | McGill University |
| Source Sets | Library and Archives Canada ETDs Repository / Centre d'archives des thèses électroniques de Bibliothèque et Archives Canada |
| Language | English |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Format | application/pdf |
| Coverage | Doctor of Philosophy (Department of Epidemiology and Biostatistics.) |
| Rights | All items in eScholarship@McGill are protected by copyright with all rights reserved unless otherwise indicated. |
| Relation | alephsysno: 001323901, proquestno: NN87704, Theses scanned by UMI/ProQuest. |
Page generated in 0.0019 seconds