Cardiovascular safety of oral contraceptives: a critical commentary
Sturtevant, F.M.
International Journal of Fertility 36(Suppl 3): 32-36
1991
ISSN/ISBN: 0020-725X PMID: 1687402 Document Number: 372344
Oral contraceptive (OC) scares began with the reported deaths of 2 women from pulmonary embolism. Consequently, for the past 30 years OCs have been evaluated for their effects on the cardiovascular system. A critical appraisal is made of epidemiologic and prospective studies of the relationship of OC use to deleterious cardiovascular effects and the experimental evidence of cardiovascular safety. Case control methodology changed the evaluations of the safety of OCs. In 1967, the Royal College of General Practitioners produced case control results which correlated OC use with thrombophlebitis. Reports of Inman and Vessey, and Vessey and Doll, reinforced these results. Computerized multivariate analysis, which would have controlled for confounding factors, was not available. Other documentation of the effects of OC use on ischemic strokes and thromboembolism by Sartwell and Heyman have been criticized by Goldzieher and Associates. Reports in 1970 of a dose-response relationship led to fear of OC use. In 1980, the US Food and Drug Administration concluded that there was such a relationship and prohibited marketing of OCs which contained more than 50 mg of estrogen. Mann and Inman found a relationship with myocardial infarction for women over the age of 40. The result has been to restrict OC use for women smokers over the age of 35 years, in spite of the critiques by Goldzieher and the author. Studies on the frequency of thrombophlebitis and related deaths were published first by Fuertes-de la Haba in 1970 and critiqued by Drill who found no differences in the OC population from the general population. Clifford Kay produced the first large-scale, controlled study in 1974 which showed the risk of thrombotic disease from OC use and smoking, but that research is considered to be flawed due to the lack of control for smoking. As recently as 1989 Vessey analyzed prospectively the risk among family planning clinic users and found no difference between OC users and nonusers. The US Walnut Creek Study found no support for the earlier findings of Vessey nor did the Puget Sound study. The new findings reflect the improved analytical power of multivariate analysis, the reluctance of doctors to prescribe to those with risk factors, and improved diagnostic accuracy.