Mechanism of antihypertensive action of ketanserin in man.

paper on mortality5 published in 1977, which she quotes, there did appear to be a significantly increased risk after five years of use. This article was superseded, however, by the publication in 19816 of a paper based on a larger database, which failed to show any evidence of an effect of duration. There is thus now no convincing evidence that duration of use has an important influence on the excess risk of vascular disease associated with oral contraceptive use. This statement needs a minor qualification. There is tentative evidence from a study of perimenopausal former users that their current risk of myocardial infarction became materially related to former use of oral contraceptives longer than 10 years.7 Our own analysis of vascular morbidity8 (fatal and non-fatal disease) suggested a possible relationship of cerebrovascular disease to duration of use, but the trend was not statistically significant. In the absence of sound statistical guidance I would personally consider critically the situation of a woman who had been using oral contraceptives for longer than 10 years. Even here, in the absence of any other risk factors for vascular disease and a marked preference for continuing oral contraceptives, I would be inclined to do so. There has never been any evidence that periodic rests from oral contraceptive use are beneficial, and I would have thought that the attendant substantial risks of an unwanted pregnancy were obvious. Finally, although Dr Patel rightly draws attention to the increased risk of pill use for a woman over the age of 35 who smokes cigarettes, she fails to make the point that a comparable risk for a non-smoker is not reached until the woman is well into her fifth decade. The relevant data are presented in our 1981 mortality analysis,6 which, because it contains more good than bad news for pill users, has never received the publicity I believe it deserves. CLIFFORD R KAY