The use of any oral contraceptive significantly raises the likelihood of a first myocardial infarction, according to a nationwide study from the Netherlands.¹ The results of the study also suggest, although inconclusively, that women who use third-generation pills (those containing the progestogen gestodene or desogestrel) are less likely to have a heart attack than those using pills containing the second-generation drug levonorgestrel, and that the use of third-generation pills is not associated with the risk of heart attack.

The investigation aimed to address a flaw that existed in most similar published studies, by recruiting sufficient numbers of women using second- or third-generation pills so that effects on the risk of myocardial infarction could be compared. The researchers note that these two types of contraceptive are commonly used in the Netherlands, so the population of potential study subjects was large. The analysis also included the use of first-generation pills (those containing the progestogens lynestrenol or norethindrone).

The investigators conducted the population-based case-control study by sending a standardized questionnaire to women aged 18-49 who had been hospitalized for a first myocardial infarction between January 1990 and October 1995, and to a randomly selected group of controls who had not had a myocardial infarction. Controls were matched to women who had had a heart attack by five-year age-group and area of residence, and they were asked to respond to the questionnaire with reference to a specific year between 1990 and 1995.

In all, 248 women who had had a myocardial infarction and 925 controls completed the survey. Women in the study group were, on average, older than the controls (43 vs. 38). They were also more likely to be current smokers (84% vs. 43%); have a history of hypertension (24% vs. 6%), high cholesterol (11% vs. 3%) and diabetes (6% vs. 1%); and have a family history of cardiovascular disease (65% vs. 36%).

Women were more likely to be current users of second-generation pills than of any other type: 24% of the study group and 19% of controls, compared with 8% of the study group and 12% of controls who used third-generation pills, and 4% of the study group and 3% of controls who used first-generation pills. After adjustment for confounding factors (age, area of residence, calendar year and risk factors for cardiovascular disease), logistic regression analyses showed that current users of any pill type and current users of first- and second-generation pills were significantly more likely than nonusers to have a heart attack (odds ratios, 2.1, 2.7 and 2.5, respectively). In contrast, users of third-generation pills seemed to be as likely as nonusers but less likely than users of second-generation pills to have a heart attack (1.3 and 0.5, respectively); the researchers concede, however, that the 95% confidence intervals for these results were too wide for a definite conclusion to be drawn.

The study also analyzed the effect of other cardiovascular risk factors. Logistic regression analyses showed that the likelihood of myocardial infarction was elevated among both pill users and nonusers if they had hypertension (6.1 and 5.1, respectively) or were obese (5.1 and 3.4, respectively). The likelihood was dramatically higher for pill users, but not nonusers, if they had high cholesterol (24.7 vs. 3.3) or diabetes (17.4 vs. 4.2), or if they were current smokers (13.6 vs. 7.9).

The researchers note that only one other investigation has recruited sufficient users of third-generation oral contraceptives to be able to compare the effect of using third- and second-generation drugs on the risk of myocardial infarction. They argue that although those results suggest that third-generation pills carry the higher risk of heart attack, the 95% confidence interval again was too wide to permit a definite conclusion. The author of an editorial accompanying the Dutch study adds that recall bias may have affected that finding.² Recall bias was minimized in this study by including in the questionnaire color photographs of all available oral contraceptives in the Netherlands. In addition, the researchers found no evidence of prescription bias, and they avoided selection bias by selecting patients nationwide and according to their hospital discharge diagnosis.

The researchers admit that the absolute risk of heart attack among pill users is small, but "because all combined oral contraceptives are equally effective means of birth control, the issue of safety is paramount." In practical terms, the authors advise that "before prescribing oral contraceptives, clinicians should screen women for conventional risk factors for cardiovascular events."--T. Lane

REFERENCES

1. Tanis BC et al., Oral contraceptives and the risk of myocardial infarction, New England Journal of Medicine, 2001, 345(25):1787-1793.

2. Chasan-Taber L and Stampfer M, Oral contraceptives and myocardial infarction--the search for the smoking gun, editorial, New England Journal of Medicine, 2001, 345(25):1841-1842.