Third-Generation Oral Contraceptives and Venous Thrombosis

Abstract & Commentary

Synopsis: Third-generation oral contraceptives have laboratory evidence of increased resistance to activated protein C.

Source: Rosing J, et al. Br J Haematol 1997;97:233-238.

Rosing and colleagues from the netherlands used the laboratory test for resistance to activated protein C to compare differences in non-oral contraceptive users, users of second-generation oral contraceptives, and users of third-generation oral contraceptives. The laboratory test was performed on plasmas from 40 normal females and 50 normal males, 24 women using a triphasic oral contraceptive, 32 women using a monophasic oral contraceptive with either levonorgestrel or lynestrenol, and 40 women using monophasic oral contraceptives containing desogestrel, gestodene, or norgestimate.

The effect of resistance to activated protein C, a measure of the Leiden Factor V mutation, was assessed by measurement of thrombin generation in the plasmas. Women who used any oral contraceptive had a decreased sensitivity to activated protein C compared to non-users of oral contraceptives. Women who used the third-generation oral contraceptive were even less sensitive to activated protein C than women using the second-generation oral contraceptives. Indeed, the activated protein C resistance in the third-generation users compared to that measured in women who are heterozygous for the Factor V Leiden mutation. Rosing et al conclude that this difference in activated protein C resistance may explain the previous reports indicating an increased risk for venous thrombosis in oral contraceptive users, especially in those women using the third-generation oral contraceptives.


The publication of the series of papers in 1995-1996 indicating an increased risk of thromboembolism with third-generation oral contraceptives greater than that with older oral contraceptives sparked a great deal of controversy. There were those who believed the epidemiological evidence, and there were those who argued that the conclusions reflected what is called "preferential prescribing." Specifically, this means that the third-generation oral contraceptives perceived to be safer were given to women who were at higher risk for venous thrombosis, and thus the results could be predicted.

Thrombin formation is counterbalanced by fibrinolytic activity. Thus, congenital deficiencies in antithrombin III, protein C, or protein S are associated with an increased risk of thrombosis. It is now apparent that the most common congenital cause of increased venous thromboembolism is the Leiden mutation. The Leiden mutation consists of a single point mutation that results in glutamine being inserted into the protein structure of factor V instead of arginine. Thus, factor V becomes resistant to degradation by activated protein C. This interferes with the ability to break down thrombin. This report suggests that women who use oral contraceptives have an acquired resistance to activated protein C, and that those containing the new generation of progestational agents have a performance in the laboratory tests that was comparable to carriers of the Leiden mutation. Because it has been demonstrated that carriers of the Leiden mutation have a 30- to 50-fold increased risk of venous thrombosis if they use oral contraceptives, this degree of change is very concerning.

In the April 19 issue of The Lancet, Vandenbroucke and Rosendaal from the Leiden University Hospital in the Netherlands write an editorial that concludes that this article is "the nail in the coffin" confirming the epidemiologic evidence. As you know, I was one of the non-believers and skeptics regarding increased venous thrombosis with third-generation oral contraceptives. Are these results from the laboratory in the Netherlands sufficient to change my mind? Frankly, I don’t think the answer is as clear-cut as the editorial in The Lancet would like us to believe. If you look at the 5th-95th percentile range of the test results, there is considerable overlap among all the groups tested. It is only the group of women heterozygous for the Leiden mutation and using oral contraception that is beyond the ranges in the other groups. In fact, many of the OC users have results comparable to the non-users.

Keep in mind that if the results of the observational epidemiologic studies are real, the overall risk is small. The actual incidence of idiopathic venous thrombo-embolism would equal an increase of nine cases per 100,000 woman-years in users of third-generation OCs compared to second-generation OCs. In addition, remember that the mortality rate for these cases is 1% or less.

Several European centers have taken this a step further and have argued that screening for the Factor V Leiden mutation should be routine prior to prescribing oral contraceptives. An excellent article in JAMA (Ridker PM, et al. JAMA 1997;277:1305-1307) measured the prevalence of the Leiden mutation in the American population. The carrier frequency was 5.27% in whites, 1.23% in blacks, 0.45% in Asian Americans, and 1.25% in Native Americans. These estimates are consistent with the European assessments, indicating that this is a trait carried in people of European origin. Thus, in the United States, of the 10 million American women currently using oral contraceptives, about 423,000 are likely to carry the Factor V Leiden mutation. However, because the incidence rate of venous thromboembolism is so low, the number of women required to be screened to prevent one death is tremendously large. Furthermore, because only a small number of women even with the Leiden mutation end up having a clinical event, the finding of a positive screening test would likely be a barrier to the use of oral contraceptives, and one could expect a subsequent increase in unwanted pregnancies. Most American experts believe that screening for the Leiden mutation should only be pursued in women with a previous episode of venous thrombosis or a positive family history for venous thrombosis.

Where does that leave us with this issue of venous thrombosis and the third-generation oral contraceptives? There are those who strongly believe that there is a mechanism and an increased risk with third-generation oral contraceptives, and there are those who believe that these conclusions are biased and not real. We don’t know if acquired changes in a laboratory test as in this report from the Netherlands have clinical meaning. I think this is underscored by the fact that there is no evidence that contemporary oral contraceptive users have an increase in actual mortality rates!