EVRA and Venous Thrombosis — Part II
Abstract & Commentary
By Leon Speroff, MD, Editor, Professor of Obstetrics and Gynecology, Oregon Health and Science University, Portland, is Editor for OB/GYN Clinical Alert.
Synopsis: Insurance company statistics indicate that the risk of venous thromboembolism is greater with transdermal contraception compared with oral contraception.
Source: Cole, JA, et al. Venous thromboembolism, myocardial infarction, and stroke among transdermal contraceptive users. Obstet Gynecol. 2007;109:339-346.
Cole and colleagues from i3 Drug Safety reported that the use of the transdermal contraceptive system was associated with a 2.4-fold increase in the risk of venous thromboembolism compared with users of a 35 µg ethinyl estradiol and norgestimate oral contraceptive.1 i3 Drug Safety is a unit of Ingenix, a medical information company owned by United Health Group. United Health Group also owns UnitedHealthcare, a major national provider of health insurance. The study used insurance claims data from UnitedHealthcare to identify women using transdermal or oral contraception. There were 20 cases of venous thromboembolism with the transdermal method for a rate of 40.8 per 100,000 woman-years, compared with 37 cases with oral contraception for a rate of 18.3 per 100,000 woman-years. This difference yielded a higher rate ratio among transdermal users in the cohort analysis of 2.2 (CI = 1.2-3.8). A nested case-control study was performed in order to control for high risk factors, producing an odds ratio for transdermal users of 2.4 (CI = 1.1-5.5). Myocardial infarction and stroke occurred too rarely to allow risk estimates.
This is the second report on the risk of venous thromboembolism comparing transdermal and oral contraception. Both studies were funded by Johnson & Johnson, the parent company for the EVRA contraceptive patch. The first report was a case-control study of nonfatal venous thrombosis using information derived from a very large database that records prescriptions and diagnoses longitudinally in managed health care plans.2 The study over a 3-year time period compared new users of the contraceptive patch with new users of an oral contraceptive containing 35 µg ethinyl estradiol and norgestimate. Sixty-eight cases of venous thrombosis and 266 controls were identified and matched for year of birth and for the date of the thrombotic episode (thus providing comparable dates for exposure). A comparison of the patch to the oral contraceptive indicated no difference in the risk of venous thrombosis:
|Comparison to OCs|
|Contraceptive patch 31 cases 127 controls OR=0.9 (0.5-1.6)|
The U.S. Food and Drug Administration (FDA) issued a press release on November 10, 2005 calling attention to the fact that women using the oral contraceptive patch are exposed over time to a greater amount of estrogen. Subsequently, the patch labeling was updated to include a warning about this higher exposure. Here are the facts:
- The contraceptive patch delivers daily 20 µg ethinyl estradiol and 150 mg norelgestromin (the primary active metabolite of orally administered norgestimate).
- The serum ethinyl estradiol concentration averages 50 pg/mL, with a range of 25 to 75 pg/mL.3, 4
- The peak estrogen blood levels with the contraceptive patch are about 25% to 35% lower compared with oral products containing 30 µg or 35 µg ethinyl estradiol.5, package label
- Over time, patch users are exposed to about 60% more estrogen compared with an oral product containing 35 µg ethinyl estradiol.
Which is more important, a higher peak level or greater exposure over time? Or maybe it doesn't make a difference.
A weakness of the current report was the inability to confirm diagnoses in 100% of the cases through review of medical records (completed for 83% of the cases). A strength of the study is the thorough attempt to control for factors that influence venous thrombosis in the case-control study. A strong point of both studies was a focus on new users, eliminating the problem known as attrition of susceptibles (comparing new users, who are more likely to experience venous thrombosis, to old users would be comparing two different groups of subjects).
Comparing the two studies, it is worth noting that the negative report had a little more than 3 times as many cases as the positive report, giving it more statistical power. This is reflected in the wider confidence intervals of the current study indicating an increase in risk, an indication of a less precise conclusion. Indeed, in the overall case-control analysis there were only 20 cases among transdermal users and 37 cases among oral contraceptive users, and the odds ratio of 2.0 was close, but it did not reach statistical significance (CI = 1.0-4.1). After excluding cases and controls with high-risk factors, the odds ratio with 16 transdermal cases and 26 oral contraceptive cases achieved 2.4 with statistical significance, but a relatively very wide confidence interval, 1.1-5.5.
So where does that leave clinicians and patients? These are the first epidemiologic data on this important issue. One study is reassuring, one is disturbing. But note that the confidence interval in the second study is relatively wide, indicating imprecision of the conclusion. Because the confidence intervals of the two studies are within the same range (ie, they overlap), it is not certain that the different results do not reflect a chance finding.
The authors of the transdermal report raise a criticism of the earlier, negative report. They point out that the first report compared only first-time users of each method, and they conclude that prior users of norgestimate oral contraceptives were thus excluded from the oral contraceptive group, but could not be excluded from the transdermal group because of its recent introduction to the market.
This confusion underscores the inappropriate strategy of the FDA in issuing press releases and making conclusions public before the publications are available for assessment by clinicians. At this point in time, it seems to me, that if there is a difference in the risk of venous thrombosis comparing transdermal and oral contraception, the difference has to be very small. Certainly the available evidence does not indicate a major difference.
- Cole, JA, et al. Venous thromboembolism, myocardial infarction, and stroke among transdermal contraceptive users. Obstet Gynecol. 2007;109:339-346.
- Jick SS, et al. Risk of nonfatal venous thromboembolism in women using a contraceptive transdermal patch and oral contraceptives containing norgestimate and 35 ug of ethinyl estradiol. Contraception. 2006;73:223-228.
- Abrams LS, et al. Multiple-dose pharmacokinetics of a contraceptive patch in healthy women participants. Contraception. 2001;64:287-294.
- Abrams LS, et al. Pharmacokinetics of a contraceptive patch (Evra/Ortho Evra) containing norelgestromin and ethinyloestradiol at four application sites. Br J Clin Pharmacol. 2002;53:141-146.
- Van Den Heuvel MW, et al. Comparison of ethinylestradiol pharmacokinetics in three hormonal contraceptive formulations: the vaginal ring, the transdermal patch, and an oral contraceptive. Contraception. 2005;72:168-174.