A Final Word on Safety of Vaginal Ring Contraception?
Abstract & Commentary
By Jeffrey T. Jensen, MD, MPH
Synopsis: A large prospective study from the United States and Europe confirms that the risk of venous and arterial thromboembolism associated with use of the contraceptive vaginal ring is similar to that seen with use of combined pills.
Source: Dinger J, et al. Cardiovascular risk associated with the use of an etonogestrel-containing vaginal ring. Obstet Gynecol 2013;122:800-808.
The transatlantic active surveillance on cardiovas-cular Safety (TASC) of Nuvaring® study was a large, multinational, controlled, prospective, observational, active surveillance study designed to assess the cardiovascular safety of the ethinyl estradiol/etonogestrel releasing contraceptive vaginal ring (CVR). The study was performed at 1661 study sites in the United States and five European countries. These sites enrolled a total of 33,295 subjects who were new users (new starters, switchers, or restarters) of the CVR or of a combined oral contraceptive (OC) pill. The choice of method was made by the woman and/or her doctor. After the choice of method was completed, the subject was invited to join the study. This required the subject to complete a health history and provide contact information to the central monitoring site. Follow-up occurred for 2 to 4 years with an impressively low loss to follow-up of only 2.9%. Altogether, a total of 66,489 woman-years of exposure to the hormonal methods were recorded. The primary clinical outcomes of interest (venous and arterial thromboembolism) were validated by attending physicians and further adjudicated by an independent board. The venous thromboembolic hazard ratio (HR) for the vaginal ring compared with combined OCs was the primary statistical variable.
The crude venous thromboembolism incidence rates for the vaginal ring users and combined OC users were 8.3 and 9.2 per 10,000 woman-years, respectively. Cox regression analysis for the vaginal ring users compared with combined OCs yielded adjusted hazard ratios (HRs) of 0.8 (95% confidence interval [CI], 0.5-1.6) for venous and 0.7 (95% CI, 0.2-2.3) for arterial thromboembolism. When the analysis was restricted to compare the CVR to levonorgestrel pills only, the adjusted HR was 1.0 (95% CI, 0.3-3.3). The authors concluded that the CVR is associated with a risk of venous and arterial thromboembolic similar to that of the combined pill during routine clinical use.
Commentary
Over the last year, I have written extensively about the ongoing controversy regarding safety of hormonal contraceptives. This paper provides another excellent example of how to perform a good Phase 4 study. The methodology of the TASC study is essentially the same as that of the European Active Surveillance (EURAS) study of drospirenone-containing OCs.1 Both studies were mandated by regulatory authorities, paid for by industry (EURAS Bayer, TASC Merck), and conducted by Jürgen Dinger of the ZEG institute in Berlin. The planning, conduct, and evaluation of the study were supervised by an independent Safety Monitoring and Advisory Council, ethical approval was obtained through an independent agency, and the study was registered in the public clinical trials registry of the United States National Library of Medicine. This transparency makes it easy to follow the logic of the methods.
The results demonstrate no difference in the risk of venous thromboemolism and arterial thrombotic events associated with use of the CVR when compared to combined pills. Since the design of prospective Phase 4 studies permits the collection and analysis of important baseline confounders, these studies provide the highest quality of evidence on uncommon and rare side effects associated with hormonal contraception. I have written extensively about the limitations of database studies in assessing these same outcomes. While the series of manuscripts by Lidegaard2,3 using data from the Danish National Database show a consistent increase in risk with newer progestins such as desogestrel (and its metabolites such as etonogestrel) and drospirenone when compared with levonorgestrel, the inability to adjust for baseline confounding represents a significant limitation.4 I encourage each of you to read the excellent discussion by Dinger in the TASC paper.
Although we have high acceptance of the CVR in my area, the use of long-acting reversible contraception methods continues to challenge the market for combined hormonal methods. For our residents, the prescription of a combined method is now a "teachable moment." Combined hormonal methods provide a number of important nonhormonal benefits, and are of particular value for women who experience symptoms due to persistent ovarian follicles, ovulation events, or cycle-related hormonal symptoms. The CVR can provide the benefits of a combined method and the convenience of once per month dosing. I believe in providing many options and allowing women to make an informed choice of what is right for them. This requires accurate information of risks, so I encourage you to take the time to teach.
References
- Dinger JC, et al. Contraception 2007;75:344-354.
- Lidegaard O, et al. BMJ 2011;343:d6423.
- Lidegaard O, et al. BMJ 2012;344:e2990.
- Shapiro S, Dinger J. J Fam Plann Reprod Health Care 2010;36:33-38.
