The trusted source for
healthcare information and
Oral Contraceptive Use in Obese Women
By Jeffrey T. Jensen, MD, MPH, Editor, Leon Speroff, Professor and Vice Chair for Research, Department of Obstetrics and Gynecology, Oregon Health and Science University, Portland, is Editor for OB/GYN Clinical Alert.
Obesity is a growing problem in the united states. the pun would be funny were it not for the adverse influence that the obesity epidemic has had on health care outcomes. Obesity affects every aspect of health, including contraception. The National Institutes of Health has recognized that our understanding of the interactions between obesity and hormonal contraception is inadequate. To address this knowledge gap, NICHD convened a workshop in November 2010 to bring together experts in the field to discuss the state of the science and to plan a research agenda.1
Alan Guttmacher, MD, (nephew of the Alan Gutt-macher of the Guttmacher Institute), the new Director of NICHD, started out the meeting noting that it is imperative that the institute examine the intersection of obesity and contraception, because part of the Institute's mission is "to ensure that every person is born healthy and wanted" and "that women suffer no harmful effects from reproductive processes." Guttmacher issued a strong statement of support for research in this area, recognizing that there are few other funding entities engaged in this field and asking: "If NICHD doesn't do it, who will?"
Bliss Kaneshiro, MD (University of Hawaii), reviewed data from two nationally representative surveys, the National Survey of Family Growth (NSFG) and the Behavioral Risk Factor Surveillance System (BRFSS).2,3 Analyses of these data have demonstrated that there is no significant difference in contraceptive non-use among women of different BMIs. Patterns of use are less clear; the cycle 6 NSFG data showed that the patterns of methods used by obese and normal weight women do not differ, whereas the BRFSS survey showed that normal weight women tend to use hormonal methods of contraception, while overweight and obese women tend to use procedural methods (like surgical sterilization). Approximately 20% of American women use oral contraceptives (OC) for birth control, and use of the pill does not seem to differ by BMI group. Data from the NSFG also demonstrated no differences in most measures of sexual behavior, except that overweight and obese women are more likely than normal weight women to report an "ever" history of coitus. Since sexual behavior does not seem to differ, what about efficacy? While data from cycle 6 of the NSFG showed no association between unintended pregnancy and BMI, the Pregnancy Risk Assessment Monitoring System (PRAMS) has detected more unintended pregnancies in obese and overweight women using contraception. However, surveys using self report to evaluate pregnancy intention suffer from bias due to social acceptability.
What about when obese women become pregnant? Catherine Spong, MD (NICHD), emphasized the danger of pregnancy for obese women, beginning with the effect of obesity on pregnancy physiology; increased blood volume and reduced pulmonary compliance can lead to a greater risk of ischemia, infarct, and heart or respiratory failure. Obesity has a multiplicative effect on the risk of venous thrombosis already increased by pregnancy. Obesity also increases insulin resistance and induces changes in inflammatory processes.4 Maternal obesity makes it more difficult for obstetricians to assess maternal weight gain and detect fetal anomalies with ultrasound. The risk of miscarriage and stillbirth are both increased. It also makes labor more complicated for the mother and infant. Consequently, the risk of cesarean section is increased and with this the risk of maternal complications due to anesthesia.
Having established that pregnancy is risky, what about the effects of hormonal contraception? Data on the effect of obesity on OC efficacy is controversial. The original reports suggesting a difference were from health care databases.5 More reliable data have come from the large phase IV studies recently commissioned in the United States and Europe. Jürgen Dinger, MD, presented data from the European Active Surveillance (EURAS) and International Active Surveillance (INAS) studies. Results from the latter study were published in January 2011 in Obstetrics & Gynecology.6 Women requesting oral contraception (switch or new start) that participated in the INAS study received a prescription from their usual health care provider and were followed prospectively with outcomes validated and adjudicated. Two interesting findings emerged from this large prospective study based on an analysis of 1634 unintended pregnancies during 73,269 woman-years of oral contraceptive pills exposure. The most significant finding was that women that received a 24-day regimen of drospirenone and ethinyl estradiol had the lowest failure rate (2.1% at 1 year). The adjusted hazard ratio (HR) for failure was 0.7 (95% confidence interval [CI], 0.6-0.8) compared to all 21-day regimens. The large sample of prospectively accumulated data also permitted for an analysis of the effect of BMI. This demonstrated an elevation in the adjusted HR of 1.5 (95% CI, 1.3-1.8) for contraceptive failure in women with a BMI ≥ 35 compared with < 35. These data are consistent with the effect noted in epidemiologic studies of oral contraceptive failures from a large closed health plan presented by Victoria Holt, PhD (University of Washington).5 An increase in failure associated with obesity also has been seen in prospective studies of emergency contraception. A 2010 meta-analysis in Lancet of two randomized clinical trials with similar study designs demonstrated that the risk of emergency contraceptive failure increases with BMI.7 This increased risk is more pronounced with levonorgestrel than ulipristal acetate and the difference in efficacy between the two drugs widens for obese women compared to merely overweight women.
If obesity is associated with an increased risk for OC failure, what are the mechanisms? Frank Stanczyk, PhD (University of Southern California), gave an overview of steroid hormone metabolism. Some pathways favor increased steroid hormone metabolism, while others favor increased bioavailability. Other important considerations include comorbidities (such as hypertension and diabetes) and use of other medications that might interfere with metabolism. Recently published studies by Alison Edelman, MD (Oregon Health & Science University), and Carolyn Westhoff, MD (Columbia University), have investigated pharmacokinetics of OCs in obese women.8,9 The consistent findings from both studies are a delay in the time needed for contraceptive steroids to reach steady state. The vulnerable interval appears to be during the reinitiation of the pill after the hormone-free interval (HFI); failure to achieve an inhibitory level would theoretically increase the risk of development of a dominant follicle, ovulation, and conception. The prospective pregnancy data from INAS showing a reduction in failure by shortening the HFI to 4 days supports this hypothesis. However, there are no clinical trials of extended cycles or continuous use in obese women.
Behavior may matter more. Westhoff reported that in her studies, obesity was associated with a three-fold increase in the risk for total noncompliance in OC use, even when controlling for race/ethnicity. However, she was cautious to suggest that this is related to obesity, as obesity is linked with low socioeconomic status, and the latter may be a predictor of poor compliance. The bottom line for clinicians is the same. Whether an increase in failure is related to altered metabolism or poor compliance, moving to long-acting methods may reduce failure.
The flip side of efficacy is safety. Contraceptive steroids influence the clotting mechanism in complex ways, and the effect on a number of prothrombotic and anticoagulation proteins can be measured. However, there is no one surrogate marker or panel of markers that reliably predicts thrombosis risk with a product. It is fair to say that the risk is related to estrogen dose, and modulated by the progestin in a combination pill (the more androgenic, the lower the risk). While epidemiologic studies have suggested that levonorgestrel pills have the lowest incidence of venous thromboembolism (VTE), large scale prospective studies do not support these findings.10 Furthermore, levonorgestrel is a potent androgenic progestin capable of causing insulin resistance and may not be appropriate for obese women at risk for diabetes. Lowering the dose of ethinyl estradiol reduces the risk of VTE, but does this increase the risk of unscheduled bleeding and compound the problem of noncompliance and failure? Will the new estradiol pills be safer? Keep in mind that oral estradiol still influences hepatic globulins and, although E2 is less potent than ethinyl estradiol, much higher oral doses are needed. Results from an active surveillance study of this new product will be years away.
What about eliminating estrogen all together? Progestin-only pills are available, but we know that they are less forgiving of noncompliance and must be dosed on a 24-hour schedule to maintain contraceptive activity through the inhibition of cervical mucus. The compliance problem linked to obesity makes this a potential concern. Also, irregular bleeding may make them less acceptable. Are long-acting methods like implants and IUDs acceptable to obese women? We still have a lot of unanswered questions.
The goal of the conference was to determine if more research is needed (Answer: It is), and discussions are underway to develop a large-scale randomized trial to address questions of efficacy, acceptability, and safety. The question of safety will need to be resolved through prospective cohort studies like EURAS and INAS with the even larger sample sizes sufficient to address these outcomes with sufficient power.
Pending these results, clinicians must continue to make decisions about OC use in obese women. All obese women need to understand that BMI has a direct and independent correlation with VTE risk that multiplies the risk of OC use. For young, otherwise healthy, obese women, this risk may be acceptable. For sedentary obese women with other risk factors, probably not. However, in all cases, the risk associated with pregnancy will be higher. Using a long-acting, estrogen-free method such as an implant or IUD will always be the safest choice, but this may not be acceptable to all patients. For women choosing to use an OC, use the lowest estrogen dose, and consider extended (e.g., 24-day) cycle or continuous use.