Objective
Many observational studies indicate higher oral contraceptive failure among obese women, but most clinical trials and physiologic studies do not support these differences. Limited data indicate higher failure rates among obese contraceptive patch users. Data regarding contraceptive vaginal ring performance in obese women are needed.
Study Design
Twenty normal weight (body mass index [BMI] 19.0-24.9; median, 21.65) and 20 obese (BMI 30.0-39.9; median, 33.7) women enrolled in a prospective study of ethinyl estradiol (EE 2 ) and etonorgestrel pharmacokinetics and of ovarian follicle development, endometrial thickness, and bleeding patterns, all measured biweekly during the second cycle of contraceptive vaginal ring use.
Results
Thirty-seven women completed follow-up. Mean day 0-21 EE 2 concentrations were lower among obese vs normal weight women (15.0 vs 22.0 pg/mL, respectively, P = .004), whereas etonorgestrel concentrations were similar (1138 vs 1256 pg/mL, respectively, P = .39). Follicular development was minimal in both groups, with only 5 women achieving a maximum follicle diameter >13 mm at any time during 3 weeks follow-up (3 normal weight and 2 obese women); these women had serum progesterone levels <1.0. Obese women reported more bleeding or spotting than normal weight women (3.6 vs 1.4 days, respectively, P = .01).
Conclusion
Although obese women had lower EE 2 levels during contraceptive vaginal ring use, they had excellent suppression of ovarian follicle development, similar to normal weight women. This predicts that contraceptive vaginal ring effectiveness will be similar in women with a BMI up to 39.9. The lower serum EE 2 levels in the obese women may explain the greater reported bleeding or spotting days.
Several observational studies have reported that heavier women had higher failure rates during use of combination oral contraceptives (OC), and one clinical trial identified higher failure rates for heavier women using the contraceptive patch. In contrast, several more recent prospective studies found little or no effect of weight on OC failure. Only one analysis has evaluated the relationship between contraceptive vaginal ring (CVR) failure rates and weight; that study found no effect of weight on CVR failure rates, but it included too few obese women to provide a precise answer. A recent physiologic study of compliant OC users identified some differences in the pharmacokinetics of the contraceptive hormones between the normal weight and obese women ; however, normal weight and obese women experienced no differences in ovarian suppression or ovulation during OC use. The goal of the current study was to compare pharmacokinetics and ovarian suppression during CVR use in normal weight and obese women.
Materials and Methods
This clinical trial compared ovarian follicle development and serum ethinyl estradiol and etonogestrel levels between normal weight and obese women using the CVR (daily release of 15 mcg ethinyl estradiol [EE 2 ] and 120 mcg etonogestrel [ENG]). Participant-related activities were conducted between July and December 2008. The Columbia University Institutional Review Board approved the study, and all participants gave informed consent. We recruited participants from a cohort of women who had participated in a similar previous study of OCs several months before enrollment in this study. The women were initially recruited via advertisements online, in newspapers, and fliers.
Eligible women were aged 18-35 years with a recent history of regular, spontaneous menstrual cycles, were willing to use the CVR and commit to at least 8 biweekly study visits during the second ring cycle. We excluded women with any medical contraindications to the use of combined hormonal contraception based on the WHO Medical Eligibility Criteria and only enrolled women in WHO Category 1. Women using medications known to affect the CYPp450 system were ineligible. All participants had normal-appearing ovaries on a baseline sonogram using a TITAN (SonoSite, Inc, Bothell, WA) with a 7.5 MHz transvaginal probe, and were either normal weight (body mass index [BMI, kg/m 2 ] 19.0-24.9) or obese (BMI 30.0-39.9) based on standardized height and body weight measurements on the day of enrollment. We measured body weight (kg) using the BC-418 (Tanita Corp, Tokyo, Japan), body composition analyzer, and used the same machine for all measurements.
Participants received 2 CVRs and completed 1 ring cycle (21 days with continuous ring use, followed by 7 ring-free days) before the planned study cycle. They inserted the second ring on day 1 of the study cycle. Participants then underwent biweekly visits for venipuncture to obtain samples to measure EE 2 and ENG; we collected these samples at approximately days 3, 6, 9, 12, 15, 18, and 21 ± 1 day. Specimens were allowed to clot for at least 10 minutes at room temperature, and then separated at 3400 RPM by centrifuge. The serum was transferred to clean tubes and stored in aliquots at −80°C until analysis. For this study, the Biomarkers Core Laboratory of the Irving Institute of Clinical and Translational Research at Columbia University Medical Center developed ENG and EE 2 assays using the tandem mass spectrometry. In short, ENG and EE 2 were measured in serum by UPLC/MS/MS after liquid/liquid extraction using D8-Progesterone or D4-EE2 as the internal standards (IS) for ENG and EE 2 , respectively. EE 2 was derivatized with dansyl chloride before analysis. The steroids were quantified by positive electrospray ionization in multiple reaction monitoring mode using the Waters Xevo TQ-S system (Waters, Milford, MA). The method was linear between 50 and 2000 pg/mL and 1 and 100 pg/mL for ENG and EE 2 , respectively (LOQ: 50 pg/mL and 1 pg/mL). The intra- and interassay coefficient of variation were < 6% and < 13%, respectively, for ENG; <3.9% and <4.4%, respectively, for EE 2 .
Participants also underwent biweekly vaginal sonograms during use of the second ring to measure ovarian follicle-like structures using the same Sonosite TITAN with a 7.5 MHz transvaginal probe; we measured follicle diameter (FD) in 2 perpendicular diameters and recorded the dimensions of all follicles with a mean diameter of at least 8 mm. We also noted a corpus luteum or ruptured follicle, if present, and measured anterior-posterior endometrial thickness. Each sonogram took place on the same day as the venipunctures. We measured serum progesterone for all participants who had any ovarian follicle with a diameter of 13 mm or greater. At all visits the study physician confirmed that the CVR was present in the vagina. Participants were compensated for their time and travel costs.
Participants recorded daily bleeding and spotting throughout the study using paper diaries. The loss of blood requiring the use of a sanitary pad or tampon was noted as “bleeding” and the loss of blood requiring a panty liner or no protection as “spotting.” Participants reported the total number of bleeding and spotting days, both scheduled and unscheduled, and the investigators did not impose a definition of which days were scheduled and which were unscheduled.
We compared serum ENG and EE 2 levels using the area under the concentration-time curve (AUC) from days 0 to 21 using the linear trapezoidal approximation, assuming that the day 0 level was 0 for both hormones (as by day 0 97% of EE 2 has been eliminated and over 99% of ENG). We also compared the geometric mean serum levels for both hormones in normal weight vs obese women at the end of weeks 1, 2, and 3 using the log-rank test. We assessed ovarian suppression in this study as the proportion of participants in each BMI group who had a maximum FD exceeding prespecified thresholds, 8 mm or 13 mm. We used χ 2 tests and Fisher exact test as appropriate to assess the association between obesity and maximum FD. For each participant, we calculated mean endometrial thickness and number of bleeding and spotting days during the study cycle, and used a t test to compare these variables between the BMI groups.
A final sample size of 17 participants in each group was planned a priori to have 80% power to identify a 1 standard deviation difference in the mean serum levels of the contraceptive hormones, based on expected values for normal weight women.
Results
We enrolled 40 women into the study. Before the study cycle 1 woman fractured her pelvis in a motor vehicle accident and another experienced a venous thromboembolism following a long-haul flight (flight duration >4 hours). We excluded both of these women from continuation in the study. We also excluded 1 woman because of removal of the CVR during the study cycle, leaving 18 normal weight and 19 obese women for analysis. Table 1 presents the baseline characteristics of the participants, whose BMI differed by design, and who differed somewhat by ethnicity.
| Variable | Overall (n = 37) | Normal weight (n = 18) | Obese (n = 19) | P value |
|---|---|---|---|---|
| Age | 26.6 ± 4.5 | 27.1 ± 4.7 | 26.2 ± 4.3 | .6 |
| Hispanic | 14 (38) | 4 (22) | 10 (53) | |
| Non-Hispanic black | 12 (32) | 6 (33.5) | 6 (31) | |
| Non-Hispanic white | 9 (24.5) | 6 (33.5) | 3 (16) | .15 |
| Asian | 2 (5.5) | 2 (11) | 0 (0) | |
| Age menarche | 12.0 ± 1.9 | 12.4 ± 1.8 | 11.6 ± 1.9 | .3 |
| Ever pregnant | 18 (49) | 9 (50) | 9 (47) | |
| Never pregnant | 19 (51) | 9 (50) | 10 (53) | > .99 |
| Current smoker | 5 (41) | 2 (11) | 3 (16) | |
| Past or never smoker | 32 (86) | 16 (89) | 16 (84) | .63 |
| Height, cm | 164.4 ± 5.7 | 162.6 ± 5.2 | 166.1 ± 5.7 | .1 |
| Weight, kg | — | 56.9 ± 5.5 | 94.7 ± 10.4 | < .0001 |
| BMI | — | 21.5 ± 1.3 | 34.3 ± 3.0 | < .0001 |
The Figure shows the mean EE 2 and ENG levels during CVR use. We did not measure the serum levels immediately before ring insertion, and assume the day 0 levels to be 0 for both hormones. The day 0-21 average concentration for EE 2 was 22.0 for normal weight women and 15.0 for obese women ( P = .004). The day 0-21 average concentration for ENG was 1256 for normal weight women and 1138 for obese women ( P = .39). The EE 2 levels were lower in the obese participants during all 3 weeks of CVR use; however, the ENG levels were similar between the normal weight and obese participants ( Table 2 ).
| Variable | Normal weight Mean (ng/L, SD 1 ) | Obese Mean (ng/L, SD) | P value |
|---|---|---|---|
| EE | |||
| Week 1 | 21.8 (17.5, 27.1) | 14.8 (12.5, 17.5) | .008 |
| Week 2 | 23.5 (19.6, 28.2) | 14.9 (12.5, 17.8) | .001 |
| Week 3 | 21.9 (17.9, 26.6) | 14.8 (12.7, 17.3) | .004 |
| ENG | |||
| Week 1 | 1349 (1179, 1542) | 1190 (968, 1464) | .35 |
| Week 2 | 1360 (1196, 1547) | 1311 (1106, 1553) | .75 |
| Week 3 | 1275 (1116, 1457) | 1240 (1041, 1477) | .39 |
All participants underwent biweekly transvaginal sonograms during CVR use. As shown in Table 3 , during week 1 (which immediately followed the ring-free week), 30% of the women had a largest follicle with a mean FD >8 mm; during weeks 2 and 3, respectively, only 14% and 11% of participants had follicles >8 mm. During all 3 weeks of CVR use, only 5 participants (11%) had a largest follicle with a mean FD >13 mm and no corpora lutea were observed. All serum progesterone levels were <1.0 ng/mL. Maximum FDs were similar between the normal weight and obese participants. During follow-up, the mean endometrial thickness was 3.9 mm (standard deviation [SD] 1.6) in the normal weight participants, and 4.7 mm (SD 1.8) in the obese participants ( P = .19).
| Maximum follicular diameter | Overall n = 37 | Normal weight n = 18 | Obese n = 19 | P value |
|---|---|---|---|---|
| Week 1 | ||||
| ≥8 mm | 11 (30) | 5 (28) | 6 (32) | .80 |
| ≥13 mm | 4 (11) | 3 (17) | 1 (5) | .26 |
| Week 2 | ||||
| ≥8 mm | 5 (14) | 4 (22) | 1 (5) | .13 |
| ≥13 mm | 4 (11) | 3 (17) | 1 (5) | .26 |
| Week 3 | ||||
| ≥8 mm | 4 (11) | 2 (11) | 2 (11) | .95 |
| ≥13 mm | 4 (11) | 2 (11) | 2 (11) | .95 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
