Over the past 40 years, the doses of estrogen and progestin in oral contraceptive pills (OCPs) have been reduced substantially.1 Most of the OCPs currently prescribed in the United States contain 35 micrograms or less of ethinyl estradiol (E2), and formulations containing as little as 20 micrograms ethinyl E2 are available.2 However, although efficacy is comparable with higher-dose formulations, the use of OCPs containing only 20 micrograms ethinyl E2 has been associated with less favorable bleeding patterns, which may include intermenstrual bleeding or spotting or both, particularly during the first several months of use.3 Women who experience irregular bleeding often discontinue use of OCPs.4–6 A survey of more than 6,000 European women found that those who experienced bleeding problems during the first 3 months were twice as likely to stop OCP use as those who did not report bleeding problems during this time.4
It has been suggested that extending the duration of active hormonal treatment from 21 to 24 days may decrease the duration and intensity of withdrawal bleeding and reduce intermenstrual bleeding and spotting.7 Results of one head-to-head comparison of 21-day and 24-day regimens of norethindrone acetate/ethinyl E2 20 micrograms showed that, although the mean cumulative days of withdrawal bleeding over cycles 2 to 6 was less with the 24-day regimen than with the 21-day regimen, the mean number of days of unscheduled bleeding was not reduced significantly until cycle 6.7 Although the unscheduled bleeding pattern with a 21-day regimen of norethindrone acetate/ethinyl E2 20 micrograms appears to be similar to that of a 24-day regimen of norethindrone acetate/ethinyl E2 20 micrograms, using old definitions, it is inferior to that with the 21-day norgestimate/ethinyl E2 25-microgram regimen.7,8
Little has been published about bleeding patterns with another 20 micrograms ethinyl E2 OCP with a 24-day regimen and containing 3 mg of drospirenone (drospirenone/ethinyl E2 20 micrograms). Results of a noncomparative trial with a 24-day regimen of drospirenone/ethinyl E2 20 micrograms suggested a favorable bleeding profile9; however, a comparative trial of a 24-day regimen of drospirenone/ethinyl E2 20 micrograms and a 21-day regimen of desogestrel 150 micrograms/ethinyl E2 20 micrograms found no significant difference in bleeding/spotting episodes.10
The aforementioned studies used various definitions of measuring bleeding patterns that are not similar to those recently recommended by Mishell and colleagues and endorsed by the U.S. Food and Drug Administration’s (FDA) Reproductive Health Drugs Advisory Committee.11–13 The present study was undertaken to compare bleeding patterns with a 21/7-day regimen of norgestimate/ethinyl E2 25 micrograms (ORTHO TRI-CYCLEN LO, Ortho-McNeil-Janssen Pharmaceuticals, Inc., Raritan, NJ) and a 24/4-day regimen of drospirenone/ethinyl E2 20 micrograms (YAZ, Bayer HealthCare LLC, Wayne, NJ). The FDA-endorsed terminology and definitions for bleeding variables were used for this trial.12 In addition, bleeding information was collected using an interactive voice-response system.12
MATERIALS AND METHODS
This randomized, open-label, active-controlled study was conducted at 20 centers in the United States. The protocol was approved by central or local institutional review boards. Eligible participants included healthy, nonpregnant, nonlactating, sexually active women aged 18 to 45 years seeking oral hormonal contraception. Among participants aged 35 to 45 years, only those who were nonsmokers were included. Eligible participants had regular menstrual cycles, including at least one normal menstrual period (typical in duration and amount of flow for the participant) within 35 days of the first study visit, a negative Chlamydia test (urine or cervical swab), and a Pap test without evidence of moderate to severe dysplasia or any malignancy within the preceding 12 months. All eligible participants gave written informed consent. Exclusion criteria included a history or presence of disorders commonly accepted as contraindications to steroid hormone therapy, an untreated thyroid disorder, or a body mass index greater than 40 kg/m.2 Also excluded were participants who previously discontinued norgestimate/ethinyl E2 or drospirenone/ethinyl E2 owing to breakthrough bleeding, received depot medroxyprogesterone acetate or other hormonal injectables within 6 months of screening or other hormonal implants in place or removed within 60 days of screening, or had used a steroid-containing intrauterine device within 3 months before screening.
The sample size for this study was based on information from prior studies on total number of bleeding days across cycles 1 to 3.14,15 The mean for the total bleeding days was expected to be 22.2 (standard deviation 7.8) for drospirenone/ethinyl E2 and 18.9 (standard deviation 6.9) for norgestimate/ethinyl E2. Based on a mean difference of 3.3 days and a conservative pooled standard deviation of 7.8, 119 participants in each group would have 90% power to show a difference between treatment groups at an alpha level of 0.05 (two-tailed).
Using a computer-generated randomization schedule, participants were randomly assigned 1:1 to treatment with study medication for three 28-day cycles, with randomization stratified by center and recent hormonal contraceptive exposure (ie, fresh starts compared with switchers). Fresh starts were defined as women who had not been on a hormonal contraceptive for 60 days or more, and switchers were defined as those who had been on a hormonal contraceptive within the previous 60 days.
Bleeding patterns and adherence were evaluated from data recorded daily from day 1 to day 85 using an interactive voice-response system–based diary. Participants responded to the following questions: 1) Did you take your birth control pill yesterday? 2) Did you have any vaginal blood loss yesterday? If yes, 3) Please describe the blood loss (reported as light, medium, or heavy), 4) Did you use any sanitary protection (for example, tampon, pad, or pantiliner) for your blood loss yesterday? If participants missed calling on a particular day, they were contacted daily by the study site with a reminder. Participants were allowed to enter diary data into the interactive voice-response system for up to 7 days after the expected date (backfilled data). The following questions were asked for backfilled data: 1) Did you take your birth control pill on (read out day, date)? 2) Did you have any vaginal blood loss on that date? Data not entered within 7 days were classified as missing.
Bleeding was defined according to the 2007 FDA’s Reproductive Health Drug Advisory Committee–endorsed criteria with one exception—an unscheduled bleeding episode was defined using the Belsey/World Health Organization definition (see Table 1).12,16 Because days 1 through 7 of cycle 1 were excluded from the analysis of bleeding per the FDA-endorsed criteria, only those participants who took study medication for at least 7 days and had postbaseline bleeding data after day 7 were included in the analysis of cycle control.12 The primary efficacy variable was the participant’s total number of unscheduled bleeding days during cycles 1 through 3 as recorded daily using the interactive voice-response system diary. Secondary efficacy variables analyzed for each of cycles 1 to 3 and cumulatively included: 1) incidence of unscheduled bleeding, 2) mean number of unscheduled bleeding episodes, 3) incidence of scheduled bleeding, and 4) mean number of scheduled bleeding days. In the cycle analysis, unscheduled bleeding episodes may be bounded by scheduled bleeding days. Other secondary efficacy variables included the incidence of total bleeding during days 8 to 84 and participant satisfaction with the OCP regimen rated at the end of the study as very satisfied, somewhat satisfied, neither satisfied nor dissatisfied, somewhat dissatisfied, or very dissatisfied.
Adverse-event data were collected and reported according to the Guideline for Good Clinical Practice.17 Vital-signs data (blood pressure and pulse), physical examinations, and body weight measurements (including body mass index) were obtained at the screening visit and at the end of treatment. In addition, as noted in the prescribing information for drospirenone/ethinyl E2 20 micrograms, participants randomly assigned to this group who were taking any medications that might increase serum potassium levels had their level measured during the first month of concomitant medication use (between days 15 and 28 of starting study medication).
The evaluation of bleeding patterns was based on the intent-to-treat population, defined as all randomly assigned participants who took the study drug and for whom there was postbaseline bleeding data after day 7. The mean, median, and frequency of unscheduled, scheduled, and total bleeding days were presented by treatment for each cycle and across all three cycles. The number of unscheduled and scheduled bleeding days, number of unscheduled bleeding episodes, and overall participant satisfaction were compared between the two treatment groups using the Mann-Whitney-Wilcoxon test. The incidence of scheduled and unscheduled bleeding in the two treatment groups was compared using Fisher exact test. The extent of exposure to study medication in each treatment group was determined by calculating the number of days exposed to study medication using the date of the last and first treatment day. Safety was evaluated in all participants who took the study medication and provided postbaseline safety data. Treatment-emergent adverse events in both treatment groups were summarized. Mean changes in blood pressure and body weight in the two treatment groups were compared by a two-sample t test. A post hoc analysis of the primary endpoint was performed in which the demographic variables smoking and weight were incorporated separately as covariates.
The participant flow diagram shows the number of participants enrolled in the study and their disposition (Fig. 1). Between May 27, 2008, and June 30, 2008, a total of 355 women were assigned randomly to either a 21/7-day norgestimate/ethinyl E2 (n=178) or a 24/4-day drospirenone/ethinyl E2 (n=177) OCP regimen. Eleven women in the 21/7-day group and 10 women in the 24/4-day group failed to meet inclusion criteria. All 334 women who took the study medication (n=167 in each study group) were included in the safety analysis.
The treatment groups were comparable in age, racial distribution, and previous contraceptive history (Table 2). More than 50% of participants in each treatment group had used hormonal contraception within the 60 days before enrollment. Participants in the 24/4-day drospirenone/ethinyl E2 group weighed more and had a greater body mass index than did participants in the 21/7-day norgestimate/ethinyl E2 group. There were twice as many smokers in the 24/4-day drospirenone/ethinyl E2 group compared with the 21/7-day norgestimate/ethinyl E2 group (11 participants compared with 22 participants). However, the majority of participants in each treatment group were nonsmokers.
There were 165 efficacy-evaluable participants in the 21/7-day norgestimate/ethinyl E2 group and 167 in the 24/4-day drospirenone/ethinyl E2 group. Two participants in the 21/7-day norgestimate/ethinyl E2 group were excluded from the efficacy analysis because bleeding data beyond day 7 were not available. The mean number of days exposed to the study drug was 80.9±14.82 in the 21/7-day norgestimate/ethinyl E2 group and 82.0±11.33 in the 24/4 drospirenone/ethinyl E2 group.
More than 95% of participants in both groups called in one or more data responses late (backfilled within 7 or fewer days of scheduled date). However, fewer than 5% of participants in both groups had any missing data. Of more than 13,000 data points in each treatment group, approximately 19% were backfilled and fewer than 0.1% of the data points in either group were missing. Hence, use of the interactive voice-response system resulted in complete data for all but 15 participants, seven in the norgestimate/ethinyl E2 group and eight in the drospirenone/ethinyl E2 group.
Bleeding data for the two treatment groups are summarized in Table 3. The mean number of days of unscheduled bleeding in cycles 2 and 3 and across all three cycles was significantly greater in the 24/4-day drospirenone/ethinyl E2 group compared with the 21/7-day norgestimate/ethinyl E2 group (P=.005 and P=.001, cycle 2 and 3, respectively, P=.003 cumulatively). Over the three cycles, participants in the 21/7-day norgestimate/ethinyl E2 group experienced approximately 1.5 fewer days of unscheduled bleeding than did those in the 24/4-day drospirenone/ethinyl E2 group (4.6 compared with 6.1 days). Cumulatively, 55 (33.3%) participants in the 21/7-day norgestimate/ethinyl E2 group and 29 (17.4%) participants in the 24/4-day drospirenone/ethinyl E2 group did not experience any unscheduled bleeding episodes (Fig. 2). The number of unscheduled bleeding episodes was significantly less in the 21/7-day norgestimate/ethinyl E2 group than in the 24/4-day drospirenone/ethinyl E2 group over the course of three cycles (1.47 compared with 2.01, P=.001). In a post hoc analysis controlling for weight and smoking separately, no difference from the primary endpoint finding was seen.
Participants in the 21/7-day norgestimate/ethinyl E2 group experienced more days of scheduled bleeding than did those in the 24/4-day group throughout the cycles; however, data collection stopped on day 28 of cycle 3, resulting in truncation of results on the duration of scheduled bleeding (Table 3). The mean number of days of scheduled bleeding was significantly greater in the 21/7-day norgestimate/ethinyl E2 group than in the 24/4-day drospirenone/ethinyl E2 group at cycle 1 (4.3 compared with 3.2 days, P<.001) and at cycle 2 (4.0 compared with 2.8 days, P<.001). The percentage of women who reported scheduled bleeding was higher in the 21/7-day norgestimate/ethinyl E2 group than in the 24/4-day drospirenone/ethinyl E2 group at cycle 1 (89% compared with 73%, respectively) and at cycle 2 (85% compared with 62%, respectively). Among the subset of participants who reported scheduled bleeding (responder analysis) at cycles 1 and 2 (n=147 and 136, respectively, for the 21/7-day norgestimate/ethinyl E2 group and 122 and 102, respectively, for the drospirenone/ethinyl E2 group), the mean number of scheduled bleeding days appeared to be similar between the two OCP regimens. Because the data were truncated at cycle 3, the results for this cycle cannot be interpreted. Absence of scheduled bleeding was significantly lower (P<.001) for the 21/7-day norgestimate/ethinyl E2 group compared with the 24/4-day drospirenone/ethinyl E2 group in each cycle (10.9% compared with 26.9% at cycle 1, 15% compared with 38.2% at cycle 2, 21% compared with 41.5% at cycle 3). These findings were not affected by data truncation at cycle 3. Overall, 84% (134 of 159) of participants in the 21/7-day norgestimate/ethinyl E2 group and 91.0% (147 of 162) of participants in the 24/4-day drospirenone/ethinyl E2 group were somewhat to very satisfied with their OCP regimen (P=.07).
The changes from baseline in systolic and diastolic blood pressure and weight were not significantly different for the two treatment groups. Mean weight increased 0.7±5.49 lb in the 21/7-day norgestimate/ethinyl E2 group (n=160) and 0.4±6.15 lb in the 24/4-day drospirenone/ethinyl E2 group.
Treatment-emergent adverse events were reported by 34 (20.4%) of 167 participants in each treatment group. Two participants in the 21/7-day norgestimate/ethinyl E2 group reported serious adverse events that were not considered to be related to the study drug: one participant had a lumbar disc protrusion, and one subject had a uterine leiomyoma, ruptured ovarian cyst, and pelvic hemorrhage leading to study discontinuation. One participant in the 24/4-day drospirenone/ethinyl E2 group had a pulmonary embolism that was considered to be possibly related to study drug and discontinued the study. Pregnancy occurred in cycle 3 while on the study drug in one participant in the 21/7-day norgestimate/ethinyl E2 group. No participants in the 24/4-day drospirenone/ethinyl E2 group became pregnant during the study period. One participant in the 21/7-day norgestimate/ethinyl E2 group discontinued treatment owing to nausea and vomiting considered to be probably related to treatment. In the 24/4-day drospirenone/ethinyl E2 group, treatment was discontinued by one participant owing to irritability and by another owing to mood swings, both of which were considered to be possibly related to treatment.
In this study, we prospectively compared patterns of bleeding with two different OCP regimens using the 2007 FDA-endorsed criteria and employed collection of daily compliance and bleeding data using an interactive voice-response system as also recommended by the FDA. We found less unscheduled bleeding among users of the 21/7-day norgestimate/ethinyl E2 OCP compared with users of the 24/4-day drospirenone/ethinyl E2 OCP in the first three cycles. Cumulatively, participants using the 21/7-day OCP regimen experienced approximately 1.5 fewer days of unscheduled bleeding than did those using the 24/4-day OCP regimen. Moreover, nearly twice as many participants in the 21/7-day norgestimate/ethinyl E2 group than in the 24/4-day drospirenone/ethinyl E2 group reported no episodes of unscheduled bleeding during the 84-day trial period.
Although clinicians may be more familiar with the commonly presented breakthrough-bleeding analyses, the FDA-endorsed criteria provide a standard by which all hormonal contraceptives can be assessed more objectively. Compared with the more common practice of paper diaries, the interactive voice-response system is easier to use and offers fast data turnaround and high participant compliance, therefore representing a superior technology for collecting daily diary data. The interactive voice-response system represents an important and low-cost advance in diary collection tools. The high compliance with both pill taking and diary completion seen in this study may be attributed to the ease of use of the interactive voice-response system. This high compliance also may have contributed to the higher overall reported rates of bleeding seen in this study.
The FDA-endorsed bleeding methodology applied in this study has been shown to result in higher reported rates of unscheduled bleeding.18 The bleeding data from this study also were analyzed using definitions from an earlier 21/7-day norgestimate/ethinyl E2 25-microgram study.8,12 In this earlier study, breakthrough bleeding/spotting was defined as “bleeding and spotting occurring during the active pill-taking interval, excluding bleeding contiguous with menses.”8 In addition, breakthrough bleeding was limited to “bleeding requiring sanitary protection of more than one pad or tampon on any day.”8 In the present trial, unscheduled bleeding was defined as blood loss that occurred while taking active hormones, with the exception of blood loss beginning during the hormone-free interval and continuing through days 1 to 4 of the subsequent active cycle (see Table 1).
Results of the two analyses are compared in Figure 3. Using the FDA-endorsed criteria resulted in higher reported rates of unscheduled bleeding than did the analysis using the old criteria for breakthrough bleeding or spotting.8,12 Hence, any bleeding or spotting that began during active pills and continued into the placebo days or any bleeding or spotting that began during the placebo days and continued into the next cycle of active pills was classified as withdrawal bleeding and not breakthrough bleeding.8 Both analyses found a significantly higher incidence of unscheduled bleeding or breakthrough bleeding/spotting with the 24/4-day drospirenone/ethinyl E2 OCP during cycles 2 and 3 compared with the 21/7-day norgestimate/ethinyl E2 OCP. Women do not expect to bleed while taking active hormones; however, those taking cyclical regimens do expect to experience bleeding when active hormones are withdrawn. Although many women can adjust easily to expect bleeding during the first days of their next pill pack, these differences should be addressed, particularly when counseling fresh-start patients.
More scheduled bleeding days were expected in the norgestimate/ethinyl E2 group based on the longer hormone-free interval with the 21/7 OCP regimen.7 In a responder analysis, the subset of participants who reported scheduled bleeding exhibited a similar duration of scheduled bleeding. The high incidence of absence of scheduled bleeding among women receiving drospirenone/ethinyl E2 in this study was an unexpected finding.
A limitation of this study is the duration. The FDA-endorsed criteria recommend a six-cycle study for understanding bleeding patterns. The three-cycle study reported here is appropriate to characterize cycle-control patterns; a six-cycle study would have provided more information on bleeding over time. However, a six-cycle study may have resulted in higher dropout rates because of the burden of daily interactive voice-response system reporting.
In addition, because of the three-cycle duration, a full picture of scheduled bleeding cannot be understood from this study. The data collection stopped after day 28 of cycle 3 (day 84 of the reference period). Because the FDA-endorsed criteria consider days 1 to 4 of active pills part of scheduled bleeding (if contiguous with menses), we were unable to complete our collection in cycle 3. Although the primary endpoint of this study was unscheduled bleeding, it is our recommendation to add an additional week of observation to any study on cycle control to account fully for scheduled bleeding in the final cycle.
Finally, both the FDA-endorsed criteria and the unprecedented use of an interactive voice-response system in a study on cycle control may have resulted in higher reported rates of bleeding than in previous studies. In conclusion, in a three-cycle study, the 21-day norgestimate/ethinyl E2 regimen resulted in less unscheduled bleeding and more scheduled bleeding compared with the 24-day drospirenone/ethinyl E2 regimen. These findings can be a helpful guide for the clinician in counseling patients on what to expect when starting these 21-day or 24-day OCP regimens.
1.Blumenthal PD, Edelman A. Hormonal contraception. Obstet Gynecol 2008;112:670–84.
2.Kaunitz AM. Clinical practice. Hormonal contraception in women of older reproductive age. N Engl J Med 2008;358:1262–70.
3.Gallo MF, Nanda K, Grimes DA, Lopez LM, Schulz KF. 20 microg versus >20 microg estrogen in combined oral contraceptives for contraception. The Cochrane Database of Systematic Reviews 2008, Issue 4. Art. No.: CD003989. DOI: 10.1002/14651858.CD003989.pub3.
4.Rosenberg MJ, Waugh MS, Meehan TE. Use and misuse of oral contraceptives: risk indicators for poor pill taking and discontinuation. Contraception 1995;51:283–8.
5.Rosenberg MJ, Waugh MS. Oral contraceptive discontinuation. A prospective evaluation of frequency and reasons. Am J Obstet Gynecol 1998;179:577–82.
6.Sabatini R, Cagiano R. Comparison profiles of cycle control, side effects and sexual satisfaction of three hormonal contraceptives. Contraception 2006;74:220–3.
7.Nakajima ST, Archer DF, Ellman H. Efficacy and safety of a new 24-day oral contraceptive regimen of norethindrone acetate 1 mg/ethinyl estradiol 20 μg (Loestrin 24 Fe). Contraception 2007;75:16–22.
8.Hampton RM, Short M, Bieber E, Bouchard C, Ayotte N, Shangold G, et al. Comparison of a novel norgestimate/ethinyl estradiol contraceptive (Ortho Tri-Cyclen Lo) with the oral contraceptive Loestrin Fe 1/20. Contraception 2001;63:289–95.
9.Bachmann G, Sulak PJ, Sampson-Landers C, Benda N, Marr J. Efficacy and safety of a low-dose 24-day combined oral contraceptive containing 20 μg ethinylestradiol and 3 mg drospirenone. Contraception 2004;70:191–8.
10.Klipping C, Marr J. Effects of two combined oral contraceptives containing ethinyl estradiol 20 μg combined with either drospirenone or desogestrel on lipids, hemostatic parameters and carbohydrate metabolism. Contraception 2005;71:409–16.
11.Mishell DR Jr, Guillebaud J, Westhoff C, Nelson AL, Kaunitz AM, Trussell J, et al. Combined hormonal contraceptive trials: variable data collection and bleeding assessment methodologies influence study outcomes and physician perceptions. Contraception 2007;75:4–10.
12.Mishell DR Jr, Guillebaud J, Westhoff C, Nelson AL, Kaunitz AM, Trussell J, et al. Recommendations for standardization of data collection and analysis of bleeding in combined hormone contraceptive trials. Contraception 2007;75:11–5.
13.U.S. Food and Drug Administration. Advisory Committee for Reproductive Health Drugs. Food and Drug Administration Center for Drug Evaluation and Research Volumes I and II. Available at: http://www.fda.gov/ohrms/dockets/ac/cder07.htm
. Retrieved May 6, 2009.
16.Belsey EM, Machin D, d’Arcangues D. The analysis of vaginal bleeding patterns induced by fertility regulating methods. Contraception 1986;34:253–60.
17.International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use. ICH Harmonised Tripartite Guideline. Guideline for Good Clinical Practice E6(R1). Current Step 4 version. Available at: http://www.emea.europa.eu/pdfs/human/ich/013595en.pdf
. Retrieved June 7, 2009.
© 2009 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.
18.Hampton RM, Fisher AC, Pagano S, LaGuardia KD. Scheduled and unscheduled bleeding patterns with two combined hormonal contraceptives: application of new recommendations for standardization. Fertil Steril 2009;92:434–40.