Contraception
Ila Dayananda
S. Jean Emans
Alisa Goldberg
Providing targeted sexuality education and contraceptive counseling for adolescents is a key component of the clinical visit. Adolescence is often a time when young people experiment. By the time they reach the age of 19, 7 in 10 teenagers will have had sexual intercourse (1). The 2009 Youth Risk Behavioral Surveillance report (2) found that 46% of female high school students had had sexual intercourse (29% of 9th-grade girls to 65% of 12th-grade girls). By 12th grade, 19% of girls had had four or more partners. Adolescents who become sexually active in their early teens have more lifetime partners and are often involved in other risk-taking activities such as substance abuse and smoking. Given that 40% of sexually active teenagers do not use condoms, this population is at an increased risk for sexually transmitted infections (STIs) and pregnancy. A sexually active adolescent who does not use contraception has a 90% chance of being pregnant in 1 year (3). Approximately 82% of teenage pregnancies are unintended, and teenage pregnancy and abortion rates are higher in the United States than in other developed countries (4). Nearly 800,000 teenagers in the United States become pregnant every year, and nearly one-third of these pregnancies end in abortion (5). These statistics emphasize the importance of continued improvement in sex education and access to contraceptive services. Correct and effective use of contraception by adolescent girls and young women is dependent on a number of factors, including the personal characteristics of the user (age, educational level, socioeconomic status, parity, and social context), the characteristics of the method (dosage, side effects, ease of use, and cost), and the service system providing the method (access, counseling, provider characteristics, educational methods, and choices offered) (6). Too often clinicians think only of the user without reflecting on the other two important areas that have a major impact on compliance.
Adolescents and Contraception
Overall, the number of teenage pregnancies has declined over the past two decades, from a peak in 1990 to a low of 69.5 per 1,000 females in 2005. This trend is encouraging and can be attributed to an increase in contraceptive use, long-acting reversible contraception (LARC), and delay in sexual activity. Eight-three percent of female adolescents reported using contraception at last intercourse (5). Sexually active teenagers are more likely to seek contraception if they are older and have a positive outlook toward contraception, long-term plans, educational goals, and/or support from their social network to use contraception (7,8).
Of the adolescents who are using contraception, over 50% rely on the combined oral contraceptive pill and close to 30% rely on condoms as their primary method of contraception (9). The fact that sexually active adolescents are using highly effective methods such as subdermal implants and IUDs, so-called LARC, should have a positive impact on teen pregnancy rates. Many teenagers who begin using contraception are consistent users, and over two-thirds of adolescents report long-term uninterrupted contraceptive use (10). However, studies have shown that most adolescents seeking contraception do so months after initiating sexual activity. Teenagers who do not use a contraceptive method at first intercourse are twice as likely to become pregnant as those who do use a method at coitarche (11). It is important to highlight that there is no evidence that provision of contraception leads to adolescents having sex earlier or more frequently. Educating adolescents about postponement of sexual intercourse and contraceptive methods is an important part of office practice.
Contraceptive Counseling
In the practice setting, it is important to build rapport with a teenager and ensure a climate of confidentiality. An open, interactive discussion about healthy sexuality and prevention of pregnancy and sexually transmitted infection is an important part of the adolescent visit (12). When seeing a teenager who is not anticipating becoming sexually active in the near future, abstinence should be reinforced as a healthy choice. It is important to develop a direct confidential relationship with the patient, and if possible interact with both the adolescent and her parents to obtain a detailed history.
Adolescents should be asked about performance in school, family life, and risk-taking behaviors. History taking should include questions to evaluate for medical contraindications to hormonal methods and identify lifestyle factors that may influence contraceptive compliance. Blood pressure and weight and height should be measured and body mass index (BMI) calculated. Combined hormonal contraceptive methods are contraindicated in adolescents who are at high risk for thromboembolism or who have migraines with accompanying auras, uncontrolled hypertension, a history of complicated valvular or ischemic heart disease, liver adenoma/hepatoma/uncompensated cirrhosis or breast cancer (13,14). Further, the adolescent should be asked what, if any, contraceptive methods she has used in the past, what her experience has been with those methods, what she has heard about various methods, what she is worried about, and which methods she feels she can reliably use (Table 24-1). It is important to gain the adolescent’s trust, value her opinion, obtain her verbal consent, and ensure her understanding of risks and benefits before prescribing contraception. Currently 21 states and the District of Columbia allow all minors to give informed consent to receive contraceptive services without requiring parental permission (see Chapter 32). In an additional 21 states, adolescents who are emancipated or “mature minors” can also give informed consent to receive contraception without parental permission. Four states have no relevant policy or law (15). Each clinician should be familiar with his or her state laws. Teens should be
encouraged to open a dialogue with their parents. Many teens can share the need for birth control with one or both parents. The teen who does not feel comfortable sharing the information about birth control with her parents still needs to think through with the clinician what she would do if her parents found the contraceptive pills or device. In volatile situations, the clinician should offer to be the mediator to help both sides come to a solution around responsible sexuality. The clinician can facilitate communication by asking one or both of the parents and the daughter to come in for counseling at the end of the office day. Each health care provider has his or her own style of communication, but it is helpful to empathize with and acknowledge the parents’ genuine concerns for their daughter’s health. Additionally, the clinician can discuss with parents the issues faced by teens such as media messages about sexuality and the challenges of establishing healthy relationships. Parents should be congratulated on raising a daughter who is being responsible about contraception instead of becoming pregnant, as happens with so many of her peers. Acknowledge to parents how difficult it is to raise adolescents, and listen to their issues. This can facilitate solutions acceptable to both the parent(s) and the adolescent.
encouraged to open a dialogue with their parents. Many teens can share the need for birth control with one or both parents. The teen who does not feel comfortable sharing the information about birth control with her parents still needs to think through with the clinician what she would do if her parents found the contraceptive pills or device. In volatile situations, the clinician should offer to be the mediator to help both sides come to a solution around responsible sexuality. The clinician can facilitate communication by asking one or both of the parents and the daughter to come in for counseling at the end of the office day. Each health care provider has his or her own style of communication, but it is helpful to empathize with and acknowledge the parents’ genuine concerns for their daughter’s health. Additionally, the clinician can discuss with parents the issues faced by teens such as media messages about sexuality and the challenges of establishing healthy relationships. Parents should be congratulated on raising a daughter who is being responsible about contraception instead of becoming pregnant, as happens with so many of her peers. Acknowledge to parents how difficult it is to raise adolescents, and listen to their issues. This can facilitate solutions acceptable to both the parent(s) and the adolescent.
Table 24-1 Questions to Pose to Teens Choosing a Contraceptive Method | |||||||||
|---|---|---|---|---|---|---|---|---|---|
|
Among the barriers to seeking contraception are the fear of the pelvic examination and misperceptions about the side effects of hormonal contraceptive methods (16). The pelvic examination is not required for the initiation of hormonal contraception in adolescents and tests for STIs can be done using urine or self-obtained vaginal swabs (see Chapter 18) (17). Common misconceptions surrounding weight gain, fertility, cancer, and risk of infection should be addressed. The risks associated with pregnancy and STIs should be weighed against the potential side effects of contraceptive methods.
Contraceptive effectiveness and evidence-based guidelines
The full range of contraceptive options should be presented to the adolescent patient. As always, the use of condoms to help prevent STIs should be encouraged and the availability and use of emergency contraception (EC) should be explained. The best method of contraception is one that the adolescent patient will correctly and consistently use. If the patient has effectively used contraception in the past, the same method may be employed. The importance of consistent use should be discussed with an emphasis on failure rates and pregnancy risk.
Contraceptive effectiveness is a measure of the probability of pregnancy in a given interval with the use of that method. Failure rates of contraceptive methods are often expressed by the Pearl Index, which describes the number of pregnancies in 100 women-years of contraceptive use. Theoretical effectiveness represents failure rates in a highly motivated population, such as in a clinical trial, and reflects perfect use. Use effectiveness reflects failure rates with typical use as can be expected in the general population. Table 24-2 outlines typical and perfect use failure rates for each contraceptive method during the first year of use. For some teenagers, the pregnancy rate with the combined oral contraceptive (COC) may climb to 9 to 12 pregnancies per 100 woman-years or even higher because of missed pills, frequently underreported (18), and misunderstanding the directions.
In order to assess the risks and benefits of various methods with the individual patient, the clinician needs to be knowledgeable about the CDC U.S. Medical Eligibility Criteria for Contraceptive Use, 2010 (13). These guidelines are adapted from the World Health Organization (WHO) Medical Eligibility Criteria for Contraceptive Use (14), a set of comprehensive, evidence-based guidelines developed after a critical review of the medical literature. Although some domains lack rigorous trials and evidence, this comprehensive approach has been widely adopted (13,14,18,33). Category levels 1 to 4 are noted in Table 24-3. The WHO and CDC classify a medical condition as category 3 when the “theoretical or proven risks usually outweigh the advantages of using the contraceptive method,” and category 4 when there is “an unacceptable health risk if the contraceptive method is used.” (14). As an example, Table 24-4 lists category 3 and 4 conditions for the use of combined hormonal contraceptives. In the United States, prescribing a contraceptive method for a woman with a category 3 condition for that method requires clinical judgment and access to close follow-up care. While U.S. Medical Eligibility Criteria for Contraceptive Use 2010 (13) is based on the WHO guidelines, some recommendations were modified for use in the United States, including recommendations about contraceptive use for women with venous thromboembolism, valvular heart disease, ovarian cancer, and uterine fibroids and for postpartum and breastfeeding women. Recommendations were added to the U.S. guidance for women with rheumatoid arthritis, history of bariatric surgery, peripartum cardiomyopathy, endometrial hyperplasia, inflammatory bowel disease, and solid organ transplantation. Table 24-4 also includes some of the changes in classification in the CDC guidelines for combined hormonal contraceptives, and includes category 3 and 4 recommendations for specific medical conditions (full CDC guidelines on contraceptive eligibility can be found in Appendix 1).
Contraceptive Initiation
Instructions on use of the chosen contraceptive method should be simple and clear, and demonstrations of actual packages or devices should be used as possible when providing information on use. Hormonal contraceptive methods can be started on any day of the menstrual cycle. Popular approaches are “Quick Start,” “Sunday Start,” and “first-day start.” With the Quick Start approach, the adolescent starts the contraceptive method on the day she is given the prescription, as long as pregnancy is reasonably excluded (19); 7 days of backup contraception should
be used. This approach increases the 3-month continuation rate but has not been shown to increase long-term continuation rates or decrease pregnancy rates; however, it may result in greater patient satisfaction (19,20). An alternative is the Sunday Start approach, in which the patient starts the contraceptive method on the Sunday after menses begins. This method was popularized as many oral contraceptive pills are packaged for initiation on Sunday to avoid a withdrawal bleed over the weekend. First-day start is another approach. If the hormonal contraceptive method is started within 5 days of the onset of menses, backup contraception is not necessary (21). However, to keep instructions standardized and provide teenagers the best protection from pregnancy, we advise that backup contraception should be used for 7 days when starting hormonal contraception. For the teen who has been successfully using condoms and now wishes to use hormonal contraception, discussion about the need for continuing protection against STIs is essential. Some girls are reluctant to tell their partner about taking COCs because of concern that his knowledge of her method will negatively affect his condom use. Discussing this issue at the same time as the prescription is provided allows the teen girl to think through her options.
be used. This approach increases the 3-month continuation rate but has not been shown to increase long-term continuation rates or decrease pregnancy rates; however, it may result in greater patient satisfaction (19,20). An alternative is the Sunday Start approach, in which the patient starts the contraceptive method on the Sunday after menses begins. This method was popularized as many oral contraceptive pills are packaged for initiation on Sunday to avoid a withdrawal bleed over the weekend. First-day start is another approach. If the hormonal contraceptive method is started within 5 days of the onset of menses, backup contraception is not necessary (21). However, to keep instructions standardized and provide teenagers the best protection from pregnancy, we advise that backup contraception should be used for 7 days when starting hormonal contraception. For the teen who has been successfully using condoms and now wishes to use hormonal contraception, discussion about the need for continuing protection against STIs is essential. Some girls are reluctant to tell their partner about taking COCs because of concern that his knowledge of her method will negatively affect his condom use. Discussing this issue at the same time as the prescription is provided allows the teen girl to think through her options.
Table 24-2 Percentage of Women Experiencing an Unintended Pregnancy During First Year of Typical Use and First Year of Perfect Use of Contraception and the Percentage Continuing Use at the End of the First Year, United States | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Table 24-3 Four Category Classification for WHO and CDC Guidelines | ||
|---|---|---|
|
Strategies to improve compliance, such as setting a phone alarm or taking a pill after brushing one’s teeth, can be individualized to each patient. The adolescent should be given simple instructions, a handout of basic information, Internet resources, and a contact phone number in case of questions or need for refills. A follow-up appointment should be scheduled within 3 months. Family members or partners who are present and supportive can be helpful in the counseling process and facilitate better contraceptive adherence (22).
Combined Hormonal Contraception
Combined Oral Contraceptives
The COC pill is the most requested method of contraception by teenagers (23). As shown in Table 24-2, the failure rate with perfect use is <1%, although typical failure rates are 8%. COCs are primarily composed of a synthetic estrogen and a synthetic progestin. Table 24-5 lists the oral contraceptives (excluding some generic forms) currently available in the United States, ranked by dose of estrogen and grouped by type of progestin. Of the two synthetic estrogens that have been available in the United States, ethinyl estradiol (EE) is the most commonly used and is available in doses ranging from 20 to 50 μg per pill (a 10 μg EE pill has been recently introduced—see below). Mestranol is found only in a few 50-μg pills. EE is rapidly absorbed, with peak levels in 60 to 120 minutes (24). Mestranol is converted in the liver to ethinyl estradiol; therefore, peak serum levels of ethinyl estradiol are lower and occur later after ingestion of mestranol than after ingestion of the same dose of ethinyl estradiol. Although it is difficult to assess the exact potency of these two compounds in relation to each other, 50 μg of mestranol is generally equivalent to 35 μg of ethinyl estradiol (25). The plasma levels of ethinyl estradiol seen with a COC containing 50 μg mestranol and 1 mg norethindrone were the same as those with a pill containing 35 μg ethinyl estradiol and 1 mg norethindrone (26). However, there is significant variability among individual women and even within the same woman at different sampling times. COCs that contain ≤35 μg of estrogen are termed “low-dose” pills and should be first line when prescribing COCs. Initial trials of COCs containing β-estradiol as the estrogen provided adequate ovulation suppression but not good cycle control; COCs that combine β-estradiol with newer progestins are showing promising results with improved bleeding patterns (27). A recently approved COC (Natazia) includes varying doses of estradiol valerate, a synthetic prodrug of β-estradiol, and dienogest over 26/28 days, but requires 9 days of backup contraception at initiation and has lowered efficacy with a number of drugs.
Most of the progestins used in COCs are 19-nortestosterone derivatives and are related to 19-carbon androgens. These include norethindrone, norethindrone acetate, norethynodrel, ethynodiol diacetate, norgestrel, levonorgestrel, desogestrel, norgestimate, and gestodene. Norgestrel is a racemic mixture of dextro- and levonorgestrel, with the levonorgestrel being the active isomer (thus 0.3 mg of norgestrel can be considered equivalent to 0.15 mg of levonorgestrel). Most progestins have varying qualities of being estrogenic, antiestrogenic, progestational (anabolic), and androgenic. Potencies are extremely controversial because of the varying tests used, including animal models, delay of menses, and ability to induce glycogen vacuoles in human endometrium (26,28). Generally, the estranes—norethindrone, norethindrone acetate, and ethynodiol diacetate—are considered fairly equipotent. Norgestrel, a gonane, is estimated to be 5 to 10 times more potent, and levonorgestrel is 10 to 20 times more potent. Gestodene is considered less androgenic and more progestational than levonorgestrel. Norgestimate is estimated to have three times the potency of norethindrone using a test that measures the percentage of rabbits ovulating (levonorgestrel is seven to eight times more potent than norethindrone on this test). Norgestimate has four to five times less binding affinity for rabbit uterine receptors than levonorgestrel and much less affinity for androgen receptors and sex hormone-binding globulin than levonorgestrel. Desogestrel is more progestational than levonorgestrel but less androgenic; this compound also binds to corticoid receptors (6,24). Dienogest appears to be progestational without having androgenic, estrogenic, or mineralocorticoid effects in animal and in vitro studies. In general, the doses of progestins have been adjusted to be roughly equipotent across all COCs, and none of the progestins are truly androgenic in the quantities contained in COCs and when the estrogen content of the OC is taken into consideration.
Progestins are sometimes classified by “generations” on introduction to the market, although the classification is not particularly helpful. “First generation” COCs contain estranes, such as norethindrone, and “second generation” COCs contain a gonane, such as norgestrel and levonorgestrel. Norgestimate has been grouped as a “third generation” COC along with desogestrel and gestodene, but it is closer to a “second generation” progestin because it is metabolized to levonorgestrel (LNG) and
LNG metabolites. Drospirenone has been called a “fourth generation” progestin. The half-lives of the progestins have been estimated to be about 7 to 8 hours (range, 4 to 11 hours) for norethindrone, 10 to 12 hours for gestodene, 12 hours for dienogest, 16 hours (range, 8 to 30 hours) for levonorgestrel, 45 to 71 hours for norgestimate, and 20 hours (range, 11 to 24 hours) for desogestrel. However, there are variations in elimination half-lives between subjects and within the same subject, as well as evidence of differences when the progestin is given alone, when it is given for multiple doses, and when it is administered with estrogen (26). Some progestins, such as norethindrone, have a loss of drug with the first pass through the liver after absorption, and others, such as levonorgestrel, do not. Long half-lives and other aspects of binding and metabolism of certain progestins likely contribute to the lowered incidence of breakthrough bleeding.
LNG metabolites. Drospirenone has been called a “fourth generation” progestin. The half-lives of the progestins have been estimated to be about 7 to 8 hours (range, 4 to 11 hours) for norethindrone, 10 to 12 hours for gestodene, 12 hours for dienogest, 16 hours (range, 8 to 30 hours) for levonorgestrel, 45 to 71 hours for norgestimate, and 20 hours (range, 11 to 24 hours) for desogestrel. However, there are variations in elimination half-lives between subjects and within the same subject, as well as evidence of differences when the progestin is given alone, when it is given for multiple doses, and when it is administered with estrogen (26). Some progestins, such as norethindrone, have a loss of drug with the first pass through the liver after absorption, and others, such as levonorgestrel, do not. Long half-lives and other aspects of binding and metabolism of certain progestins likely contribute to the lowered incidence of breakthrough bleeding.
Table 24-4 Selected Category 3 and 4 Recommendations for Combined Oral Contraceptives from the 2009 WHO Guidelines and the 2010 CDC Guidelines (including 2011 CDC Updates) for Adolescents and Young Women | ||||||||
|---|---|---|---|---|---|---|---|---|
|
Table 24-5 Common Oral Contraceptives Available in the United States | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
Drospirenone is a synthetic progestin chemically related to 21-carbon 17α-spironolactone with antimineralocorticoid (antialdosterone) and antiandrogenic activity (29,30). It blocks the stimulation of the renin–angiotensin–aldosterone system by estrogen (29). Serum half-life is 31 to 33 hours (31). The 3-mg dose of drospirenone, equivalent to 25 mg spironolactone, may be helpful in the treatment of acne and mild water retention (32). Because of the possibility for potassium retention, drospirenone-containing COCs should be avoided in those with adrenal insufficiency or renal or hepatic disease, and used with monitoring of potassium in those taking potassium-sparing diuretics, potassium supplements, angiotensin-converting enzyme (ACE) inhibitors, nonsteroidal anti-inflammatory drugs (NSAIDs), heparin, or angiotensin II receptor antagonists. Other progestins are under study. The risk of venous thromboembolism varies with the dose of estrogen and the type of progestin and is part of ongoing evaluation by the FDA (see section on side effects).
Mechanism of Action
Combined oral contraceptive pills inhibit ovulation by suppressing gonadotropin secretion. The progestin component is the key player in the prevention of ovulation as it suppresses the midcycle luteinizing hormone (LH) surge necessary for ovulation. Progestin also causes thickened cervical mucus, which impedes sperm transport and prevents conception. It further acts to hinder peristalsis of the fallopian tubes and thus the transport of sperm and ova, providing an additional contraceptive effect. The estrogen component of the COC pill suppresses follicle-stimulating hormone (FSH) secretion and thus follicular development. In addition, it plays a role in increasing intracellular progestin receptors, and thus potentiates the action of the progestin, allowing for lower doses of progestin in the COC. Finally, estrogen is responsible for stabilizing the endometrium in order to minimize irregular bleeding.
Use of Combined Oral Contraceptives
The majority of adolescents are healthy and thus COCs are rarely contraindicated in this population. Most medical conditions in and of themselves do not necessarily preclude use of COCs and the risks of a high-risk pregnancy to the mother must be weighed against the risk of the contraceptive method.
For most adolescents, the COC with a low-dose estrogen (20 to 35 μg EE) and progestin that has the fewest side effects should be prescribed. COCs with 20 μg EE are generally associated with a higher rate of intermenstrual bleeding than 35-μg pills, but 20-μg EE formulations have been associated with lower rates of breast soreness and nausea (34). There have been conflicting studies on whether adolescents on OCs in general or on OCs with lower doses of estrogen acquire less bone mass (35,36,37,38,39,40,41). While a prospective study of adult women ages 18 to 36 years did not find any difference between users and nonusers (36), a cross-sectional study suggested that among women ages 18 to 24 years, initiation within 3 years of menarche and use for >2 years might have an adverse impact (37). Another cross-sectional study (35) found no difference in bone mineral density in
14- to 18-year-old adolescents taking COCs with <30 μg EE versus 30 to 35 μg EE; however, lower bone density was noted in the 19- to 30-year-old women taking the lower dose COC for longer duration. Lack of exposure to the midcycle estradiol concentrations coupled with decreased ovarian and adrenal androgens has been hypothesized to play a role in possibly lessening bone acquisition. At the same time as these cross-sectional studies, a randomized trial of a 20-μg EE versus a 30-μg EE levonorgestrel pill found no difference in bone acquisition for adults ages 20 to 35 years (39). A similar trial of two EE doses with drospirenone also found no difference in bone density at one year (38). Progestins such as norethindrone and levonorgestrel appear to be beneficial (39,40). Overall, a systematic review (41) generally found lower bone mass among adolescent and young adult COC users, higher bone mass among premenopausal users, and, importantly, no difference among former users and never users. Thus, there is currently insufficient data at this time to state definitively if higher-dose 30-μg to 35-μg pills (more suppressive of endogenous estrogen but higher EE dose) are better than 20-μg EE pills, especially in younger patients. Given the mixed evidence, clinicians may be more likely to prescribe COCs with 30- 35-μg EE to young patients but more longitudinal data are needed.
14- to 18-year-old adolescents taking COCs with <30 μg EE versus 30 to 35 μg EE; however, lower bone density was noted in the 19- to 30-year-old women taking the lower dose COC for longer duration. Lack of exposure to the midcycle estradiol concentrations coupled with decreased ovarian and adrenal androgens has been hypothesized to play a role in possibly lessening bone acquisition. At the same time as these cross-sectional studies, a randomized trial of a 20-μg EE versus a 30-μg EE levonorgestrel pill found no difference in bone acquisition for adults ages 20 to 35 years (39). A similar trial of two EE doses with drospirenone also found no difference in bone density at one year (38). Progestins such as norethindrone and levonorgestrel appear to be beneficial (39,40). Overall, a systematic review (41) generally found lower bone mass among adolescent and young adult COC users, higher bone mass among premenopausal users, and, importantly, no difference among former users and never users. Thus, there is currently insufficient data at this time to state definitively if higher-dose 30-μg to 35-μg pills (more suppressive of endogenous estrogen but higher EE dose) are better than 20-μg EE pills, especially in younger patients. Given the mixed evidence, clinicians may be more likely to prescribe COCs with 30- 35-μg EE to young patients but more longitudinal data are needed.
Continuation rates, pregnancy rates, and bleeding patterns are generally similar for monophasic and triphasic pills (42). Pills have fairly similar estrogen/progestin and thus most are beneficial for lessening hirsutism and acne. Traditionally COCs are taken for 21 or 24 days (some formulations have active pills for 26 days or 28 days) of the cycle starting either on the first day of menses or on the first Sunday after the onset of menses. The same-day Quick Start approach can also be used (see p. 448). Women who use this method do not experience an increase in breakthrough bleeding (43). Women are best protected from pregnancy if they use a backup method of contraception, preferably condoms, during the first week of COC use. Withdrawal bleeding occurs on the days on which hormone-free pills are taken. Hot flashes are reported by some patients during the hormone-free week and can be ameliorated by continuous COCs or low-dose transdermal estradiol for the week.
Patients also may desire to avoid the pill-free interval and take the pill continuously. An extended 84/7-day oral contraceptive, consisting of 0.15 mg levonorgestrel and 30 μg EE, demonstrated similar contraceptive effectiveness compared to a cyclic COC. Irregular bleeding was higher with the 84/7 regimen initially, but was comparable by the fourth cycle (44). A 12-month trial of a continuous 20-μg EE/0.1-mg levonorgestrel versus a cyclic pill showed fewer bleeding days when using the continuous regimen (45). Although there are extended regimen COCs on the market, monophasic COCs can be used similarly, taking the active pills continuously and skipping the placebo weeks. The number of continuous months can be preset to 2 to 4 months. COC pills can also be taken continuously until bleeding occurs, after which the COC can be stopped for 3 to 7 days to allow for a withdrawal bleed, and then subsequently restarted. Alternatively, COCs can be taken continuously until bleeding has occurred for 7 days, stopped for 3 days, and then resumed or the COCs can be continued daily as the patient waits for the bleeding to cease spontaneously. Taking pills continuously is associated with less dysmenorrhea, premenstrual syndrome, and headaches but more unscheduled bleeding.
Studies are conflicting on whether body weight plays a role in increasing the risk of OC failure. In one study of women in the Group Health Cooperative of Puget Sound (46), women of the highest body weight quartile (>70.5 kg) had increased risk of OC failure (odds ratio [OR] 1.6, 95% confidence interval [CI]: 1.1 to 2.4) compared to women of the lower weight; higher risk was seen with the very low-dose OCs for the highest quartile (OR 4.5, 95% CI: 1.4 to 4.4). However, other studies have not demonstrated a higher risk for obese patients.
Managing Missed Pills
In order to achieve ovulation suppression and protection from pregnancy, COCs should be taken every day at approximately the same time. However, studies have shown that irregular pill taking is a common occurrence. When compliance was measured using an electronic measuring device, only 33% of women did not miss any pills in their first cycle of use. By cycle 3 one-third of women missed three or more pills (18). Unfortunately, the failure rate of oral contraceptives among adolescents is higher than for adults.
If one pill is missed, the patient should be instructed to take it immediately and to continue to take pills daily, including the regularly scheduled pill for that day. No backup contraception is necessary. If two pills are missed, then the patient should take the missed pills as soon as possible, and continue taking one pill a day as scheduled. No backup contraception is needed. If three or more pills are missed, the patient should take one pill immediately and continue to take one pill per day as scheduled. The patient should use backup contraception or abstain from intercourse for 7 days. Additionally, if she is in her third week of pills and misses three or more pills, the patient should take one pill immediately and then continue taking one pill a day until she has taken all the active pills in the pack. She should then skip the placebo pills and start a new package. If the patient has had intercourse during a period of missed pills or has missed pills within the first week of starting a new package, she should be encouraged to use Emergency Contraception (EC). For teens who are using 20-μg EE pills and miss two pills, the instructions for missing 3 days of pills should be followed (47). Some adolescents may not be able to remember these details and may find written instructions helpful (www.youngwomenshealth.org). In the event that a patient misses a pill and cannot remember what to do, she should be instructed to take her pill and call the clinician’s office for guidance as soon as possible.
Side Effects
Breakthrough bleeding is a common problem during the first few cycles of COC use. The incidence ranges from 10% to 30% in the first month to <10% in the third month of use, and rates are higher as the dose of estrogen decreases (48). Bleeding usually occurs because the endometrium is thin and not maintained for the entire cycle. The first line of management is education and support as the bleeding usually resolves after three cycles. The adolescent should continue to take the COCs as prescribed and should be reassured that breakthrough bleeding does not correlate with decreased efficacy (49) with the exception of patients’ taking medications that may interfere with metabolism, having gastroenteritis, or missing pills. A backup method of contraception is advisable in these circumstances.
If irregular bleeding occurs a few months after the initiation of COCs or is persistent, further evaluation is necessary. Pregnancy and STIs, especially chlamydia infections, should be strongly considered in the differential diagnosis, and questions
should be asked regarding missed pills (“Some patients find it hard to remember all their pills. Have you missed any?”) and consistency of pill taking, medication changes, pelvic pain, new sexual partners, and symptoms of pregnancy. A pregnancy test and screening for chlamydia and gonorrhea are often indicated. Smoking also increases spotting and breakthrough bleeding when on COCs (50).
should be asked regarding missed pills (“Some patients find it hard to remember all their pills. Have you missed any?”) and consistency of pill taking, medication changes, pelvic pain, new sexual partners, and symptoms of pregnancy. A pregnancy test and screening for chlamydia and gonorrhea are often indicated. Smoking also increases spotting and breakthrough bleeding when on COCs (50).
In contrast, some patients may experience amenorrhea months or years after initiation. The thin endometrial lining may lack sufficient tissue to yield withdrawal bleeding. In cases of new-onset amenorrhea, a pregnancy test should be done. Once pregnancy is excluded, the patient should be reassured that the presence of amenorrhea while on COCs is not harmful. If the patient desires a monthly withdrawal bleed, a COC with higher estrogen content (20 μg to 30 or 35 μg EE) or less progestin (e.g., changing from 1.0 mg norethindrone [NET] to one containing 0.4 mg or 0.5 mg NET) can be initiated. After discontinuation of the pill, 1% to 2% of patients have postpill amenorrhea, similar to the incidence of amenorrhea in control populations. Approximately 95% of these patients revert to regular periods within 12 to 18 months. Most patients with postpill amenorrhea have irregular cycles before initiating COC use, and thus the cause of oligomenorrhea should be investigated before a COC is prescribed. For example, a patient with polycystic ovary syndrome and hirsutism may be appropriately treated with a COC, but she will likely experience oligomenorrhea again after the COC is discontinued. Patients who lose weight or engage in endurance sports while taking COCs appear to be more susceptible to postpill amenorrhea just as they would be if they were not taking COCs.
Nausea is generally considered an estrogenic side effect. Patients may experience nausea for the first few days of each cycle; however, the nausea usually abates after 3 months. If patients take the pill after dinner or with a snack at bedtime rather than in the morning, nausea is less likely to be a problem. However, if the nausea is persistent or bothersome, the estrogen dose can be decreased. Patients who have vomiting or gastroenteritis may risk pill failure and should use a backup method for 7 days. If vomiting occurs within 1 hour of taking the pill, taking an extra pill from a different pill pack is recommended.
Adolescents are particularly worried about weight gain with contraceptive methods. Weight gain has not been associated with low-dose COCs (8,51). Despite this evidence, some individual teenagers, especially young teens treated for dysfunctional bleeding, may be prone to gain weight with the use of hormonal methods, perhaps because of increased appetite. Some perceived weight gain might be due to bloating or water retention. Often an adolescent believes she has gained 5 to 10 lb, when in fact the gain is a perception and not an actual change. Thus, if teens call about significant weight gain, it is important to have them come to the office for a weight measurement. If weight gain occurs, a trial of a lower-calorie, lower-salt diet; less screen time; and an increase in exercise should be instituted. If diet and exercise do not control the problem, a lower-dose OC or an alternative form of contraception can be considered.
Subjective symptoms such as depression, nervousness, sleep disturbance, or emotional lability have been associated with COC use in some studies, but not in others (52,53). Some patients report new onset of symptoms of irritability or depression that appear to coincide with the initiation of COC use. Depressive symptoms are common during adolescence, and thus the clinician is often faced with trying to assess whether the COC is part of the problem. It is useful to have a baseline history of psychiatric symptoms prior to COC use. The severity of depressive symptoms and acceptability of alternative contraceptive methods should guide management.
Table 24-6 Questions to Ask to Ascertain a Personal/Family History of Thromboembolism | ||||
|---|---|---|---|---|
|
Serious side effects from COCs are estrogen related and rare, especially among the adolescent population. Reported events include thromboembolic events, myocardial infarction, stroke, hepatic adenoma, hypertension, cervical dysplasia/carcinoma, exacerbation of cholelithiasis, and vascular headache (51,52,53). Many of these adverse events were initially observed in patients taking COC containing >50 μg of estrogen.
Venous thromboembolism (VTE) has received the most press in young women (54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69). Screening questions for a predisposition to VTE are outlined in Table 24-6. The risks of hormonal preparations related to VTE may vary depending on the dose of estrogen, type of progestin, age, family history, low antithrombin III, hypercoagulability states (e.g., polycythemia, protein C and protein S deficiency, factor V Leiden, prothrombin 20210 mutations), travel, immobilization, traumatic injury, surgery, smoking, postpartum, lupus erythematosus, obesity (BMI >35; OR 3.8, 95% CI: 1.8 to 8.0), and chronic diseases such as diabetes. The relative risk for thrombosis in patients taking COCs is three- to fivefold compared to patients not taking COCs (66). While the relative risk of thrombosis when thrombophilia and COCs are combined can be much higher (e.g., 10- to 35-fold for factor V Leiden heterozygotes plus COCs [67]), the absolute risk of thrombosis remains very low. Given the low incidence of VTE in healthy adolescents and young adults, most patients with thrombophilia (who comprise approximately 5% to 7% of whites) do not have thrombotic complications on COCs. Thrombosis risk increases with increasing estrogen dose to 50 μg: 0.8 (0.5 to 1.2) for 20 μg, 1.0 for 30 μg (referent), and 1.9 (1.1 to 3.4) for 50 μg EE (66). Thrombosis is more prevalent during the first 12 months (particularly the first 3 months) of initiating COCs (67); however, it may occur at any time. The odds ratio for COC users compared to nonusers for <1 year was 7.0 (5.1 to 9.6); 1 to 5 years, 3.6 (2.7 to 4.8); >5 years, 3.1 (2.5 to 3.8) (68). Compared to levonorgestrel, progestins such as norethindrone and norgestimate have similar rate ratios of thrombosis (norethindrone: 0.98; range, 0.71 to 1.37; norgestimate: 1.19; range, 0.96 to 1.47). However, higher rate ratios have been observed for desogestrel (1.82; range, 1.49 to 2.22) and drospirenone (1.64; range, 1.27 to 2.10) (69). Desogestrel had previously been shown to have a slightly higher risk, but two previous studies of drospirenone showed no increased risk and the limitations of the recent studies are outlined in the Yaz and Yasmin package inserts. Two studies published in 2011 also found elevated risk for drospirenone leading to an FDA review of the data (58,59).
There are variable estimates of the annual absolute risk for nonfatal VTE. Estimates that have been used are approximately 4 per 100,000 for healthy young women, 10 to 15 per 100,000 for women taking NET- and LNG-containing pills, 20 to 30 per 100,000 for women taking desogestrel- or gestodene-containing pills, and 60 per 100,000 for pregnant women. Age is an important risk factor, with estimates of 1 to 10 per 100,000 for the pediatric age, 10 to 46 per 100,000 for the childbearing age, and 123 per 100,000 for the perimenopausal age. Given the conflicting data, a COC with levonorgestrel, norgestrel, norethindrone, and norgestimate may be reasonable first choices for initiating COC use, although patient choice plays an important role in compliance (55). Adolescents on COCs taking long flights (or long land travel) should maintain hydration and exercise their lower extremities; those at higher risk should use graduated stockings and consider thromboprophylaxis (see Table 24-7). Patients having major surgery that will require bedrest should discontinue COCs 4 weeks preoperatively (balancing the risks and benefits of pregnancy with higher risks of VTE). OCs are contraindicated in patients with systemic lupus erythematosus (SLE) who have antiphospholipid antibodies, cyanotic heart disease, and pulmonary artery hypertension. The presence of varicose veins is not a contraindication to COC use.
At most, 1% to 2% of normotensive individuals develop hypertension (a blood pressure >140/90 mm Hg) within weeks to several months of starting an OC, with the risk of hypertension increasing with age, parity, and obesity (70). The WHO and CDC guidelines for blood pressure need to be adjusted to adolescent standards based on age and height. In girls in whom new-onset hypertension develops while taking OCs, the elevated blood pressure usually returns to normal within 2 to 12 weeks after the pill is discontinued. Preferably, nonhormonal contraception should be used until the blood pressure returns to normal, but a progestin-only method may be used. For example, if mild hypertension develops in a patient taking a 35-μg ethinyl estradiol pill and subsequently resolves, a pill lower in estrogen (20 μg) and progestin or a progestin-only pill may be prescribed, with monitoring of the blood pressure.
Table 24-7 Risk Estimates for Several Common Prothrombotic Risk Factors and Prevention Strategies | ||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||||||||||||||||||||
Headaches occur in 10% to 30% of healthy adolescents on a weekly basis. Thus, it is important for the clinician to obtain a history of headache type, frequency, and severity before prescribing an OC. Headaches may be precipitated by caffeine and alcohol use, fatigue, stress at school, psychosocial issues, sinus infections, medications including hormones, or rarely an intracranial process. Although migraine with aura is considered level 4 for COCs, it should be noted that ascertainment and the definition of migraine in studies are not uniform. In a study of 86 cases of ischemic stroke and 214 controls, Donaghy and colleagues (71) found that the adjusted risk of ischemic stroke was significantly associated with the following: migraine of more than 12 years’ duration, OR 4.61 (95% CI: 1.27 to 16.8); initial migraine with aura, OR 8.37 (95% CI: 2.33 to 30.1); and attacks more frequent than 12 times per year, OR 10.4 (95% CI: 2.18 to 49.4). A systematic review by Curtis and colleagues (72) found that among those women with a history of migraine, COC users were two to four times as likely to have an ischemic stroke as nonusers. When compared with nonusers without migraine, the odds ratio was 6 to almost 14 for COC users with migraine. The absolute risk of stroke in women aged 20 years who have migraine and use COCs has been estimated to be 10 per 100,000, compared to 100 per 100,000 in women aged 40 years (73).
Multiple studies have found no long-term increased risk of breast cancer with COC use, and COCs are considered safe for young women with a family history of breast cancer (74). In an analysis of 54 epidemiologic studies of over 53,000 women with breast cancer and over 100,000 women without breast cancer, the Collaborative Group on Hormonal Factors in Breast Cancer concluded that there was a small relative risk of having breast cancer diagnosed in current users of 1.24 (95% CI: 1.15 to 1.33), which decreased to 1.16 (95% CI: 1.08 to 1.23) 1 to 4 years after stopping OCs, and to 1.01 (95% CI: 0.96 to 1.05) by 10 years after use (75,76). The cancers diagnosed in OC users were less advanced compared to never users. It may be that women receiving OC prescriptions are more likely to have breast cancer diagnosed early. There was no effect on duration of use, age at first use, type of formulation, parity, reproductive or family
history, or a number of other factors. In a large case-control study of women 35 to 64 years old, Marchbanks and colleagues (74) found a relative risk of 1.0 (95% CI: 0.8 to 1.3) for current users and 0.9 (95% CI: 0.8 to 1.0) for previous users. Neither initiation of OCs at a young age nor a family history of breast cancer increased risk. Studies are conflicting on risk for women with BRCA1/2, but the OR is 1.0 or close to 1.0, with one study noting an increase related to duration of OC use before first pregnancy (77,78,79).
history, or a number of other factors. In a large case-control study of women 35 to 64 years old, Marchbanks and colleagues (74) found a relative risk of 1.0 (95% CI: 0.8 to 1.3) for current users and 0.9 (95% CI: 0.8 to 1.0) for previous users. Neither initiation of OCs at a young age nor a family history of breast cancer increased risk. Studies are conflicting on risk for women with BRCA1/2, but the OR is 1.0 or close to 1.0, with one study noting an increase related to duration of OC use before first pregnancy (77,78,79).
Several studies have demonstrated no increase in seizures with current formulations (80,81). In fact, patients with a history of an increase in seizures in the premenstrual and/or menstrual phase of the cycle may benefit from COC use given cyclically or continuously. The major problem with the use of OCs in adolescents with seizure disorders is the potential for intermenstrual bleeding and possibly lowered efficacy of the OC and decreased anticonvulsant levels. Anticonvulsants such as phenobarbital, phenytoin, carbamazepine, primidone, and oxcarbazepine increase the metabolism of synthetic steroids by increasing conjugation in the gut and enzyme induction in the liver (82). In addition, these drugs increase the production of sex hormone-binding globulin to which the progestin is bound. Felbamate and topiramate also appear to have an effect on OC metabolism (83). Unlike Depo-Provera, use of progestin-only pills or subdermal implants does not provide sufficient efficacy for patients on enzyme-inducing anticonvulsants. Sodium valproate has not been associated with OC failure (84). Most of the new antiepileptic agents such as gabapentin, lamotrigine, and levetiracetam and the benzodiazepines do not interfere with COCs. However, COCs lower plasma levels of lamotrigine when used as monotherapy and cessation of hormones during the pill-free week can lead to higher levels of lamotrigine, with reported average increases of 54% and 84% (range, 29% to 129%) (85,86
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
