Herbert B. Peterson
Kathryn M. Curtis
Olav Meirik
Catherine d'Arcangues
Reproductive rights embrace certain human rights that are already recognized in national laws, international human rights documents, and other relevant consensus documents. These rights rest on the recognition of the basic right of all couples and individuals to decide freely and responsibly the number and spacing and timing of their children and to have the information and means to do so, and the right to attain the highest standard of sexual and reproductive health.
—Beijing Platform for Action, 1995
All women have a basic human right to determine whether and when they will become pregnant. Sexually active women and men have a wide variety of contraceptive choices for implementing this right, ranging from temporary to permanent methods. Most women have the potential to become pregnant over at least three decades and most men are fertile for an even longer period. Thus, a series of contraceptive decisions may be necessary over a reproductive lifetime. Since 2000, the range of contraceptive choices in the United States has increased, with the approval of four new contraceptive methods: An estrogen–progestogen injectable, a vaginal ring, a contraceptive patch, and a progestogen-releasing intrauterine device (IUD). This chapter focuses on information to help women and men make the contraceptive choices that best address their needs and circumstances.
CONTRACEPTIVE USE IN THE UNITED STATES AND AROUND THE WORLD
Data from the U.S. National Survey of Family Growth indicate that, in 1995, 65% of women between the ages of 15 and 44 used some method of contraception. Among women who used contraception, permanent methods were the most popular; 28% had undergone tubal sterilization and 11% had partners who had undergone vasectomy. Oral contraceptives were the most popular temporary method, but the proportion of pill users decreased from 31% in 1988 to 27% of contraceptive users in 1995, while the proportion of condom users increased from 15% in 1988 to 20% in 1995. Only 3%, 1%, and 1%, respectively, used the injectable depot medroxyprogesterone acetate, levonorgestrel implants, and IUDs. Despite this level of reported contraceptive use, 49% of pregnancies concluding in 1994 were unintended; 54% of those ended in abortion.
Globally, according to the Demographic and Health Surveys, tubal sterilization is the most common method of contraception (in 1997, 20% of couples in which the woman was of reproductive age and who used a contraceptive method, used tubal sterilization), followed by IUDs (15%), oral contraceptives (8%), and condoms (5%).
CHOOSING A METHOD OF CONTRACEPTION
Although sterilization is the most widely used method of contraception in the United States and in much of the rest of the world, it is appropriate only for women and couples who have made a decision to permanently prevent pregnancy. The initial contraceptive choices for most women and men are temporary. Factors that influence the appropriateness of any contraceptive choice include the relative safety and effectiveness of the method for the individual, the frequency and acceptability of side effects, the willingness and ability to use the method consistently and correctly, cost, and the role of religious or cultural beliefs in method acceptability. Other factors include the frequency of coitus, the length of time pregnancy is to be delayed, the impact on lactation and on the breast-fed infant, and any potential impact of the method on future fecundity. In addition, many women are at risk for sexually transmitted infections, including human immunodeficiency virus (HIV) infection, and the need for dual protection against pregnancy and sexually transmitted infections is an issue of great importance for them. Sexually active women at risk for sexually transmitted infections are best advised to consistently and correctly use condoms for both pregnancy and disease prevention, or to use another method of contraception for pregnancy prevention together with condoms for disease prevention.
The need for such dual protection produces special challenges, as there is some evidence that the use of a highly effective method of contraception other than condoms may affect a person's willingness or ability to use condoms for disease prevention.
CONTRACEPTIVE EFFECTIVENESS
Contraceptive effectiveness is measured as reduction in the probability of conception with use of a contraceptive method over a defined period; it cannot be measured directly, largely because studies cannot determine the proportion of women who would have become pregnant during that time had they not been using contraception. By contrast, contraceptive failure rates can be directly determined and are the most clinically useful measures of effectiveness.
Widely used estimates of contraceptive failure rates are shown in Table 30.1. These rates are provided in two categories: “typical use” and “perfect use.” The former is similar to “use effectiveness” and is characteristic of a typical couple starting to use a method (some use the method properly and others do not); the latter is similar to “method effectiveness,” which is the result of consistent and correct use of the method. For some methods, the typical and perfect use rates are substantially different, which indicates that the user's willingness and ability to use these methods consistently and correctly are more important than they are for those methods that have similar typical and perfect use rates. For example, according to Trussell and colleagues (see Table 30.1), although women who take a combined oral contraceptive pill each day should have a near-zero probability of pregnancy, among typical couples in which the woman initiates use of oral contraceptives, 8% experience an unintended pregnancy during the first year if they do not discontinue pill use for any other reason. By contrast, the typical and perfect use failure rates are virtually the same for IUD users and for users of levonorgestrel implants. Published estimates of contraceptive failure rates apply to groups and may vary substantially among individuals within those groups, particularly among those who use methods with major differences in estimates between typical and perfect use. Most failure rates address the risk of pregnancy within 12 months of starting to use the method. The risk of pregnancy for some methods, particularly those that depend on proper use, is likely to decline over time.
|
|
|
TABLE 30.1. Percentage of women experiencing an unintended pregnancy during the first year of typical use and the first year of perfect use of contraception and the percentage continuing use at the end of the first year in the United States |
ORAL CONTRACEPTION
Combined Oral Contraceptives
Combined oral contraceptives (COCs), containing both estrogen and progestogen, have been available in the United States since 1960, have been used by millions of women worldwide, and have been extensively studied. Formulations for oral contraceptives have changed over time from higher to lower doses of steroid hormones. Today, most COCs prescribed in the United States contain ≤35 µg of ethinyl estradiol. The progestogen component of COCs varies and may include norethindrone, norethindrone acetate, ethynodiol diacetate, norgestrel, levonorgestrel, norethynodrel, and the progestogens of “third generation pills”: desogestrel, norgestimate, and gestodene (gestodene is not available in the United States). Monophasic pills have constant doses of estrogen and progestogen, whereas multiphasic pills vary the doses throughout the cycle.
The primary mechanism of COCs is inhibition of ovulation by suppressing follicle-stimulating hormone and luteinizing hormone. In addition, the progestogen thickens and decreases the amount of cervical mucus, which impedes the ascent of sperm into the upper genital tract. COCs may also act by affecting the endometrium and tubal transport. Although COCs are highly effective in preventing pregnancy when used consistently and correctly, 29% of COC users reported missing one or more pills in the last 3 months in the 1995 National Survey of Family Growth. COCs are substantially less effective when used inconsistently (i.e., as seen with many typical users) (see Table 30.1).
Metabolic Effects
The estrogen and progestogen components of COCs produce metabolic changes, but for most healthy women, the changes associated with the available low-dose COCs have little or no clinical significance.
Estrogens generally produce changes in lipid metabolism that are considered beneficial, including slightly increasing levels of high-density lipoprotein (HDL) and decreasing low-density lipoprotein (LDL). Depending on their level of androgenicity, progestogens can counteract these effects, decreasing HDL and increasing LDL. Therefore, the net effect depends on the doses of both estrogen and progestogen as well as the type of progestogen; however, most changes are within the normal range and not clinically relevant.
Although some studies of older formulations of high-dose COCs reported progestogen-associated elevated glucose and insulin levels and higher rates of relative peripheral insulin resistance, clinical studies of low-dose COCs have not found clinically significant effects on glucose metabolism. A large, cross-sectional study of U.S. women found no elevation in hemoglobin A1C, fasting glucose, insulin, and C-peptide levels among current COC users compared with those who never used COCs. Large, prospective studies have not found increased risks of diabetes mellitus among COC users with either high-dose or low-dose pills.
Cardiovascular Diseases
Epidemiologic research confirms that the overall risk of serious cardiovascular complications attributable to COC use is extremely low for the vast majority of users of modern ethinyl estradiol preparations (≤35 µg). However, individual user characteristics and the type of COC modify the risk. Cardiovascular complications associated with COC use occur while the pill is being used; once pills are discontinued, risk levels return to baseline.
Hypertension
COC use may slightly increase blood pressure among normotensive women, but studies suggest this increase is reversible when COC use is discontinued. Among women with mild hypertension, COC use has also been associated with increased blood pressure. In a cross-sectional study of 94 women with mild hypertension, COC users had significantly increased daytime and nighttime ambulatory systolic blood pressure values (mean 8.3 mm Hg increase for daytime and mean 6.1 mm Hg increase for nighttime). No significant differences in diastolic blood pressure values were found.
Venous Thromboembolism
The incidence rate of venous thromboembolism in healthy nonpregnant Caucasian women is estimated to be 5 per 100,000 woman-years. When such women use COCs, the incidence increases to between 15 and 30 per 100,000 woman-years, or three to six times higher than non-COC users, depending on personal characteristics and the type of pill. By comparison, the risk of thrombosis associated with pregnancy is about 60 per 100,000 pregnant women. The case fatality rate of venous thromboembolism is 1 to 2 per 100 and the risk of death from thromboembolism attributable to COC use is extremely low at 1 to 5 per million per year of use.
The most important risk factor for thromboembolism is family history of the disease, which is often related to genetic thrombophilia. The most common genetic cause of thrombophilia is resistance to activated protein C (factor V Leiden), which occurs in about 5% of Caucasians but is rare in non-Caucasians. The presence of factor V Leiden substantially increases the risk of COC-associated venous thrombosis; screening for the condition before starting pill use has been considered. However, cost–benefit considerations lead most authorities to recommend against screening before starting COCs unless women have a family history of thrombosis.
The third-generation low-dose pills with the progestogens desogestrel or gestodene have been associated with a nearly two-fold increased risk of thrombosis relative to “second- generation” pills, which contain levonorgestrel or norethindrone. Low-dose pills containing cyproterone acetate appear to have risks similar to desogestrel and gestodene pills. Although these elevated risks are noteworthy, they are nonetheless low in absolute terms—an estimated increased risk of 4 deaths per 1 million woman-years of use.
Arterial Diseases
Older age, smoking, diabetes, and hypertension are major risk factors for stroke and myocardial infarction. For women with these risk factors, COC use increases their already elevated baseline risk; thus, women with multiple major risk factors for cardiovascular disease generally should not use COCs.
Stroke
The annual incidence rates of hemorrhagic and ischemic stroke in healthy, nonsmoking women between 20 and 24 years of age are estimated to be 2 and 0.5 per 100,000, respectively. In women between 40 and 44 years of age, these annual rates are 5 and 1 per 100,000, respectively. Nonsmoking women less than 35 years old with normal blood pressure who use COCs have no extra risk of hemorrhagic stroke and only a marginally increased risk of ischemic stroke. Among such women 35 years or older, the risk of hemorrhagic or ischemic stroke is increased 1.5- to 2-fold during use of COCs. Smoking and hypertension increase the COC-associated risk of hemorrhagic and ischemic stroke. Use of COCs by women with classic migraine further elevates their moderately increased risk of ischemic stroke.
Myocardial Infarction
In healthy, nonsmoking women of reproductive age with normal blood pressure, the risk of myocardial infarction is extremely low and the use of COCs causes little, if any, increase in risk. On the other hand, among women who smoke, have hypertension, or have hyperlipidemia, their baseline elevated risk for myocardial infarction is substantially increased by COC use. Myocardial infarction is rare in women under 35 years old, even if they have risk factors; for women 30 to 34 years old who use COCs and smoke, the risk is estimated to be 2 per 100,000 woman-years. By contrast, women 40 to 44 years old who use COCs and smoke have an estimated risk of 25 per 100,000 woman-years. Some studies suggest that women who take pills containing the progestogens desogestrel or gestodene may have an even lower risk of myocardial infarction than that associated with the very low risk of pills containing levonorgestrel or norethindrone, but findings are inconsistent.
Malignant Neoplasia
COCs clearly reduce the risk of some cancers and may increase the risk of some others. Most data are based on COCs with higher doses of estrogens and progestogens than in the currently available pills; studies suggest that lower-dose preparations are likely to have similar effects on cancer risk.
Breast Cancer
A pooled analysis of 54 studies found a small increased risk of breast cancer diagnosis (relative risk = 1.24) while COCs were being used and recently after use. The excess risk was among women with localized disease, and there was a corresponding decrease in metastatic disease. These findings argue that women who use COCs are:
· simply more likely to have existing breast cancers diagnosed because they are more likely to have clinical exams or mammograms
· more likely to have late-stage promotion of tumors that subsequently are more likely to remain localized to the breast.
The observation that the duration of COC use had no effect on breast cancer risk argues for the former explanation. The excess risk of breast cancer disappears 10 years after cessation of pill use. Thus, women who use the pill from age 15 to age 35 years have the same breast cancer risk at age 50 as comparable women who never took COCs. Because the incidence of breast cancer is low at ages when COC use is most common, any effect would affect a relatively small number of women. For example, among women who stop using COCs at 25 years of age, the cumulative risk from ages 25 through 34 years is estimated to be 1 excess cancer diagnosed per 10,000 women. In women who stop COC use at age 40, when incidence rates are higher, an estimated 19 excess cancers will be diagnosed from ages 40 through 49 years. As noted, even these excess cancers may represent only an earlier detection of existing disease; at worst, they represent a small absolute excess occurrence of localized disease.
Cervical Cancer
Several studies in the past suggested that using COCs for 5 years or longer increased the risk of cervical cancer 1.3- to 1.8-fold, but the causality of the association was uncertain. More recent data indicate that COC use may not be associated with cervical cancer in women without persistent genital human papilloma virus (HPV) infection or in women with HPV infection who use COCs for fewer than 5 years. However, in women with persistent HPV infection who have used COCs for 5 or more years, the risk for cancer of the cervix may well increase up to 2-fold. Available data suggest that the increased risk decreases over time after stopping pill use. Questions remain about whether the risk is related to pill use per se, to other characteristics of pill users, such as sexual behavior, or to circumstances associated with pill use, such as cytologic screening. Regardless, where screening services are available, COC users should avail themselves of these services as advised for other women.
Ovarian Cancer
Use of COCs reduces the risk of epithelial ovarian cancer, the most common cancer of the ovary. Risk reduction is related to the duration of use. Using COCs for 5 years or longer confers at least a 50% reduced risk that persists for up to 15 years after cessation of use.
Endometrial Cancer
COC use reduces the risk of cancer of the endometrium. As with ovarian cancer, risk reduction is related to the duration of use; after about 5 years of use, the risk of endometrial cancer is at least 50% lower than in women who never used COCs. Risk reduction persists for 15 to 20 years after stopping pill use.
Liver Cancer
Primary cancer of the liver is rare in populations where hepatitis B or C is not endemic. Some studies in the 1980s pointed to a substantially increased risk of primary liver cancer after long-term use of COCs. Subsequent studies suggest that, among women free of hepatitis B or C, long-term use (>6 years) of mostly less than 50 µg ethinyl estradiol pills may be associated with a modest increased risk of primary liver cancer. In populations where chronic hepatitis is common, pill use has not been associated with primary liver cancer. Liver cancer is rare and usually fatal within 1 year of diagnosis; thus, the fact that there has been no increase in liver cancer deaths in the United States in the four decades that COCs have been in widespread use argues against a substantial risk.
Other Cancers
Some studies have found a lower incidence of cancer of the colon among women who have used COCs, but it is unclear whether the association is causal. Previous claims of increased risk of tumors of the pituitary and the skin in users of COCs have not been confirmed.
Medical Eligibility Criteria for COC Use
Most women can safely use COCs. There are some conditions, however, under which COCs should not be used. These conditions include the following:
· breast-feeding and less than 6 weeks postpartum
· age ≥35 years and smoking ≥15 cigarettes per day
· multiple risk factors for arterial cardiovascular disease (e.g., older age, smoking, diabetes, hypertension)
· elevated blood pressure of ≥160 mm Hg systolic or ≥100 mm Hg diastolic
· current or history of deep vein thrombosis or pulmonary embolism
· major surgery with prolonged immobilization
· current or history of ischemic heart disease
· stroke
· complicated valvular heart disease
· migraine with focal neurologic symptoms (migraine with aura)
· migraine without focal neurologic symptoms and age ≥35 years
· current breast cancer
· diabetes with nephropathy, retinopathy, neuropathy, vascular disease, or diabetes of more than 20 years duration
· severe cirrhosis
· liver tumors.
In addition, there are several other conditions for which women generally should not use combined oral contraceptives (Table 30.2).
|
TABLE 30.2. WHO medical eligibility criteria for contraceptive use |
Noncontraceptive Benefits of COC Use
In addition to protecting against pregnancy, including ectopic pregnancy, and the noted protection against endometrial and ovarian cancers, COC use provides other health benefits including preventing and treating menstrual abnormalities (including bleeding problems and pain); reduced risk of symptomatic pelvic inflammatory disease (PID) relative to use of no contraception (although there is no protection against lower genital tract infection and HIV infection); reduced risk of benign breast cysts; reduced risk of iron deficiency anemia; and treatment of acne. There is also some, albeit inconsistent, evidence that COCs may protect against osteoporosis and uterine fibroids. The protection against ovarian cysts, seen with higher dose COCs, is reduced in low-dose monophasic COCs and may not be present for triphasic COCs.
Return to Fertility After Discontinuing COCs
Women who discontinue COCs have no overall reduction in fertility. There may be a very short delay in time to conception compared with women not using COCs. This delay in time to conception is temporary, and by 3 to 12 months after discontinuation there are no differences in fertility rates.
Side Effects
Nausea and breakthrough bleeding are common side effects of COC use, although they often diminish or disappear after the first 3 months of use. Breakthrough bleeding may be due to the specific pill formulation, but it also may be caused by missed pills. If breakthrough bleeding lasts beyond the first 3 months of use, other gynecologic causes have been ruled out, and the bleeding is a problem for the woman, it may be beneficial to switch to a different COC formulation. Other approaches to managing breakthrough bleeding include adding oral estrogen along with the current COCs (1.25 mg conjugated equine estrogen or its equivalent, daily for 1 week at the time the breakthrough bleeding occurs), doubling up on active pills for 2 or 3 days until the breakthrough bleeding stops, doubling up on active pills through the end of the cycle, or using active pills in a continuous regimen. Nonsteroidal antiinflammatory drugs (NSAIDs) may also help alleviate breakthrough bleeding. Other common side effects associated with COCs are breast tenderness and headaches. Weight changes are often reported but most recent studies suggest that little, if any, weight gain can be directly attributed to COC use.
Progestogen-only Oral Contraceptive Pills
Progestogen-only pills are an appropriate alternative for women who desire an oral contraceptive but who are not candidates for COCs. They are commonly used by women who are breast-feeding after 6 weeks postpartum. Progestogen-only oral contraceptives act on the cervical mucus and may act on the endometrium to prevent pregnancy, but do not consistently inhibit ovulation. Progestogen-only pills are estimated to be as effective as COCs when taken consistently (see Table 30.1), but they are less forgiving of inconsistent use. The most common side effects associated with progestogen-only pill use are changes in vaginal bleeding patterns.
PROGESTOGEN-ONLY AND ESTROGEN–PROGESTOGEN INJECTABLES
Progestogen-only Injectables
Progestogen-only injectables first became available to women in the United States with approval of the U.S.
Food and Drug Administration (FDA) in 1992. The injectable, containing 150 mg of depot medroxyprogesterone acetate, inhibits ovulation, produces an atrophic endometrium, and alters cervical mucus to decrease sperm penetration. Intramuscular injections are administered every 3 months (within a window of 2 weeks on either side). Most women who want to use a progestogen-only injectable can safely do so. Women with current breast cancer should not use this method. Several other conditions for which women generally should not use progestogen-only injectables are listed in Table 30.2.
The most common side effects experienced by progestogen-only injectable users are changes in vaginal bleeding patterns. During the first few months of use, many women experience irregular and prolonged bleeding. However, as use continues, bleeding becomes less frequent and up to half the women become amenorrheic by 1 year of use. Some women also report weight gain during the first several years of use; studies suggest a mean weight gain of 3 to 5 lbs per year.
Because of the highly effective suppression of ovulation leading to amenorrhea in many women, there is also a significant delay in return to fertility after discontinuation of progestogen-only injectables. The median time to pregnancy is 6 to 9 months after the last injection. Progestogen-only injectables are the only temporary contraceptives with a substantial delay in return to fertility.
Concern has been raised that progestogen-only injectable use may adversely affect bone mineral density. This question may be particularly important for adolescents, who have not yet reached peak bone mass. A meta-analysis of 12 studies concluded that women who were currently using depot medroxyprogesterone acetate had a lower average bone mineral density than non-users, with some suggestion that women with longer duration of use had greater reductions in bone mineral density. No differences in bone mineral density were found between past users and those who had never been users. One study has found that premenopausal women who discontinue progestogen-only injectables can regain much of their lost bone mass. However, whether adolescents who are still developing peak bone mass can regain bone density to normal levels has not been studied. Likewise, whether perimenopausal use will precipitate the decline in bone mass associated with menopause is unclear.
Estrogen–Progestogen Injectables
In 2001, the FDA approved the first combined estrogen-progestogen injectable for use in the United States. This injectable is administered intramuscularly on a monthly basis and contains 25 mg depot medroxyprogesterone acetate and 5 mg estradiol cypionate. Combined injectables appear to be similar to combined oral contraceptives in their mechanism of action, eligibility criteria, side effects, and safety profile. Vaginal bleeding patterns are more stable with combined injectables than with progestogen-only injectables, and most women experience a monthly withdrawal bleed. Return to fertility is not as delayed as it is with progestogen-only injectables.
LEVONORGESTREL IMPLANTS
Currently, the levonorgestrel implants in use in the United States consist of six capsules containing 36 mg of levonorgestrel for release over 5 years, with potential for effectiveness up to 7 years. A similar system, composed of two rods with 75 mg of levonorgestrel for release over 5 years, has been approved by the FDA but is not yet marketed in the United States. Both systems are inserted subcutaneously in the woman's upper inner arm, under local anesthesia. Although complications with insertions and particularly with removals have been reported, complication rates are related to the training and experience of the provider; studies show significantly fewer complications and faster removal times for the two-rod system than for the six capsules. A woman should have ready access to implant removal on request.
Levonorgestrel implants prevent pregnancy by disrupting normal ovulation in most women, including changes that range from insufficient luteal function to inhibition of ovulation. In addition, the cervical mucus becomes viscous and scanty, which inhibits sperm penetration.
Both systems are highly effective long-term methods, with little action needed by the contraceptive user. However, among the small number of women who become pregnant during use, the proportion of pregnancies that are ectopic is higher than among women who use no contraception.
Most women can safely use levonorgestrel implants. However, women with current breast cancer should not use them. Other conditions for which women generally should not use levonorgestrel implants are listed in Table 30.2.
An international cohort study of over 16,000 women followed for up to 5 years confirmed the safety of levonorgestrel implant use. This study found no significant excess risk of adverse effects for levonorgestrel users compared with controls (IUD users and women with tubal sterilization), other than small excess risks of gallbladder disease and hypertension.
Most levonorgestrel implant users experience changes in vaginal bleeding patterns, including prolonged or irregular bleeding. Although these changes are most pronounced in the first months of use and tend to diminish over time, they persist for many women. As bleeding changes are among the most common reasons for discontinuation of implants, women should be counseled about these bleeding changes before initiating implant use. Other reported side effects include weight gain, headaches, acne, and mood changes. Some women also experience persistent ovarian follicles because of the incomplete suppression of ovulation. These follicles generally disappear within 1 to 2 months and, thus, most do not require surgery.
CONTRACEPTIVE PATCH
In 2001, the FDA approved the first transdermal contraceptive patch. The patch is designed to release 20 µg ethinyl estradiol and 150 µg norelgestromin, the active metabolite of norgestimate, daily for 1 week. The patch acts similarly to oral contraceptive pills in that the user applies the patch weekly for 3 weeks, followed by a patch-free week to allow for menstrual bleeding. The patch is also similar to oral contraceptive pills in effectiveness, side effects, and incidence of breakthrough bleeding. It may be less effective among women who weigh 90 kg or more. In a study that compared the patch with an oral contraceptive, patch users experienced higher rates of breakthrough bleeding or spotting and reported more breast discomfort in the first 2 months of use, but not thereafter. Dysmenorrhea and headaches were infrequent reasons for discontinuation but were more frequent among patch users than among COC users. In clinical trials, patch users have demonstrated high compliance with method use, significantly better than COC users in one study. Over time, the patch may prove to be more effective than oral contraceptives with typical use because of increased compliance.
VAGINAL RING
Most steroid hormones are absorbed efficiently through the vaginal epithelium and can be released from vaginal rings made out of polymers such as Silastic or ethylvinyl acetate. As a delivery system, the vaginal ring is the only long-acting method that is under the user's immediate control. It can be easily inserted, checked, removed, and replaced by the user. Other advantages include the following:
· its use is not related to coitus
· it provides a constant rate of drug release, resulting in a steady plasma level of the minimum dose required for contraception
· metabolic side effects are reduced by avoiding the first-pass effect through the liver
· in the case of accidental pregnancy or if protection is no longer required, plasma levels fall rapidly to zero.
Fertility returns promptly after the ring is removed. In 2001, the FDA approved the first contraceptive ring for use in the United States. The ring, which releases 15 µg ethinyl estradiol and 120 µg etonogestrel daily, is left in place for 3 weeks and then is removed for 1 week to allow for withdrawal bleeding. Like the pill, it acts mainly by inhibiting ovulation and it is highly effective when used correctly. Experience to date suggests that it is well accepted by users and their partners and that it does not cause untoward local effects.
IUDS
Although IUDs are the most commonly used reversible method of contraception worldwide, their use in the United States has plummeted from 7.1% of married women in 1981 to just 0.5% of women between the ages of 15 to 44 years in 1995. Much of the decrease in use of this long-term, highly effective method of contraception can be attributed to concerns about the risk of pelvic infections stemming from studies of IUDs in the 1970s. In particular, the Dalkon Shield IUD was associated with an increased risk of infection and was removed from the market in 1975. Today, pelvic infection among users of modern IUDs is primarily associated with exposure to sexually transmitted infections; the risk among IUD users at no risk for sexually transmitted infections is extremely low.
Currently, two IUDs are available for use in the United States. The most commonly used IUD is the copper T 380A, which is made of polyethylene with fine copper wire wound around the stem and copper sleeves on the two arms of the “T”. It remains effective for at least 10 years. The copper IUD is the most cost-effective method of contraception in the United States over 5 years of use. The FDA approved a progestogen-releasing IUD in 2000; this IUD releases 20 µg levonorgestrel per day for 5 years. The progestogen plays a role in the mechanism of action and decreases the amount and duration of menstrual bleeding and dysmenorrhea. Both IUDs are highly effective, with typical use failure rates ranging between 0.1% and 0.8% in the first year of use (see Table 30.1). Over the long term, IUD failure rates are similar to those of tubal sterilization, which makes it an ideal reversible alternative to sterilization for many women. Both the copper and levonorgestrel IUDs also protect against ectopic pregnancy. However, among women who become pregnant while using an IUD, the proportion of pregnancies that are ectopic is higher than it is among women who use no contraception.
The primary mechanism of action of IUDs appears to be prevention of fertilization as demonstrated by studies finding significantly decreased numbers of fertilized ova in the fallopian tubes of copper IUD users compared with women who use no contraception. IUDs also act by stimulating an inflammatory response in the uterine cavity, which decreases sperm transport, impedes the ability of sperm to fertilize the ovum, and may be spermicidal. The copper in copper-bearing IUDs enhances this response. In the progestogen-releasing IUD, the hormone also acts on the cervical mucus and ovarian function to prevent fertilization. IUDs may also act to prevent implantation if fertilization occurs.
The primary health concern with the use of IUDs has been the risk of upper genital tract infection. Early studies overestimated the association between IUD use and PID due to inappropriate control groups and lack of control for confounding factors. However, studies of modern IUDs among women at low risk for sexually transmitted infections have shown that the risk of infection is primarily associated with insertion of the IUD, is low (1 per 1,000 woman-years), and is largely limited to the first 20 days after insertion (Fig. 30.1). A randomized clinical trial of 1,833 U.S. women having IUDs inserted compared antibiotic prophylaxis to placebo and found very low rates of infection (1 participant in each group) during the 90 days after insertion. A large international cohort study reported an incidence rate of acute PID of 0.6 per 1,000 woman-years of copper IUD use. Some studies of levonorgestrel IUDs have shown decreased risk of PID compared with copper IUDs, possibly due to thickening of the cervical mucus, which could act as a barrier to bacteria; this finding needs to be confirmed in future studies. In addition, recent studies reduce concerns about any association between IUD use and tubal infertility.
|
|
|
FIG. 30.1. Incidence of pelvic inflammatory disease (PID) by duration of intrauterine device use. The 95% confidence intervals are shown. (From Farley TMM, Rosenberg MJ, Rowe PJ, et al. Intrauterine devices and pelvic inflammatory disease: an international perspective. Lancet 1992;339:785–786.) |
The IUD is an appropriate method of contraception for many women. Advantages of IUD use include it being a long-term, highly effective, reversible method, with little action required of the user. Exposure to sexually transmitted infections is one of the most important factors to assess when a woman is considering IUD use, and women with current sexually transmitted infections, PID, or increased risk of sexually transmitted infections are generally not appropriate candidates for IUDs. Young age and nulliparity are not by themselves factors that restrict IUD use, although studies have shown a higher risk of expulsion among young and nulliparous women. Some studies have shown that IUD use in nulliparous women is not associated with tubal infertility, and there is no delay in return to fertility after IUD removal. Therefore, young women and nulliparous women may be appropriate candidates for IUD use, if they are at low risk of sexually transmitted infections. Conditions for which women should not use an IUD include the following:
· pregnancy
· puerperal sepsis
· immediate postseptic abortion
· anatomic abnormalities that distort the uterine cavity
· unexplained bleeding, suspicious for a serious condition
· malignant gestational trophoblastic disease
· cervical cancer
· breast cancer (for progestogen-releasing IUDs only)
· endometrial cancer
· uterine fibroids with distortion of the uterine cavity
· PID—current or within the last 3 months
· sexually transmitted infections—current or within the last 3 months
· known pelvic tuberculosis.
Other conditions for which women generally should not use IUDs are listed in Table 30.2.
IUDs can be inserted at any time during the menstrual cycle if the woman is not pregnant. For postpartum women, immediate IUD insertion (within 10 minutes of expelling the placenta) may reduce the risk of expulsion. The IUD can also be inserted after 4 weeks postpartum without an increased risk of perforation or expulsion. However, for breast-feeding women, the progestogen-releasing IUDs generally should not be inserted until 6 weeks postpartum because of theoretical concerns about exposing the newborn to steroid hormones. An IUD can also be inserted immediately after an uncomplicated first trimester spontaneous or induced abortion, but it is recommended that insertion be delayed until involution is complete after second trimester abortions.
A pelvic exam is necessary before IUD insertion to rule out pregnancy and pelvic infection and to identify the position and mobility of the uterus. Accurate determination of uterine position is necessary to prevent perforation; the incidence of perforations with IUD insertion is about 1 per 1,000 insertions. Because the main risk of infection occurs with insertion, careful aseptic technique is required. Randomized clinical trials have not shown differences in infection rates or early discontinuation of IUD use when comparing antibiotic prophylaxis with placebo; however, there was a significant decrease in unscheduled return visits among the group who received antibiotics in one trial. Use of NSAIDs before insertion may reduce discomfort; local anesthesia is another option that is infrequently needed. Once the IUD is inserted, the woman should check for the IUD string so that she will later be able to confirm its presence and length.
IUD users have high rates of continuation—approximately 80% at 1 year of use in the United States. Increased menstrual bleeding and pain are the most common side effects reported with copper IUD use and are a primary reason for discontinuing the method. Because levonorgestrel-releasing IUDs generally reduce the amount of cramping and bleeding, women with heavy menstrual bleeding or dysmenorrhea may benefit from these IUDs. Treatment with NSAIDs has been shown to effectively reduce pain and blood loss. In contrast, amenorrhea is a major cause of discontinuation of the levonorgestrel-releasing IUD; it occurs in about 5% of users in the first year of use.
IUD expulsions occur in about 5% of users. Risk factors for expulsion include young age, nulliparity, and heavy bleeding. IUD users should check for the presence and length of the IUD strings frequently, at least after every menstrual period, to confirm that the IUD is still in place. Although pregnancy in an IUD user is an unusual event, when it does occur and the IUD strings are visible, the IUD should be removed to decrease the risk of spontaneous abortion, preterm delivery, and sepsis.
MECHANICAL BARRIER METHODS
The male latex condom is the only method of contraception proven to be highly effective in preventing both pregnancy and HIV infection when used consistently and correctly. Male condom use also reduces the risk of gonorrhea and chlamydial infection. Inconsistent or incorrect use, however, undermines this great potential benefit. Polyurethane male and female condoms have been less fully evaluated for contraception and disease prevention but consistent and correct use of these methods should also provide substantial dual protection. The potential for condoms to be highly effective if used consistently and correctly is in stark contrast to the long-standing notion that condoms are not highly effective. In fact, as shown in Table 30.1, perfect use of male condoms is more effective than typical use of oral contraceptives, which highlights the role of male condom users. It is now clear that true condom failure is infrequent but that failure to consistently and correctly use condoms is not.
Diaphragms and cervical caps may also provide some protection against some sexually transmitted infections, particularly gonorrhea and chlamydial infections, but there is substantially less evidence for this protection than for male condoms. Pending further evidence, protection against sexually transmitted infections should be presumed to be less complete than for male condoms and protection against HIV infection should not be assumed at all.
Male Condom
The Centers for Disease Control and Prevention provide the following instructions for correct use of male condoms:
1. Use a new condom with each act of sexual intercourse.
2. Handle the condom carefully to avoid damaging it with fingernails, teeth, or other sharp objects.
3. Put the condom on after the penis is erect and before any genital contact with the partner.
4. Ensure that no air is trapped in the tip of the condom.
5. Ensure adequate lubrication during intercourse, possibly requiring the use of exogenous lubricants.
6. Use only water-based lubricants with latex condoms (oil-based lubricants can weaken latex).
7. Hold the condom firmly against the base of the penis during withdrawal, and, to prevent slippage, withdraw while the penis is still erect.
Condom breakage rates in the United States are low, about 2 broken condoms per 100 used. Proper use reduces the risk of condom breakage. Condom slippage may be more likely to contribute to pregnancies than condom breakage, which highlights the importance of following instruction 7 given above. Most condom failures are the result of inconsistent or incorrect use rather than failures of the product per se.
Female Condom
In 1993, the FDA approved the female condom for marketing and distribution in the United States. The device is a prelubricated, loose-fitting polyurethane sheath that is 17 cm in length and has a flexible ring at each end. One ring is loose inside the condom and is used to insert the condom into the vagina; the other ring lines the opening of the condom, remains outside the vagina, and covers a portion of the external genitalia during intercourse. The woman should insert the condom by holding the sheath at the closed end and grasping the inner ring; the inner ring should then be inserted toward the apex of the vagina by the index finger, making sure that the sheath is not twisted and that the outer ring remains outside the vagina. It is important that the penis is guided into the opening of the sheath to reduce the likelihood that penetration around the sheath occurs during intercourse. Thus, although this condom is female inserted, its proper use, like that of the male condom, requires the cooperation of both partners.
Diaphragm
The diaphragm is a latex, dome-shaped cup with a flexible spring rim. Diaphragms range in size from 50 to 95 mm and require vaginal examination for proper fitting. An appropriately placed diaphragm should completely cover the cervix; the posterior rim should lie in the posterior fornix and the anterior rim should fit posteriorly to the symphysis pubis, with about 1 cm between the rim and the symphysis pubis. In general, the largest diaphragm that is comfortable for the woman should be prescribed. If the diaphragm is too small, it may be displaced during coitus; if it is too large, it may cause discomfort or trauma. Diaphragms are used with spermicidal creams or jellies applied to the inside of the dome. After insertion, the diaphragm substantially reduces the risk of pregnancy for about 6 hours. If coitus occurs after that time, additional spermicidal cream or jelly should be placed intravaginally without removing the diaphragm. Likewise, additional spermicide should be placed intravaginally if coitus is repeated within 6 hours. The diaphragm should remain in place for at least 6 hours after coitus but for no longer than 24 hours to reduce the risk of toxic shock syndrome.
Cervical Cap
The cervical cap is a latex cup with a firm rim that covers the cervix and fits snugly around its base. Four sizes of the cap are available (with inner diameters of 22, 25, 28, and 31 mm). The properly fitted cap has a rim with an inner diameter that is almost the same as the diameter of the base of the cervix, so the cap remains in close contact with the cervix. The cap should be long enough to cover the entire cervix without resting on the cervical os. If the cap is too large, it is more likely to be displaced during coitus; if it is too small, it may cause trauma. The device is used with spermicide placed in the cap before insertion and is effective for 48 hours after insertion. It should be removed no longer than 48 hours after insertion to reduce the risk of toxic shock syndrome.
SPERMICIDES
Spermicides, formulated as vaginal foams, gels, creams, film, suppositories, and tablets, usually contain the surfactant nonoxynol 9 in the United States; the surfactants octoxynol and benzalkonium chloride are widely available elsewhere. Spermicides are not highly effective for contraception when used alone, even when used consistently and correctly. When used alone, or with a mechanical barrier (condoms, diaphragms, or cervical caps), spermicides should be inserted just before coitus.
The hope that vaginal use of nonoxynol 9 would reduce the risk of sexually transmitted infections, including HIV infection, has met with disappointment. Three randomized, controlled trials have failed to demonstrate any benefit in preventing HIV infection and the most recent of these suggests that frequent use of spermicides containing nonoxynol 9 may actually enhance the risk of acquiring infection; the U.S. Centers for Disease Control and Prevention and the World Health Organization now recommend that spermicides containing nonoxynol 9 not be used for prevention of sexually transmitted infections, including HIV infection. Evidence suggests that vaginal spermicides containing nonoxynol 9 are not effective in preventing cervical gonorrhea or chlamydial infection.
FERTILITY AWARENESS-BASED METHODS
Fertility awareness-based methods use signs and symptoms to estimate a woman's fertile period, during which time the couple should refrain from unprotected intercourse. Calendar methods estimate the fertile days based on the length of a woman's usual menstrual cycle; the other three fertility awareness-based methods are based on monitoring physiologic changes throughout the menstrual cycle. About 25% of typical users of fertility awareness-based methods become pregnant in the first year of use. Failure rates are based on couples who abstain from sex during the fertile period and may be different for couples who choose to use barrier methods during that time. The relatively high failure rates are partially due to inconsistent use of the method, but they are also due to the lack of reliability in estimating the fertile window.
One study that measured timing of ovulation with daily urine samples found that in only 30% of women did the actual fertile window fall within the days of the fertile window as estimated from calendar calculation. Training in how to use these methods, as well as monitoring the menstrual cycles for several months, is necessary for successful use.
Calendar Methods
Calendar-based methods require abstaining from unprotected intercourse during the woman's fertile period, which is estimated from the typical length of her menstrual cycle and from assumptions about the timing of ovulation, the length of time the ovum is capable of being fertilized, and the length of time sperm can survive in the female genital tract. One method estimates the fertile period to be from the length of the shortest cycle minus 18 to the length of the longest cycle minus 11. For example, if the woman's shortest cycle was 25 days and her longest cycle was 29 days, her fertile period (i.e., when abstinence is required) would be from day 7 through day 18. Another version, the standard days method, simply states that for women whose cycle lengths are between 26 and 32 days, the fertile period is from days 8 through 19 of the cycle.
Cervical Mucus Method
The cervical mucus method entails checking the quality and quantity of cervical mucus each day; the fertile period is indicated by clear, wet, and slippery mucus. Abstinence from unprotected intercourse is necessary during menses, every other day during the preovulatory period (as intercourse interferes with interpreting the cervical mucus signs), and from the time fertile mucus appears through 3 days after the last day of fertile mucus.
Basal Body Temperature Method
The basal body temperature method is based on temperature changes throughout the menstrual cycle. A rise of 0.4 to 0.8 degrees above the mean temperature of the preovulatory phase for 3 days indicates ovulation has occurred. Therefore, abstinence is required from the time of menses until 3 days after the rise in temperature.
Symptothermal Method
The symptothermal method is the use of at least two of the previously described methods and may also rely on other changes during the menstrual cycle, including midcycle pain and bleeding, as well as position and texture of the cervix.
EMERGENCY CONTRACEPTION
Emergency Contraceptive Pills (ECP)
Emergency contraception was once called “the best kept secret of family planning.” In the early 1970s, Yuzpe and colleagues tested combinations of ethinyl estradiol and norgestrel for postcoital contraception, based on a single dose of 50 µg ethinyl estradiol and 500 µg norgestrel. Subsequent clinical trials led them to conclude that the most successful regimen consisted of two doses of 100 µg ethinyl estradiol and 500 µg levonorgestrel, with the first dose taken within 72 hours of unprotected intercourse, and the second dose taken 12 hours later.
When started within 72 hours of unprotected intercourse, combined ECPs prevent about 75% of pregnancies. About half of users report nausea and about 20% experience vomiting. A randomized clinical trial showed that pretreatment with 50 mg of meclizine significantly reduces the frequency and severity of nausea and vomiting.
Various doses of levonorgestrel have also been tested since the 1970s for emergency contraception. A large multinational, randomized, double-blind clinical trial was conducted comparing the Yuzpe regimen and levonorgestrel treatment (one 750 µg dose followed by a second 750 µg dose 12 hours later) when started up to 72 hours after unprotected intercourse. Results showed that levonorgestrel was not only better tolerated but was also more effective than the Yuzpe regimen, preventing 89% of expected pregnancies among women who correctly used the treatment. It also showed that the earlier either treatment was taken after the act of unprotected intercourse, the more effective it was. Since then, it has been demonstrated that both 750 µg pills can be taken together in a single dose (1.5 mg levonorgestrel) with the same effectiveness and side effects as the two dose regimen.
A primary mechanism of action of ECPs is to inhibit, delay or otherwise interfere with normal ovulation, thereby preventing fertilization. ECPs may also alter the endometrium and impair implantation, although one study found that use of levonorgestrel ECPs did not impair endometrial morphology. ECP use does not disrupt an established pregnancy.
ECPs are safe for use by almost all women. Concerns about ECP distribution have included those regarding advance provision and multiple administrations. Studies have shown that women who were given an advance supply of ECPs had the same rates of unprotected intercourse as controls, followed the treatment regimen correctly, were more likely to use one dose of ECPs, but were no more likely than the control group to use ECPs repeatedly.
Emergency Use of Copper-Bearing IUDs
Copper-bearing IUDs are extremely effective when used as emergency contraception (99%). The IUD must be inserted within 5 days of unprotected intercourse. This method is most appropriate for women who want long-term, highly effective contraception, and have no contraindications to IUD use.
STERILIZATION
Vasectomy and tubal sterilization are intended to be permanent and are appropriate only for women and men who have made a fully informed and well-considered decision to permanently prevent pregnancy. Although some sterilizations are potentially reversible, depending on factors including the sterilization procedure, the length of normal fallopian tube remaining, the age of the woman (tubal sterilization), and the length of time between the sterilization and reversal procedures (vasectomy), many sterilizations are not reversible. Further, the costs of sterilization reversal or in vitro fertilization are prohibitive for many who regret having had a sterilization procedure.
Although vasectomy and tubal sterilization have been shown to be safe and highly effective, vasectomy is somewhat safer, partly because most vasectomies are done with local anesthesia and tubal sterilizations are intraabdominal procedures. Vasectomy is believed to be somewhat more effective than tubal sterilization and its effectiveness is easily assessed by semen analysis. The long-term effectiveness of vasectomy, however, has been less completely studied than that of tubal sterilization; some trials suggest that the effectiveness of vasectomy, like tubal sterilization, may vary according to the method of occlusion.
Counseling of those considering sterilization is critically important. Such counseling should include not only the anticipated risks and benefits of the surgery but also the possibility of later regretting having had the procedure. The strongest predictors of later regretting tubal sterilization are young age at the time of sterilization and substantial conflict between the woman and man at the time of sterilization. Women considering tubal sterilization should understand that pregnancy is possible even remote from sterilization, and they should understand that a high proportion of pregnancies after tubal sterilization are ectopic gestations.
Vasectomy
Vasectomy, or transection and occlusion of the vas deferens, is usually performed in an outpatient setting with local anesthesia and without premedication. After incising or puncturing the skin of the scrotum, the vas is identified, dissected free of its fascial sheath, and divided. The cut ends of the vas are then occluded by ligation, coagulation of the mucosa, or, rarely, clip application. Fascial interposition is frequently performed after ligation or coagulation by pulling the sheath over one of the vas ends and suturing it—to reduce the likelihood of subsequent spontaneous anastomosis. The most frequent complications of vasectomy are hematoma formation and infection, each of which occurs in about 2% of procedures. Death and serious morbidity are rare. No long-term serious adverse health effects have been documented, and the best available evidence suggests no effect on risk of cardiovascular disease, prostate cancer, or testicular cancer.
Although there is a substantial decrease in the concentration of sperm in the semen by 3 days after vasectomy, complete absence of sperm may take 16 weeks or more, depending in part on the number of ejaculations. Semen analysis should be performed, where feasible, after 12 weeks or 20 ejaculations and temporary contraception should be used in the interim. The time to azoospermia after vasectomy is under study and may vary according to surgical technique.
Tubal Sterilization
The proportion of sterilizations done on an interval basis (not pregnancy-associated) varies from country to country. Approximately one half of tubal sterilizations in the United States are performed after vaginal delivery or at cesarean section and the other half are performed at a time unrelated to pregnancy. The timing of the procedure with respect to pregnancy can affect the surgical approach, the method of tubal occlusion, and the choice of anesthetic. For example, interval procedures are usually performed in the United States by laparoscopy using coagulation (the bipolar technique is illustrated in Fig. 30.2), silicone rubber bands (Fig. 30.3), or clips (the spring clip technique is illustrated in Fig. 30.4) under general anesthesia; laparoscopic sterilization can be even more safely performed with local anesthesia. By contrast, procedures at the time of cesarean section require no additional anesthesia and usually involve partial salpingectomies (the modified Pomeroy method is illustrated in Fig. 30.5); procedures postvaginal delivery are performed by minilaparotomy with subumbilical incisions and likewise usually involve partial salpingectomies.
|
|
|
FIG. 30.2. Bipolar method. A 3-cm minimum zone of isthmic tube is desiccated with bipolar forceps. The paddles of the forceps extend across the tube onto the mesosalpinx. (From Peterson HB, Warshaw J, Pollack A. Tubal sterilization. In: Rock JA, Thompson JD, eds. TeLinde's operative gynecology, eighth ed. Philadelphia: JB Lippincott Co, 1997:529–547.) |
|
|
|
FIG. 30.3. Silicone rubber band method. The isthmic portion of the tube is retracted into the applicator barrel using grasping tongs, which should completely surround the tube. To avoid excessive traction on the tube and its mesentery, the applicator barrel is advanced toward the tube during this retraction process. (From Peterson HB, Warshaw J, Pollack A. Tubal sterilization. In: Rock JA, Thompson JD, eds. TeLinde's operative gynecology, eighth ed. Philadelphia: JB Lippincott Co, 1997:529–547.) |
|
|
|
FIG. 30.4. Spring clip method. The clip is applied to the mid-isthmus at a 90-degree angle to the long axis of the tube. The hinge of the clip should be pressed against the tube, and the tips of the clip should extend onto the mesosalpinx. (From Peterson HB, Warshaw J, Pollack A. Tubal sterilization. In: Rock JA, Thompson JD, eds. TeLinde's operative gynecology, eighth ed. Philadelphia: JB Lippincott Co, 1997:529–547.) |
|
|
|
FIG. 30.5. Pomeroy method. A: A loop of the isthmic portion of the tube is elevated and ligated at its base with one or two ties of no. 1 plain catgut suture. If performed through a minilaparotomy incision, these ties should be held long to prevent premature retraction of the tubal stumps into the abdomen when the loop of tube is transected. B: A fenestration is bluntly created through the mesentery within the tubal loop, and each limb of the tube on either side of this fenestration is individually cut. The cut ends of the tube are inspected for hemostasis and allowed to retract into the abdomen. (From Peterson HB, Warshaw J, Pollack A. Tubal sterilization. In: Rock JA, Thompson JD, eds. TeLinde's operative gynecology, eighth ed. Philadelphia: JB Lippincott Co, 1997:529–547.) |
The likelihood of pregnancy after tubal sterilization varies according to age at the time of the procedure and sterilization technique (Table 30.3). The chance that pregnancies are ectopic varies by method. Among women in the U.S. Collaborative Review of Sterilization, the highest proportion of pregnancies that were ectopic occurred after bipolar coagulation (65%), followed by interval partial salpingectomy (43%), silicone rubber band application (29%), postpartum partial salpingectomy (20%), unipolar coagulation (17%), and spring clip application (15%). Further, the proportion of pregnancies that were ectopic for all methods combined increased over time, being three times as high (61%) in 4 to 10 years after sterilization as in the first 3 years (20%).
|
|
|
TABLE 30.3. Life-table cumulative probability of pregnancy among women undergoing tubal sterilization in the U.S. collaborative review of sterilization by agea |
Death from tubal sterilization is rare, 1 to 2 deaths per 100,000 procedures, with most deaths attributable to complications of general anesthesia. Major complications, principally unintended laparotomies, occur in 1% to 2% of procedures.
The potential for late sequelae of tubal sterilization has been assessed. There is strong evidence against a purported post-tubal ligation syndrome of menstrual abnormalities. Sterilized women in the U.S. Collaborative Review of Sterilization were no more likely than women whose partners underwent vasectomy to have menstrual abnormalities within 5 years after sterilization. In the same study, sterilized women were more likely to undergo hysterectomy than women whose partners were sterilized, but nonbiologic factors were likely explanations.
Young age at sterilization, regardless of the number of children, is a strong predictor of later regretting having had the procedure. Among women sterilized at 18 to 24 years who were followed in the U.S. Collaborative Review of Sterilization, the cumulative probability of requesting information about reversal within 14 years was 40%.
SUMMARY POINTS
· The effectiveness of some contraceptive methods—including oral contraceptives, barrier methods, and fertility awareness-based methods—is highly dependent on whether they are used consistently and correctly. The effectiveness of other methods—including IUDs, implants, and injectables—is less dependent on user characteristics.
· Oral contraceptives have been studied more extensively than any other method. They are nearly 100% effective if taken daily, but they are substantially less effective with typical use. For healthy, nonsmoking women, the health benefits of oral contraceptive use far exceed the health risks.
· Four new contraceptive methods are available in the United States—a contraceptive patch, a combined injectable, a vaginal ring, and an IUD. The first three deliver both an estrogen and a progestogen; the fourth delivers a progestogen alone.
· Many women who desire contraception are also at risk for sexually transmitted infections, including HIV infection. For such women, consistent and correct use of a male latex condom is highly effective in protecting against both pregnancy and some sexually transmitted infections, including HIV infection. Spermicides containing nonoxynol 9 are no longer recommended for preventing sexually transmitted infections, including HIV; frequent use of spermicides may actually enhance the risk of acquiring HIV infection.
· IUDs are highly effective and provide long-term protection against pregnancy. Women at no risk of sexually transmitted infections are at exceedingly low risk of pelvic inflammatory disease from IUD use.
· Male and female sterilization are highly effective but should be chosen only by those who wish to permanently prevent pregnancy. Young age at sterilization is a key risk factor for regretting having had a tubal sterilization.
SUGGESTED READINGS
Centers for Disease Control and Prevention. 1998 guidelines for treatment of sexually transmitted diseases. MMWR 1998;47:1–116.
Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and hormonal contraceptives: collaborative reanalysis of individual data on 53,297 women with breast cancer and 100,239 women without breast cancer from 54 epidemiological studies. Lancet 1996;347:1713–1727.
Farley TMM, Collins J, Schlesselman JJ. Hormonal contraceptives and risk of cardiovascular disease. Contraception 1998;57:211–230.
French RS, Cowan FM, Mansour D, et al. Levonorgestrel-releasing (20 µg/day) intrauterine systems (Mirena) compared with other methods of reversible contraceptives. Br J Obstet Gynaecol 2000;107:1218–1225.
Glasier A. Implantable contraceptives for women: effectiveness, discontinuation rates, return of fertility, and outcome of pregnancies. Contraception2002;65:29–37.
Grimes DA. Intrauterine device and upper-genital-tract infection. Lancet 2000;356:1013–1019.
Hillis SD, Marchbanks PA, Tylor LR, et al. Higher hysterectomy risk for sterilized than nonsterilized women: findings from the U.S. Collaborative Review of Sterilization. Obstet Gynecol 1998;91:241–246.
Hillis SD, Marchbanks PA, Tylor LR, et al. Poststerilization regret: findings from the United States Collaborative Review of Sterilization. Obstet Gynecol1999;93:889–895.
Hubacher D, Lara-Ricalde R, Taylor DJ, et al. Use of copper intrauterine devices and the risk of tubal infertility among nulligravid women. N Engl J Med2001;345:561–567.
International Agency for Research on Cancer (IARC). Hormonal contraception and postmenopausal hormone therapy. Summary of data reported and evaluation. IARC Monographs, Vol 72. Lyon: International Agency for Research on Cancer, 1999.
Kaunitz AM. Injectable contraception. New and existing options. Obstet Gynecol Clin North Am 2000;27:741–780.
Kaunitz AM, Garceau RJ, Cromie MA, and the Lunelle Study Group. Comparative safety, efficacy, and cycle control of Lunelle monthly contraceptive injection (medroxyprogesterone acetate and estradiol cypionate injectable suspension) and Ortho-Novum 7/7/7 oral contraceptive (norethindrone/ethinyl estradiol triphasic). Contraception 1999;60:179–187.
Meirik O, Farley TMM, Sivin I, for the International Collaborative Post-Marketing Surveillance of Norplant. Safety and efficacy of levonorgestrel implant, intrauterine device, and sterilization. Obstet Gynecol 2001;97:539–547.
Moreno V, Bosch FX, Mu–oz N, et al. for the IARC Multicentric Cervical Cancer Study Group. Oral contraceptives and cervical cancer: pooled analysis of the IARC multicentric case-control study. Lancet 2002;359:1093–1101.
Peterson HB, Xia Z, Hughes JM, et al. The risk of ectopic pregnancy after tubal sterilization. N Engl J Med 1997;336:762–767.
Peterson HB, Jeng G, Folger SG, et al. The risk of menstrual abnormalities after tubal sterilization. New Engl J Med 2000;343:1681–1687.
Petitti DB, Piaggio G, Mehta S, et al. Steroid hormone contraception and bone mineral density: a cross-sectional study in an international population. Obstet Gynecol 2000;95:736–744.
Rivera R, Yacobson I, Grimes D. The mechanism of action of hormonal contraceptives and intrauterine contraceptive devices. Am J Obstet Gynecol1999;181:1263–1269.
Roumen FJME, Apter D, Mulders TMT, et al. Efficacy, tolerability and acceptability of a novel contraceptive vaginal ring releasing etonogestrel and ethinyl estradiol. Hum Reprod 2001;16:469–475.
Schmidt JE, Hillis SD, Marchbanks PA, et al. Requesting information about and obtaining reversal after tubal sterilization: findings from the U.S. Collaborative Review of Sterilization. Fertil Steril 2000;74:892–898.
Sibai BM, Odlind V, Meador ML, et al. A comparative and pooled analysis of the safety and tolerability of the contraceptive patch (Ortho Evra/Evra). Fertil Steril 2002;77[Suppl 2]:19–26.
Task Force on Postovulatory Methods of Fertility Regulation. Randomised controlled trial of levonorgestrel versus the Yuzpe regimen of combined oral contraceptives for emergency contraception. Lancet 1998;352:428–433.
Trussell J, Ellertson C, Dorflinger L. Effectiveness of the Yuzpe regimen of emergency contraception by cycle day of intercourse: implications for mechanism of action. Contraception 2003;67:167–171.
Van Damme L, Ramjee G, Alary M, et al. Effectiveness of COL-1492, a nonoxynol-9 vaginal gel, on HIV-1 transmission in female sex workers: a randomized controlled trial. Lancet 2002;360:971–977.
Vandenbroucke JP, Rosing J, Bloemenkamp KWM, et al. Oral contraceptives and the risk of thrombosis. N Engl J Med 2001;344:1527–1535.
von Hertzen H, Piaggio G, Ding J, et al. Low dose mifepristone and two regimens of levonorgestrel for emergency contraception: a WHO multicentre randomised trial. Lancet 2002;360:1803–1810.
Wilcox AJ, Dunson D, Baird DD. The timing of the “fertile window” in the menstrual cycle: day specific estimates from a prospective study. Br Med J2000;321:1259–1262.
Zieman M, Guillebaud J, Weisberg E, et al. Contraceptive efficacy and cycle control with the Ortho Evra/Evra transdermal system: the analysis of pooled data. Fertil Steril 2002;77[Suppl 2]:13–218.
