CHAPTER 5: Contraception and Sterilization

Today, an ever-increasing variety of effective methods is available for fertility regulation. Although none is completely without side effects or potential danger, it remains axiomatic that contraception poses fewer risks than pregnancy (Table 5-1). Contraceptive availability is paramount for the care of women, as approximately half of pregnancies in the United States are unintended (Finer, 2014). Moreover, half of these women are using contraception at the time of conception (Henshaw, 1998). These statistics have prompted a reexamination of contraceptive counseling to prevent unplanned pregnancy (American College of Obstetricians and Gynecologists, 2011; Steiner, 2006).

Methods are now grouped according to their effectiveness. Top-tier or first-tier methods are those that are most effective and are characterized by their ease of use (Fig. 5-1). These methods require only minimal user motivation or intervention and have an unintended pregnancy rate less than 2 per 100 women during the first year of use (Table 5-2). As expected, these first-tier methods provide the longest duration of contraception after initiation and require the fewest number of return visits. Top-tier methods include intrauterine contraceptive devices, contraceptive implants, and various methods of male and female sterilization. A reduction in unintended pregnancies can be better achieved by increasing top-tier method use. Thus, although counseling is provided for all contraceptive methods, common misperceptions regarding some of the top-tier methods—especially intrauterine contraception—can also be dispelled.

Second-tier methods include systemic hormonal contraceptives that are available as oral tablets, intramuscular injections, transdermal patches, or transvaginal rings. In sum, their expected failure rate is 3 to 9 percent per 100 users during the first year. This higher rate likely reflects failure to redose at the appropriate interval. Automated reminder systems for these second-tier methods have been repeatedly shown to have limited efficacy (Halpern, 2013).

Third-tier methods include barrier methods for men and women and fertility awareness methods such as cycle beads. Their expected failure rate is 10 to 20 percent per 100 users in the first year. However, efficacy increases with consistent and correct use.

Fourth-tier methods include spermicidal preparations, which have a failure rate of 21 to 30 percent per 100 first-year users. The withdrawal method is so unpredictable that some conclude that it does not belong among other contraceptive methods (Doherty, 2009).

The World Health Organization (WHO) (2010) has provided evidence-based guidance for the use of all highly effective reversible contraceptive methods by women with various health factors. These guidelines were intended to be modified by individual countries to best serve their populations specific circumstances. Thus, the Centers for Disease Control and Prevention (2010, 2011) published United States Medical Eligibility Criteria (US MEC) for contraceptive use in the United States. These US MEC guidelines are available and updated regularly at the CDC website: http://www.cdc.gov/reproductivehealth/UnintendedPregnancy/USMEC.htm. In the US MEC, many contraceptive methods are classified into six groups by their similarity: combination oral contraceptive (COC), progestin-only pill (POP), depot medroxyprogesterone acetate (DMPA), implants, levonorgestrel-releasing intrauterine system (LNG-IUS), and copper intrauterine device (Cu-IUD). For a given health condition, each method is categorized 1 through 4. The score describes a method’s safety profile for a typical woman with that condition: (1) no restriction of method use, (2) method advantages outweigh risks, (3) method risks outweigh advantages, or (4) method poses an unacceptably high health risk.

Lactation

Among others, lactation is one factor addressed in the US MEC guidelines. Approximately 20 percent of breast-feeding women will ovulate by 3 months postpartum. Ovulation often precedes menstruation, and these women are at risk for unplanned pregnancy. For women who breast feed intermittently, effective contraception should begin as if they were not breast feeding. Moreover, contraception is essential after the first menses unless pregnancy is planned.

Of available methods, Cu-IUD in breast-feeding women has a category 1 or 2 rating (Table 5-3). Women are counseled that effects of the etonogestrel-releasing contraceptive implant (Nexplanon) or LNG-IUS and breast feeding are not known, but studies have mostly shown no adverse association (Gurtcheff, 2011). Because POPs have little effect on lactation, they are also preferred by some for use up to 6 months in women who are exclusively breast feeding. According to the American Academy of Pediatrics and the American College of Obstetricians and Gynecologists (2012), POPs and DMPA may be initiated prior to discharge regardless of breast-feeding status. For the etonogestrel implant, insertion is delayed until 4 weeks postpartum for those exclusively breast-feeding but can be inserted anytime for those not nursing. Combination hormone contraception may begin at 6 weeks following delivery, if breast feeding is well established and the infant’s nutritional status is surveilled. The CDC (2011) revised the US MEC guidelines regarding the use of combined hormonal contraception during the puerperium due to the higher risk of venous thromboembolism (VTE) during these weeks.

Concerns regarding contraceptive steroids and use with breast feeding are based on the theoretical and biologically plausible—but unproven—possibility that systemic progestins may interfere with initial breast milk production. Importantly, contraceptive steroids are not purported to harm the quality of breast milk. Minute quantities of the hormones are excreted in breast milk, but no adverse effects on infants have been reported. In two reviews, authors describe the lack of evidence to support a negative impact of hormonal contraception on lactation (Tepper, 2015; Truitt, 2003). The reviewers concluded that all the studies were of poor to fair quality and that randomized trials are needed.

Adolescence and Perimenopause

Females at both ends of the reproductive spectrum have unique contraceptive needs, which are discussed in Chapters 14 and 21. With adolescents, since the mid-1800s, the age of menarche has dropped. Thus, reproductive function is established many years earlier than psychosocial comprehension regarding the consequences of sexual activity. Such early sexual development may result in intermittent spontaneous sexual encounters with a naïve perception of pregnancy and sexually transmitted disease (STD) risks (Sulak, 1993). Importantly, adolescents have unintended pregnancy rates that approach 85 percent (Finer, 2014). Thus, effective contraception counseling ideally is provided before the onset of sexual activity. In most states, minors have explicit legal authority to consent to contraceptive services, and in many areas, publicly funded clinics provide free contraception to adolescents (Guttmacher Institute, 2014). Moreover, contraception may be provided without a pelvic examination or cervical cancer screening.

In the perimenopause, ovulation becomes irregular and fertility wanes. However, pregnancies do occur, and in women aged >40 years, nearly half of all pregnancies are unintended (Finer, 2011). Importantly, pregnancy with advanced maternal age carries an increased risk for pregnancy-related morbidity and mortality. Women in this group may also have coexistent medical problems that may preclude certain contraceptive methods. Finally, perimenopausal symptoms may be present in this group and may be improved with hormonal contraceptive methods.

Intrauterine Contraception

Fears and concerns with legal liability caused this method to become almost obsolete. However, intrauterine contraception (IUC) has again gained popularity, and IUC use increased from 2 percent in 2002 to 10 percent in 2008 (Fig. 5-2) (Mosher, 2010). Still, this is much lower compared with the worldwide IUC use rate of 14 percent, and specifically with that of China (40 percent) and northern Europe (11 percent) (United Nations, 2013).

Some barriers to IUC use in the United States include cost, politics, and provider failure to offer or encourage use of this method. To reduce the high proportion of unplanned pregnancies, the American College of Obstetricians and Gynecologists (2013b) encourages use of long-acting reversible contraceptives (LARC) for all appropriate candidates, including adolescents. Despite higher up-front costs, the extended span of effective IUC use results in competitive cost effectiveness compared with other contraceptive forms.

Levonorgestrel-releasing Intrauterine System

Three levonorgestrel-releasing intrauterine contraceptives are Food and Drug Administration (FDA)-approved in the United States. Named Mirena, Skyla, and Liletta, devices are T-shaped polyethylene structures with the stem encased by a cylinder containing polydimethylsiloxane and levonorgestrel (Fig. 5-3). The cylinder has a permeable membrane that regulates continuous daily hormone release. The Mirena is currently approved for 5 years following insertion, but evidence supports use for 7 years (Thonneau, 2008). Liletta and Skyla are currently approved for 3 years. In addition to having a lower dose of progestin, Skyla is also marginally smaller in size. Mirena and Liletta have a length of 32 mm and a width of 32 mm, but with Skyla, these same dimensions measure 28 mm.

There are several progestin-mediated mechanisms by which LNG-IUS may prevent pregnancy. The progestin renders the endometrium atrophic; it stimulates thick cervical mucus that blocks sperm penetration into the uterine cavity; and it may decrease tubal motility, thereby preventing ovum and sperm union. The progestin may also inhibit ovulation, but this is not consistent (Nilsson, 1984).

Shown in Table 5-4 are the manufacturer’s contraindications to use of LNG-IUS. Women who have had a previous ectopic pregnancy may be at increased risk for another because of diminished tubal motility from progestin action. In women with uterine leiomyomas, placement of the LNG-IUS may be problematic if the uterine cavity is distorted. In their metaanalysis, Zapata and associates (2010) reported the expulsion rate to be approximately 10 percent in women with coexistent leiomyomas. However, in affected women who retained the device, menstrual blood loss will be lessened in most.

Copper-T 380A Intrauterine Device

Marketed as ParaGard, this device is composed of a stem wrapped with 314 mm² of fine copper wire, and each arm has a 33-mm² copper bracelet—the sum of these is 380 mm² of copper. As shown in Figure 5-3, two strings extend from the base of the stem. The Cu-T 380A is approved for 10 years of continuous use, although it has been shown to prevent pregnancy with continuous use for up to 20 years (Bahamondes, 2005).

The intense local inflammatory response induced in the uterus by copper-containing devices leads to lysosomal activation and other inflammatory actions that are spermicidal (Alvarez, 1988; Ortiz, 1987). In the unlikely event that fertilization does occur, the same inflammatory actions are directed against the blastocyst. And finally, the endometrium becomes hostile for implantation.

Counseling for Intrauterine Contraception

Infection

During the modern renaissance of IUC, several improvements have resulted in safer and more effective models. That said, there are still some unwanted side effects and misconceptions surrounding their use.

First, fear of IUD-associated infections precluded use in the past by young women and those of low parity. Improved device design has mitigated these concerns appreciably. In addition, several well-designed studies have shown that sexual behavior and STDs are important risk factors.

With current devices, insertion generally does not increase the risk for pelvic infection. There is no evidence that prophylactic antibiotics are necessary with insertion for women at low risk for STDs (American College of Obstetricians and Gynecologists, 2014b; Walsh, 1998). Of the less than 1 in 100 women who develop an infection within 20 days of IUD insertion, most have a concomitant unrecognized cervical infection. Accordingly, women at higher risk for sexually transmitted lower genital tract infections are screened either before or at the time of IUD insertion (Centers for Disease Control and Prevention, 2015; Faúndes, 1998; Grimes, 2000). Alternatively, a small number of pelvic infections are presumed to be caused by intrauterine contamination with normal flora at the time of insertion. Thus, antibiotics selected for treatment of any pelvic infection within the early weeks following IUD insertion should be broad-spectrum to adequately cover all these organisms.

Long-term IUC use is not associated with an increased pelvic infection rate in women at low risk for STDs. Indeed, these long-term users have a pelvic infection rate comparable with that of COC users. Any pelvic infection after 45 to 60 days is considered sexually transmitted and appropriately treated as described in Chapter 3. For women who develop an infection associated with an IUD, evidence is insufficient to recommend device removal, although this is commonly done. However, close clinical reevaluation is warranted if an IUD remains (Centers for Disease Control and Prevention, 2015). In women who develop a tuboovarian abscess, the device is removed immediately after parenteral antibiotic therapy is begun.

Special concerns have arisen for women in whom Actinomyces species are identified in the lower genital tract, most commonly during Pap smear cytology reporting. Fiorino (1996) noted a 7-percent incidence in the Pap smears of IUD users compared with a 1-percent incidence in nonusers. Symptomatic pelvic actinomycosis is rare but tends to be indolent and severe.

Currently, in the absence of symptoms, incidental identification of Actinomyces species in cytologic specimens has uncertain significance. Treatment options reviewed by the American College of Obstetricians and Gynecologists (2013b) include: expectant management, an extended course of antibiotics, IUD removal, or antibiotics plus IUD removal. For women with symptomatic infection, the IUD is removed and intensive antibiotic therapy given. Actinomyces is susceptible to antibiotics with gram-positive coverage, notably the penicillins.

Low Parity and Adolescents

Nulliparous IUD candidates were previously precluded from IUC use because of fears of pelvic infection and induced sterility. Current studies indicate that the pelvic infection rate is not different from that discussed earlier (Lee, 1998; Society of Family Planning, 2010). Moreover, expulsion rates in nulliparas are similar to those in multiparas. A higher proportion of nulliparas will request removal of the device because of pain or bleeding, but overall, this population reports high levels of satisfaction with IUC. Specifically, after the first year, 75 to 90 percent continue use. Revised labeling now places no restrictions on IUC use based on parity. In addition, for the same reasons, adolescent IUD candidates may also appropriately select IUC (American College of Obstetricians and Gynecologists, 2014a). Counseling includes clear explanations of the anticipated periprocedural cramping and discomfort.

Human Immunodeficiency Virus-Infected Women

Intrauterine contraception is appropriate for affected women who are otherwise IUC candidates. Neither device type is associated with higher IUD complication rates if used in this population. Moreover, IUDs do not appear to adversely affect viral shedding or antiretroviral therapy efficacy (American College of Obstetricians and Gynecologists, 2012a).

Postabortal or Postpartum Placement

An ideal time to improve successful provision of contraception is immediately following abortion or delivery. For women with an induced or spontaneous first- or second-trimester abortion, IUC can be placed immediately after uterine evacuation.

Insertion techniques depend upon uterine size. After first-trimester evacuation, the uterine cavity length seldom exceeds 12 cm. In these instances, the IUD can be placed using the inserter provided in the package. If the uterine cavity is larger, the IUD can be placed using ring forceps with sonographic guidance. In women for whom an IUD is placed immediately after induced abortion, the repeat induced abortion rate is only one third of the rate of women not choosing immediate IUD placement (Goodman, 2008; Heikinheimo, 2008). As perhaps expected, the risk of IUC expulsion is slightly higher when placed immediately after abortion or miscarriage, but the advantages of preventing unplanned pregnancies seem to outweigh this (Bednarek, 2011; Fox, 2011; Okusanya, 2014).

Insertion of an IUD immediately following delivery at or near term has also been studied. Placement by hand or by using an instrument has a similar expulsion rate (Grimes, 2010). As with postabortion insertion, expulsion rates by 6 months are higher than those in women whose IUD is placed after complete uterine involution. In one study, the expulsion rate in the former group was nearly 25 percent (Chen, 2010). Even in these circumstances, however, immediate placement may be beneficial because in some populations up to 40 percent of women do not return for a postpartum clinic visit (Ogburn, 2005). Finally, postpartum placement is judged to be category 1 or 2 by the US MEC, that is, its advantages consistently outweigh the risks if puerperal infection is absent (see Table 5-3).

Despite these findings, many choose to delay insertion for several weeks postpartum. Insertion at 2 weeks is quite satisfactory, and in the Parkland System Family Planning Clinics, insertion is scheduled at 6 weeks postpartum to ensure complete uterine involution.

Menstrual Changes

Commonly, IUC may be associated with changes in menstrual patterns. Women who choose the Cu-T 380A are informed that increased dysmenorrhea and bleeding with menses may develop. Objectively, no clinically significant hemoglobin changes are generally expected (Tepper, 2013). Treatment with a nonsteroidal antiinflammatory drug (NSAID) will usually diminish the amount of bleeding—even normal amounts—and also relieve dysmenorrhea (Grimes, 2006).

With the LNG-IUS, women are counseled to expect irregular spotting for up to 6 months after insertion and thereafter to expect monthly menses to be lighter or even absent. Specifically, the Mirena device is associated with progressive amenorrhea, which is reported by 30 percent of women after 2 years and by 60 percent after 12 years (Ronnerdag, 1999). As noted in Chapter 8, the LNG-IUS device reduces menstrual blood loss and is an effective treatment for some women with heavy menstrual bleeding (American College of Obstetricians and Gynecologists, 2014e). This is often associated with improved dysmenorrhea.

Expulsion or Perforation

Approximately 5 percent of women will spontaneously expel their IUD during the first year of use. This is most likely during the first month. Accordingly, a woman is instructed to periodically palpate the marker strings protruding from the cervical os. This can be accomplished by either sitting on the edge of a chair or squatting down and then advancing the middle finger into the vagina until the cervix is reached. Following insertion of either IUD type, women are reappointed for a visit within several weeks, usually after completion of menses. At this meeting, any side effects are addressed, and IUD placement is confirmed by visualizing the marker strings. Some recommend barrier contraception to ensure contraception during this first month. This may be especially desirable if a device has been expelled previously.

The uterus may be perforated either with a uterine sound or with an IUD. Perforations may be clinically apparent or silent. Their frequency depends on operator skill and is estimated to be approximately 1 per 1000 insertions (World Health Organization, 1987). In some cases, a partial perforation at insertion is followed by migration of the device completely through the uterine wall. Occasionally, perforation occurs spontaneously.

Marker Strings

In some cases, the IUD marker strings may not be palpated or seen during speculum examination. During the investigation, a nonpregnant patient should use alternative contraception. Possibilities include that the device was expelled silently, the device has partially or completely perforated the uterus, the woman is pregnant and the enlarging uterus has drawn the device upward, or the marker strings are temporarily hidden within the endocervical canal. An IUD should not be considered expelled unless it was seen by the patient.

Initially, an endocervical brush or similar instrument can be used to gently draw the string out of the cervical canal. If this is unsuccessful, then at least two options are available. After pregnancy has been excluded, the uterine cavity is gently probed using an instrument such as Randall stone forceps or a rod with a hooked end. The strings or device will often be found with this method. If not successful, at this juncture, or possibly as a first choice, transvaginal sonography (TVS) is performed. As described in Chapter 2, 3-dimensional TVS offers improved visualization (Moschos, 2011). If the device is not seen within either the uterine cavity or uterine walls, then an abdominal radiograph, with or without a uterine sound in place, may localize it. Another option includes hysteroscopy.

Management decisions depend upon where the device is located and whether there is a coexistent intrauterine pregnancy. First, a device may penetrate the uterine wall in varying degrees. It should be removed, and this approach varies by IUD location. Devices with a predominantly intrauterine location are typically managed by hysteroscopic IUD removal. In contrast, devices that have nearly completely perforated through the uterine wall are more easily removed laparoscopically.

For women with an intraabdominal IUD, an inert-material device located outside the uterus may cause harm, but not universally. Bowel perforations—both large and small—as well as bowel fistulas have been reported. Once identified laparoscopically, these inert devices can easily be retrieved via laparoscopy or less commonly by colpotomy. Conversely, an extrauterine copper-bearing device induces an intense local inflammatory reaction with adhesions. Thus, they are more firmly adhered, and laparotomy may become necessary (Balci, 2010).

In those with pregnancy and an IUD, early pregnancy identification is important. Up to approximately 14 weeks’ gestation, the IUD strings may be visible within the cervix, and if seen, they are grasped to remove the entire IUD. This action reduces subsequent complications such as late abortion, sepsis, and preterm birth (Alvior, 1973). Tatum and colleagues (1976) reported an abortion rate of 54 percent with the device left in place compared with a rate of 25 percent if it was promptly removed. More recently, a study from Israel by Ganer and coworkers (2009) reported pregnancy outcomes from 1988 to 2007 in 292 women who conceived with a Cu-IUD in place. Outcomes were compared in the two groups of women with and without IUD removal as well as with the general obstetrical population. As shown in Table 5-5, in general, the group of women with an IUD left in place had the worst outcomes. Importantly, however, the group in whom the IUD was removed still had significantly worse outcomes compared with those of the general population. Of special note, Vessey and associates (1979) had previously reported that fetal malformations were not increased in pregnancies in which the device was left in place. In the Ganer study, it is particularly worrisome that this rate was doubled compared with women in whom the device was removed. The distribution of malformations was notable in that 12 percent were skeletal malformations. In contrast, there were no chromosomal anomalies identified in fetuses born to women from the two IUD groups.

Because of these findings, if pregnancy continuation is desired, it is recommended that with early pregnancies the IUD be removed. However, if the strings are not visible, attempts to locate and remove the device may result in pregnancy loss. This risk must be weighed against the risk of leaving the device in place. If removal is attempted, TVS can be used. If attempts at removal are followed by evidence for infection, then antimicrobial treatment is begun and is followed by prompt uterine evacuation.

Ectopic Pregnancy

The risk of an associated ectopic pregnancy has been clarified over the past few years. IUC is effective in preventing all pregnancies. Specifically, the contraceptive effect of IUC decreases the absolute number of ectopic pregnancies by half compared with the rate in women who do not use contraception (World Health Organization, 1985, 1987). However, the IUC mechanisms of action are more effective in preventing intrauterine implantation. Thus, if IUC fails, a higher proportion of pregnancies are likely to be ectopic (Furlong, 2002).

Insertion Procedures

Before IUD insertion, the FDA requires that a woman be given a brochure detailing the side effects and apparent risks from its use. Timing of insertion influences the ease of placement as well as pregnancy and expulsion rates. When done toward the end of normal menstruation, when the cervix is usually softer and somewhat more dilated, insertion may be easier, and early pregnancy can be excluded. However, insertion is not limited to this time. For a woman who is sure she is not pregnant and does not want to be pregnant, insertion may be carried out any time during the menstrual cycle. Insertion immediately postpartum or postabortion is also feasible and discussed on page 110.

Prior to insertion, a pelvic examination is completed to identify uterine position and size. Abnormalities are evaluated as they may contraindicate the device. Evidence for infection such as a mucopurulent discharge or significant vaginitis is appropriately treated and resolved before insertion.

For pain management, the most effective method of analgesia has not been established, and patient and provider preference directs selection. Options include NSAIDs, topical lidocaine, or paracervical block. Misoprostol is thought to advance cervical softening to mitigate cervical dilatation pain. However, few studies have adequately evaluated these (Allen, 2009).

At the beginning of the insertion procedure, the cervical surface is cleaned with an antiseptic solution, and a tenaculum is placed on the cervical lip. The uterus is sounded to guide correct depth placement. Specific steps for IUD insertion are outlined and illustrated in Figs. 5-4 and 5-5. During insertion, if there is concern for correct IUD positioning, then placement may be checked by inspection or by sonography. If not positioned completely within the uterus, the device is removed and replaced with a new one. An expelled or partially expelled device should not be reinserted.

Progestin Implants

Contraception can be provided by a progestin-containing device that is implanted subdermally and releases hormone over many years. The devices are coated with a polymer to prevent fibrosis. Several systems have been developed, but only one is available in the United States. The initial implant, the Norplant System, releases levonorgestrel from six Silastic rods. It was withdrawn from the U.S. market, and a fund has been established by the manufacturer to ensure access to patients for removal. Supposedly, the silicone-based rods caused ill-defined symptoms that were reversed with removal. A newer two-rod levonorgestrel system, Jadelle, has received FDA approval but is not marketed or distributed in the United States (Sivin, 2002). Sino-implant II is a structurally and pharmacologically similar system to Jadelle. It is manufactured in China and approved for use by several countries in Asia and Africa (Steiner, 2010).

The implant Nexplanon is currently the only subdermal contraceptive implant marketed in the United States. It is a single-rod subdermal implant containing 68 mg of a progestin—etonogestrel—and covered by an ethylene vinyl acetate copolymer. Nexplanon has replaced the earlier etonogestrel implant, Implanon.

For Nexplanon, contraception is provided by progestin released continuously to suppress ovulation, increase cervical mucus viscosity, and induce endometrial atrophy. The etonogestrel implant will provide contraception for up to 3 years. At this time, the device is removed, and another rod may be placed within the same incision site. Contraindications for this device are similar to those cited for other progestin-containing methods. Specifically, these include pregnancy, thrombosis or thromboembolic disorders, benign or malignant hepatic tumors, active liver disease, undiagnosed abnormal genital bleeding, or breast cancer (Merck, 2014). Importantly, patients are counseled that Nexplanon causes irregular bleeding that does not normalize over time. Thus, women who cannot tolerate unpredictable and irregular spotting or bleeding should select an alternative method.

Nexplanon is inserted subdermally along the biceps groove of the inner arm and 6 to 8 cm from the elbow (Fig. 5-6). Immediately following insertion, the provider and patient should document that the device is palpable beneath the skin. When Nexplanon is removed, this superficial location allows in-office extraction of the implant. Through a small incision large enough to admit hemostat tips, the implant is grasped and removed. If desired, a new rod can be placed through this same incision.

If Nexplanon is not palpable, it can be imaged by radiography, computed tomography (CT), sonography, or magnetic resonance (MR) imaging. Norplant and Jadelle are also radiopaque. This is an advantage compared with Implanon, which is not radiopaque and requires sonography with a 10- to 15-MHz sonographic transducer or MR imaging for identification (Shulman, 2006). In the rare event that an etonogestrel implant cannot be palpated or identified radiologically, the manufacturer can be contacted and arrangements made for etonogestrel level measurement (Merck, 2014).

Permanent Contraception—Sterilization

In 2011 to 2013, surgical sterilization was one of the most commonly reported forms of contraception in childbearing-aged women in the United States (Daniels, 2014). These procedures cannot be tracked accurately because most interval tubal sterilizations and vasectomies are performed in ambulatory surgical centers. However, according to the National Survey of Family Growth, approximately 643,000 female tubal sterilizations are performed annually in the United States (Chan, 2010). The two most commonly employed forms in this country are bilateral tubal ligation—frequently via laparoscopy—and hysteroscopic tubal sterilization. The latter has become popular, and in some settings, it is used in up to half of nonpuerperal female sterilizations (Shavell, 2009).

Over the past 20 years, several important multicenter studies regarding sterilization have been performed by investigators of the Collaborative Review of Sterilization (CREST) and the Centers for Disease Control and Prevention. Data from many of these studies are subsequently described.

Female Tubal Sterilization

This is usually accomplished by occlusion or division of the fallopian tubes to prevent ovum passage and fertilization. According to the National Health Statistics Report, 27 percent of contracepting women in the United States use this method (Jones, 2012). Approximately half of tubal sterilization procedures are performed in conjunction with cesarean delivery or soon after vaginal delivery (MacKay, 2001). Accordingly, this is termed puerperal sterilization. The other half of tubal sterilization procedures are done at a time unrelated to recent pregnancy, that is, nonpuerperal tubal sterilization. This is also termed interval sterilization. In most instances, nonpuerperal tubal sterilization is accomplished via laparoscopy or hysteroscopy.

Tubal Interruption Methods

There are three methods, along with their modifications, that are used for tubal interruption. These include application of various permanent rings or clips to the fallopian tubes; electrocoagulation of a tubal segment; or ligation with suture material, with or without removal of a tubal segment. In a Cochrane review, Lawrie and colleagues (2011) concluded that all of these are effective in preventing pregnancy.

Electrocoagulation is used for destruction of a segment of tube and can be accomplished with either unipolar or bipolar current. Although unipolar coagulation has the lowest long-term failure rate, it also has the highest serious complication rate. For this reason, bipolar coagulation is favored by most (American College of Obstetricians and Gynecologists, 2013a).

Mechanical methods of tubal occlusion can be accomplished with: (1) a silicone rubber band such as the Falope Ring or the Tubal Ring, (2) the spring-loaded Hulka-Clemens clip—also known as the Wolf clip, or (3) the silicone-lined titanium Filshie clip. The steps to these procedures are described in Section 44-2 of the surgical atlas. In a randomized trial of 2746 women, Sokal and associates (2000) compared the Tubal Ring and Filshie clip and reported similar rates of safety and 1-year pregnancy rates of 1.7 per 1000 women. All of these mechanical occlusion methods have favorable long-term success rate.

Suture ligation with tubal segment excision is more often used for puerperal sterilization. Methods include Parkland, Pomeroy, and modified Pomeroy, which are illustrated in Section 43-7.

The type of abdominal entry for sterilization is also variable. Laparoscopic tubal ligation is the leading method used in this country for nonpuerperal female sterilization (American College of Obstetrics and Gynecologists, 2013a). This is frequently done in an ambulatory surgical setting under general anesthesia, and the woman can be discharged several hours later. Alternatively, some choose minilaparotomy using a 3-cm suprapubic incision. This is especially popular in resource-poor countries. With either laparoscopy or minilaparotomy, major morbidity is rare. Minor morbidity, however, was twice as common with minilaparotomy in a review by Kulier and associates (2004). Finally, the peritoneal cavity can also be entered by colpotomy through the posterior vaginal fornix, although this approach is infrequently used.

Counseling

Indications for this elective procedure for sterilization include a request for sterilization with clear understanding that this is permanent and irreversible. Each woman is counseled regarding all alternative contraceptive options and their efficacy. Each woman is also informed regarding her sterilization options, which include laparoscopic or hysteroscopic tubal occlusion or bilateral total salpingectomy. The risks and benefits of each are thoroughly discussed. Many women may also have questions or misunderstanding about possible long-term outcomes after female sterilization. As with any operation, surgical risks are assessed, and occasionally the procedure may be contraindicated.

Risk-reducing Salpingectomy

The Society of Gynecologic Oncology (2013) currently recommends consideration of bilateral total salpingectomy as a preventive measure against serous ovarian and peritoneal cancers. As discussed in Chapter 35, this may be especially relevant for women at greatest risk for these cancers, namely, women with BRCA1 or BRCA2 mutation. Most pelvic serous cancers are thought to originate in the distal fallopian tube. Total salpingectomy may confer up to a 34-percent reduction in endometrioid and serous ovarian cancer rates (Erickson, 2013; Sieh, 2013). If risk-reducing salpingectomy is elected in women with BRCA mutations, the pathology requisition form should state this genetic information. This prompts more thorough tubal specimen sectioning to search for cancer and precancerous lesions, which can be found in the tubes of BRCA mutation carriers.

In low-risk women, because the ovarian cancer risk is less than 2 percent, risk-reducing salpingectomy as an isolated procedure is likely unwarranted. However, if surgery such as hysterectomy or tubal sterilization is planned, women are counseled regarding the risks and benefits of complete fallopian tube excision (Anderson, 2013). As advantages, total salpingectomy may decrease risks for subsequent tubal surgery. As disadvantages, operating time may be increased by 10 minutes, and more importantly, the degree of long-term ovarian blood supply disruption with total salpingectomy is not clearly defined (Creinin, 2014).

Regret

Invariably, some women will later express regrets about sterilization. From a CREST study, Jamieson and coworkers (2002) reported that by 5 years, 7 percent of women undergoing tubal ligation had regrets. This is not limited to female sterilization, as 6 percent of women whose husbands had undergone vasectomy had similar remorse. The cumulative probability of regret within 14 years of sterilization was 20 percent for women aged 30 or younger at sterilization compared with only 6 percent for those older than 30 years (Hillis, 1999).

No woman should undergo tubal sterilization believing that subsequent fertility is guaranteed either by surgical reanastomosis or by assisted reproductive techniques. These are technically difficult, expensive, and not always successful. Pregnancy rates vary greatly depending upon age, the amount of tube remaining, and the technology used. Pregnancy rates range from 50 to 90 percent with surgical reversal (Deffieux, 2011). Of note, pregnancies that result after tubal sterilization reanastomosis are at risk to be ectopic.

Method Failure

Reasons for interval tubal sterilization failure are not always apparent, but some have been identified. First, surgical error may occur and likely accounts for 30 to 50 percent of cases. Second, tubal fistula may complicate occlusion methods. Although usually encountered with electro­coagulation procedures, fistulas from inadequate or defective electric current delivery are now less likely because an amp meter is used routinely. In some cases, sterilization failure may follow spontaneous reanastomosis of the tubal segments. With faulty clips, occlusion can be incomplete. Last, luteal phase pregnancy may occur and describes the situation in which a woman is already pregnant when the procedure is performed. This can often be avoided by scheduling surgery during the menstrual cycle’s follicular phase and by preoperative human chorionic gonadotropin (hCG) testing.

The overall failure rate reported from the CREST studies was 1.3 percent of 10,685 tubal sterilization surgeries. As shown in Figure 5-7, these rates vary for different procedures. And even with the same operation, failure rates vary. For example, with electrocoagulation, if fewer than three tubal sites are coagulated, the 5-year cumulative pregnancy rate approximates 12 per 1000 procedures. However, it is only 3 per 1000 if three or more sites are coagulated (Peterson, 1999). The lifetime increased cumulative failure rates over time are supportive that failures after 1 year are not likely due to technical errors. Indeed, Soderstrom (1985) found that most sterilization failures were not preventable.

With method failure, pregnancies following tubal sterilization have a high incidence of being ectopically implanted compared with the rate in a general gynecologic population. These rates are especially high following electrocoagulation procedures, in which up to 65 percent of pregnancies are ectopic. With failures following other methods—ring, clip, tubal resection—this percentage is only 10 percent (Peterson, 1999). Importantly, ectopic pregnancy must be excluded when any symptoms of pregnancy develop in a woman who has undergone tubal sterilization.

Other Effects

Several studies have evaluated the risk of heavy menstrual bleeding and intermenstrual bleeding following tubal sterilization, and many report no link (DeStefano, 1985; Shy, 1992). In addition, Peterson and coworkers (2000) compared long-term outcomes of 9514 women who had undergone tubal sterilization with a cohort of 573 women whose partners had undergone vasectomy. Risks for heavy menstrual bleeding, intermenstrual bleeding, and dysmenorrhea were similar in each group. Perhaps unexpectedly, women who had undergone sterilization had decreased duration and volume of menstrual flow, they reported less dysmenorrhea, but they had an increased incidence of cycle irregularity.

Other long-term effects have also been studied. It is controversial whether risks for subsequent hysterectomy are increased (Pati, 2000). In a CREST surveillance study, Hillis and associates (1997) reported that 17 percent of women undergoing tubal sterilization subsequently had undergone hysterectomy by 14 years. Although they did not compare this incidence with a control cohort, the indications for hysterectomy were similar to those for nonsterilized women who had undergone a hysterectomy. Women are highly unlikely to develop salpingitis following sterilization (Levgur, 2000). Tubal sterilization appears to have a protective effect against ovarian cancer, but not breast cancer (Westhoff, 2000).

Some psychological sequelae of sterilization were evaluated in a CREST study by Costello and associates (2002). These investigators reported that tubal ligation did not change sexual interest or pleasure in 80 percent of women. In the remaining 20 percent of women who reported a change, 80 percent described the changes to be positive.

Transcervical Sterilization

Mechanical Tubal Occlusion

Various methods of sterilization can be completed using a transcervical approach to reach the tubal ostia. Within each ostium, occlusion is achieved by placing either mechanical devices or chemical compounds.

Mechanical methods employ insertion of a device into the proximal fallopian tubes via hysteroscopy. One system, Essure, is FDA-approved for use in the United States. A second system, Adiana Permanent Contraception, was taken off the market in 2013.

The Essure Permanent Birth Control System consists of a microinsert made of a stainless steel inner coil that is enclosed in polyester fibers. These fibers are surrounded by an expandable outer coil made of nitinol—a nickel and titanium alloy used in coronary artery stents (Fig. 5-8). Fibroblastic proliferation within the fibers causes tubal occlusion. The Essure technique is described in Section 44-16. Analgesia provided by intravenous sedation or paracervical block will successfully alleviate pain (Cooper, 2003). In some women, general anesthesia is preferred.

By far, the overwhelming advantage of hysteroscopic sterilization is that it can be performed in the office. In addition, the procedure times average less than 20 minutes. Abnormal anatomy may preclude procedure completion. One year after placement, Essure contraceptive failure rates range from less than 1 percent to 5 percent (Gariepy, 2014; Munro, 2014).

Three months following device insertion, hysterosalpingography (HSG) is required to confirm complete occlusion (American College of Obstetricians and Gynecologists, 2012b). Prior to undergoing the procedure, patients are counseled on the importance of HSG compliance as up to half of all unplanned pregnancies after hysteroscopic sterilization may be associated with follow-up noncompliance (Cleary, 2013; Levy, 2007). Other reasons for subsequent unplanned pregnancy include incomplete occlusion (10 percent), incorrect HSG interpretation (33 percent), and an established pregnancy prior to the procedure (1 percent) (Jost, 2013; Munro, 2014). In some women, occlusion is incomplete at 3 months, and the study is then repeated at 6 months postoperatively. Until tubal occlusion is established, another method of contraception is needed. Transvaginal sonography has been investigated as an alternative confirmation tool, but currently HSG is required by the FDA (Veersema, 2011).

Pelvic pain after hysteroscopic sterilization is uncommon. If pelvic pain presents soon after the procedure, symptoms are likely to resolve by 3 months postoperatively, around the same time as the follow-up HSG (Arjona Berral, 2014; Yunker, 2015).

As with all sterilization procedures, Essure placement should be considered permanent. The success rate of subsequent spontaneous pregnancy after microsurgery tubal reversal ranges between 0 and 36 percent (Fernandez, 2014; Monteith, 2014).

Chemical Tubal Occlusion

Agents may be placed into the uterine cavity or tubal ostia to incite an inflammatory response to cause tubal occlusion. A method that has been used worldwide in more than 100,000 women consists of using an IUD-type inserter to place quinacrine pellets into the uterine fundus. It is effective, especially considering its simplicity. Pregnancy rates reported by Sokal and colleagues (2008) were 1 and 12 percent at 1 and 10 years, respectively. Although the WHO recommends against its use because of carcinogenesis concerns, it remains an important method for resource-poor countries (Castaño, 2010; Lippes, 2002).

Hysterectomy

For a woman with uterine or other pelvic disease for which hysterectomy may be indicated, this may be the ideal form of sterilization.

Male Sterilization

Vasectomy is performed each year in nearly a half million men in the United States (Magnani, 1999). The office procedure is done with local analgesia and usually takes 20 minutes or less to complete. As illustrated in Figure 5-9, a small incision is made in the scrotum, and the lumen of the vas deferens is disrupted to block sperm traveling from the testes. Compared with female tubal sterilization, vasectomy is 30 times less likely to fail and is 20 times less likely to have postoperative complications (Adams, 2009).

Sterility following vasectomy is not immediate nor is its onset reliably predictable. The time until complete expulsion of sperm stored distal to the vas deferens interruption is variable and requires approximately 3 months or 20 ejaculations (American College of Obstetricians and Gynecologists, 2013a). Thus, another form of contraception is used until azoospermia is documented. Although most recommend that semen be analyzed until two consecutive sperm counts are zero, Bradshaw and coworkers (2001) reported that a single azoospermic semen analysis is sufficient.

Sterilization by vasectomy has a failure rate less than 1 percent (Michielsen, 2010). Causes include failure from unprotected intercourse too soon after vasectomy, incomplete vas deferens occlusion, or recanalization following suitable separation.

Fertility Restoration

After vasectomy, fertility may be restored either by surgical reanastomosis techniques or by sperm retrieval from the testis. Surgical reversal techniques and perioperative evaluation have been reviewed by the American Society for Reproductive Medicine (2008). Sperm retrieval combined with in vitro fertilization techniques avoids such reversal surgeries and is described in Chapter 20. From their review, Shridharani and Sandlow (2010) concluded that microsurgical reversal is cost effective, but comparative trials with sperm retrieval methods are needed.

Long-term Effects

Regret of sterilization was discussed on page 116. Other than this, long-term consequences are rare (Amundsen, 2004). However, because antibodies directed at spermatozoa frequently develop in these men, there were initial concerns that these might cause systemic disease. Putative risks were analyzed by Köhler and coworkers (2009) and include cardiovascular disease, immune-complex disorders, psychological changes, male genital cancers, and frontotemporal dementia. Their findings and those of others are not convincing for an increased risk of cardiovascular disease or accelerated atherogenesis from vasectomy (Schwingl, 2000). Moreover, rates of testicular or prostate cancers do not appear increased with this procedure (Holt, 2008; Köhler, 2009).

Contraceptives considered to be very effective are the hormone-containing preparations that include combination oral contraceptives, progestin-only contraceptive pills, and contraceptives with estrogens and/or progestins that are made systemically available by injection, transdermal patch, or intravaginal ring. When used as intended, these methods are highly effective, however, their efficacy is user dependent. Thus, typical use considers each woman’s compliance with taking a daily pill, changing transdermal patches or rings, or presenting for an injection (see Table 5-2). Such “real world” use significantly diminishes their efficacy, and for women in the United States, these contraceptives have a first-year pregnancy rate of 3 to 9 per 100 users.

Combined Hormonal Contraceptives

These are contraceptives that contain an estrogen and a progestin. As such, several underlying conditions are considered contraindications to their use (Table 5-6). Combined hormonal contraceptives (CHCs) are available in the United States in three formats—oral contraceptive pills, the transdermal patch, and the intravaginal contraceptive ring. Because of limited data for the transdermal and transvaginal methods relative to that for COCs, their use is usually considered along with those of combined oral contraceptives.

Pharmacology

There are multiple contraceptive actions of CHCs. The most important is to inhibit ovulation by suppression of hypothalamic gonadotropin-releasing hormone, which prevents pituitary secretion of follicle-stimulating hormone (FSH) and luteinizing hormone (LH). Estrogens suppress FSH release and stabilize the endometrium to prevent intermenstrual bleeding—referred to as breakthrough bleeding in this setting. Progestins inhibit ovulation by suppressing LH, they thicken cervical mucus to retard sperm passage, and they render the endometrium unfavorable for implantation. Thus, CHCs have contraceptive effects from both hormones and, when taken daily for 3 out of every 4 weeks, provide virtually absolute protection against conception.

Until recently, there were only two estrogens available for use in oral contraceptives in the United States. These were ethinyl estradiol and its less commonly used 3-methyl ether, mestranol. In 2010, a third estrogen compound—estradiol valerate—was approved by the FDA. Most currently available progestins are 19-nortestosterone derivatives. However, drospirenone is a spironolactone analogue, and the dose of drospirenone in COCs currently marketed has properties similar to a 25-mg dose of spironolactone (Seeger, 2007). Drospirenone displays antiandrogenic activity, and its antimineralocorticoid properties may, in theory, cause potassium retention, leading to hyperkalemia. Thus, drospirenone is not prescribed for those with renal or adrenal insufficiency or with hepatic dysfunction. Moreover, monitoring of serum potassium levels is recommended in the first month for patients chronically treated concomitantly with any drug associated with potassium retention (Bayer HealthCare Pharmaceuticals, 2012). Several studies have shown improvement in symptoms for women with premenstrual dysphoric disorder (PMDD) who use the drospirenone-containing COC, Yaz (Lopez, 2012; Yonkers, 2005). For this pill, the FDA has approved its indications to include treatment of premenstrual syndrome and moderate acne vulgaris for women requesting oral contraception.

Progestins were initially selected for their progestational potency. However, they are often compared, marketed, and prescribed based on their presumed estrogenic, antiestrogenic, and androgenic effects. However, the doses of progestins used in combined contraceptive formulations are so low that none of these purported negative side effects are actually manifested clinically. In fact, an important effect of CHCs is increased production of sex hormone-binding globulin (SHBG) by the liver and production is promoted mostly by the estrogen component of CHCs. Elevated SHBG levels lower serum free testosterone levels and thereby limit 5α-reductase, the enzyme necessary to convert testosterone to its active form, dihydrotestosterone. For this reason, CHCs can be expected to have salutary effects on androgen-related conditions such as acne (del Marmol, 2004; Rosen, 2003; Thorneycroft, 1999).

Combined Oral Contraceptive Pills

Formulations

Hormone-containing contraceptive pills recently had a celebrated 50^th anniversary in this country. These various preparations—used by approximately 16 million women in the United States in 2013—are popularly known by several names (United Nations, 2013). Among others, these include: combination oral contraceptives (COCs), birth control pills (BCPs), oral contraceptives (OCs), oral contraceptive pills (OCPs), and most simply, the pill.

Combination oral contraceptives are marketed in a bewildering variety shown in (Table 5-7 and Fig. 5-10). Currently, the daily estrogen content in most COCs varies from 20 to 50 μg of ethinyl estradiol, and most pills contain 35 μg or less. Of note, in 2011, the FDA approved the first pill containing only 10 μg of ethinyl estradiol—Lo Loestrin Fe. For current formulations, the lowest acceptable dose is governed by the ability to prevent unacceptable breakthrough bleeding.

With COCs, the progestin dose can be constant throughout the cycle—monophasic pills—but the dose frequently is varied—biphasic and triphasic pills. In some of these, the estrogen dose is also varied during the cycle. Multiphasic pills were developed to reduce the amount of total progestin per cycle without sacrificing contraceptive efficacy or cycle control. The reduction is achieved by beginning with a low dose of progestin and increasing it later in the cycle. Theoretically, the lower total dose minimizes the intensity of progestin-induced metabolic changes and adverse side effects. Disadvantages of multiphasic formulations include confusion caused by the multicolored pills—in some brands there are five colors. Another side effect is breakthrough bleeding or spotting, which likely is increased compared with monophasic pills (Woods, 1992).

In a few COCs, inert placebo pills have been replaced by tablets containing iron. These have the suffix Fe added to their name. In addition, Beyaz has a form of folate—levomefolate calcium—within both its active and placebo pills.

Administration

Ideally, women would begin COCs on the first day of a menstrual cycle, in which case an additional contraceptive method is unnecessary. A more traditional schedule—the Sunday start—requires pill initiation on the first Sunday following the onset of menses. If menses begin on a Sunday, then pills are started that day. Last, a quick start method may be used in which pills are started on any day of the cycle, commonly the day prescribed. This approach improves short-term compliance (Westhoff, 2002, 2007a). Both Sunday start and quick start methods require use of an additional method for 1 week to protect against conception.

To obtain maximum efficacy and promote regular use, most manufacturers offer dispensers that provide 21 sequential color-coded tablets containing hormones, followed by seven inert tablets of another color (see Fig. 5-10B). Some newer, lower-dose pill regimens continue active hormones for 24 days, followed by 4 days of inert pills (see Fig. 5-10C). The goal of these 24/4 regimens is to improve the efficacy of very low-dose COCs. Importantly, for maximum contraceptive efficiency, each woman should adopt an effective scheme for ensuring daily—or nightly—self-administration.

During COC use, if one dose is missed, conception is unlikely with higher-dose monophasic COCs. When this is recognized, taking that day’s pill plus the missed pill will minimize breakthrough bleeding. The remainder of the pill pack is then completed with one pill taken daily.

If several doses are missed, or if a dose is missed with the lower-dose pills, then two pills are taken but an effective barrier technique is added for the subsequent 7 days. The remainder of the pack is completed with one pill taken daily. Alternatively, a new pack can be started and a barrier method added as additional contraception for a week. With any scenario of missed pills, if withdrawal bleeding does not occur during the placebo pills, the pills are continued, but the woman should seek medical attention to exclude pregnancy. Fortunately, CHCs are not teratogenic if taken accidentally during early pregnancy (Lammer, 1986).

Transdermal System

There is one transdermal system available in the United States—Ortho Evra patch. The patch has an inner layer with an adhesive and hormone matrix and an outer water-resistant layer. The patch is applied to the buttocks, upper outer arm, lower abdomen, or upper torso but avoids the breasts. It delivers daily a dose of 150 μg of the progestin norelgestromin and 20 μg of ethinyl estradiol. A new patch is applied each week for 3 weeks, followed by a patch-free week to allow withdrawal bleeding.

In a randomized trial by Audet and associates (2001), the patch was slightly more effective than a low-dose oral contraceptive—1.2 versus 2.2 pregnancies per 100 woman years. Patch replacement was required for either complete—1.8 percent, or partial detachment—2.8 percent. In approximately 3 percent of women, a severe application-site reaction precluded further use.

Pooled data suggest that women who weigh 90 kg or more are at increased risk for pregnancy with the patch (Zieman, 2002). Other metabolic and physiologic effects mirror those seen with low-dose COCs, with the caveat that accumulated experience is limited. The patch is suitable for women who prefer weekly applications to daily dosing and who meet the other criteria for CHC administration.

Concerns have been raised that CHC delivered by the patch may be associated with an increased risk for VTE and other vascular complications. This followed reports that patch use was associated with increased hepatic synthesis of procoagulants compared with COC or vaginal ring use (Jensen, 2008; White, 2006). Although peak serum estrogen levels were lower with patch versus COC use, total exposure was greater—a relatively increased net estrogen effect (Kluft, 2008; van den Heuvel, 2005). Despite lack of a convincing clinical association, in 2008, the FDA ordered labeling for the patch to state that users may be at increased risk for developing VTE. Plaintiff attorneys followed with lawsuits that inevitably curtailed use of the patch method (Phelps, 2009). To date, conclusive evidence for increased morbidity with patch use compared with other CHC use is lacking (Jick, 2006, 2010a,b).

Transvaginal Ring

There is one intravaginal hormonal contraceptive available in the United States—NuvaRing. It is a flexible polymer ring with a 54-mm outer diameter and a 50-mm inner diameter (Fig. 5-11). Its core releases a daily dose of 15 μg ethinyl estradiol and 120 μg of the progestin etonogestrel. These doses effectively inhibit ovulation, and the failure rate is reported to be 0.65 pregnancies per 100 woman-years (Mulders, 2001; Roumen, 2001).

Prior to dispensing, the pharmacy must keep rings refrigerated. Once dispensed, their shelf life is 4 months. The ring is initially inserted within 5 days after the onset of menses. It is removed after 3 weeks for 1 week to allow withdrawal bleeding. After this, a new ring is inserted. Breakthrough bleeding is uncommon. Up to 20 percent of women and 35 percent of men reported being able to feel the ring during intercourse. If bothersome, the ring may be removed for coitus, but it should be replaced within 3 hours.

Extended Cycle Contraception

The use of CHCs continuously for more than 28 days has become increasingly popular. Their benefits include decreased episodes of cyclic bleeding, fewer menstrual symptoms, and lower costs. Several formulations are available (see Table 5-7). Although these prepackaged cycle formulations are available, extended cycle contraception can also be achieved in other ways. The standard 21- or 28-day COC packs, with the placebo pills discarded, can be used continuously. Also, either the transdermal patch or the vaginal ring can be used without the 1-week hormone-free interval.

Several factors unique to extended-cycle CHCs are important. Some of these are shared with continuous progestin contraceptive methods such as implants or injections. The principal change is loss of menstrual normalcy that manifests as less frequent, lighter, and generally unpredictable bleeding episodes. For example, amenorrhea of 6 months or more is reported to affect 8 to 63 percent of extended cycle users. Although considered a benefit by most women, it is far from a guaranteed one. More often, women have fewer bleeding episodes per month. This allows repair of associated anemia in those who had heavy menstrual bleeding prior to extended-cycle use.

However, these characteristics also render some women reluctant to use this method, as it may be considered “unnatural” to miss monthly menses. Some are concerned that amenorrhea may be a sign of pregnancy, may diminish future fertility, or may increase endometrial neoplasia. Rather, findings support a decreased risk for endometrial malignancy associated with cyclic CHC use. Thus on a biological basis, it seems reasonable to conclude that this protective effect would also apply to continuous CHC use. Moreover, women who use a continuous CHC method report fewer menstrual symptoms that include headaches, fatigue, bloating, and dysmenorrhea compared with women using cyclic contraceptives (Edelman, 2014). Moreover, hypothalamic-pituitary-ovarian suppression is greater with continuous use and reduces the possibility of escape ovulation caused by delayed start of a new contraceptive cycle.

Drug Interactions

Interactions between CHCs and various other medications take two forms. First, hormonal contraceptives may interfere with the actions of some drugs shown in Table 5-8. In contrast, some drugs shown in Table 5-9 may decrease the contraceptive effectiveness of CHCs. Mechanisms for these are multiple and frequently cannot be identified. In some cases, genes coding for cytochrome oxidase system enzymes are either stimulated or suppressed.

Resulting pharmacokinetic changes can decrease serum contraceptive steroid concentrations, but the ultimate effect on ovulation suppression is unknown. However, with current information, these interactions often require that the dose of contraceptive or that of the other drug be adjusted to ensure efficacy.

Combined Hormonal Contraception and Medical Disorders

A summary of health benefits associated with CHCs is found in Table 5-10. Despite this, interactions of CHCs with some chronic medical disorders may constitute relative or absolute contraindication to CHC use. These are described in the following sections.

Obese and Overweight Women

In general, CHCs are highly effective in obese women (Lopez, 2013). However, obesity may result in altered pharmacokinetics of some CHC methods. That said, data regarding overweight women are conflicting regarding increased pregnancy risk due to decreased CHC efficacy from lowered bioavailability (Brunner, 2005; Edelman, 2009; Holt, 2005; Westhoff, 2010). Importantly, in some women, obesity may be synergistic with other conditions, described next, that may render CHCs a less optimal contraceptive method.

Excessive weight gain is a concern with use of any hormonal contraceptive. Gallo and associates (2014) again concluded in their review that available evidence was insufficient to determine the influence of CHCs on weight gain, although no large effect was obvious.

Diabetes Mellitus

Higher-dose COCs were associated with insulin antagonistic properties, particularly those mediated by progestins. However, with current low-dose CHCs, these concerns have been mitigated. In healthy women, large, long-term prospective studies have found that COCs do not increase the risk for diabetes (Rimm, 1992). Moreover, these agents do not appear to increase the risk for overt diabetes in women with prior gestational diabetes (Kjos, 1998). Last, use of these contraceptives is approved for nonsmoking diabetic women who are younger than 35 years and who have no associated vascular disease (American College of Obstetricians and Gynecologists, 2013d).

Cardiovascular Disease

In general, severe cardiovascular disorders limit the use of CHCs. For less severe disorders, however, current formulations do not increase associated risks.

First, low-dose CHCs do not appreciably increase the absolute risk of clinically significant hypertension (Chasan-Taber, 1996). However, it is common practice for patients to return 8 to 12 weeks following CHC initiation for evaluation of blood pressure and other symptoms. For those with already established chronic hypertension, CHC use is permissible in those with well-controlled, otherwise uncomplicated hypertension (American College of Obstetricians and Gynecologists, 2013d). Severe forms of hypertension, especially those with end-organ involvement, usually preclude CHC use.

Women who have had a documented myocardial infarction should not be given CHCs. That said, these contraceptives do not increase the de novo risk for myocardial ischemia in nonsmoking women younger than 35 years (Margolis, 2007; Mishell, 2000; World Health Organization, 1997). Smoking by itself, however, is a potent risk factor for ischemic heart disease, and CHCs used after age 35 years act synergistically to increase this risk.

Cerebrovascular Disorders

Women who have had either an ischemic or hemorrhagic stroke should not use CHCs. But the incidence of strokes in nonsmoking young women is low, and use of CHCs does not increase the risk for either type of stroke (World Health Organization, 1996). This form of vascular disorder is more commonly encountered in those who smoke, have hypertension, or have migraine headaches with visual aura and who use CHCs (MacClellan, 2007).

Migraine headaches may be a risk factor for strokes in some young women. Curtis and coworkers (2002) reported that women using COCs who had migraine headaches with aura had a two- to fourfold increased risk for stroke compared with nonusers. Because of this, the WHO (2010) recommends against CHC use in this subset of migraineurs. Alternatively, the American College of Obstetricians and Gynecologists (2013d), because the absolute risk is low, has concluded that CHCs may be considered for young nonsmoking women who have migraine headaches without focal neurologic changes. For many of these women, an intrauterine contraceptive method or a progestin-only pill may be more appropriate (World Health Organization, 2010).

Venous Thromboembolism

Early in CHC history, it was apparent that deep-vein thrombosis and pulmonary embolism risks were significantly increased in women who used these contraceptives (Realini, 1985). These risks were found to be estrogen-dose related and have been appreciably lowered with evolution of low-dose formulations that contain only 10 to 35 μg of ethinyl estradiol (Westhoff, 1998). Of note, a possible increased VTE risk with drospirenone-containing COCs has been shown in two studies, and the FDA has encouraged an assessment of benefits and of VTE risks in users of these pills (Food and Drug Administration, 2012; Jick, 2011; Parkin, 2011).

Mishell and coworkers (2000) concluded that VTE risk is three- to fourfold higher in current COC users compared with nonusers. However, the risk without contraception is low—approximately 1 per 10,000 woman years—and thus the incidence with CHCs is only 3 to 4 per 10,000 woman years. Importantly, these CHC-enhanced risks appear to dissipate rapidly once contraceptive treatment is discontinued. And, of equal importance, these VTE risks are still lower than those estimated during pregnancy, which has an incidence of 5 to 6 per 10,000 woman years.

Several cofactors increase the incidence of VTE in women using estrogen-containing contraceptives or those who are pregnant or postpartum. These include one or more of the many thrombophilias, which include protein C or S deficiency or factor V Leiden mutation (Chap. 39) (Mohllajee, 2006). Other factors that raise VTE risks are hypertension, obesity, diabetes, smoking, and a sedentary lifestyle (Pomp, 2007, 2008).

Older studies indicated a twofold increased risk for perioperative VTE in CHC users (Robinson, 1991). Data are lacking with the low-dose formulations currently used, and thus the American College of Obstetricians and Gynecologists (2013c,d) recommends balancing VTE risks against those of unintended pregnancy during the 4 to 6 weeks required preoperatively for thrombogenic effects of CHCs to dissipate.

Systemic Lupus Erythematosus

The use of CHCs in women with otherwise uncomplicated systemic lupus erythematosus (SLE) has been the “poster child” for evidence-based clinical research. In the past, and with good reason, CHCs were contraindicated in women with SLE. This was because of their underlying high risk to develop venous and arterial thrombosis along with the thrombogenic effects of older high-dose COC pills. The safety of the low-dose modern COCs in many women with SLE was shown in two randomized trials (Petri, 2005; Sánchez-Guerrero, 2005). Use of CHCs in women with SLE was reviewed by Culwell and colleagues (2009). Importantly, CHCs are not appropriate in women with SLE who have positive testing for antiphospholipid antibodies or have other known contraindications to CHC use. Affected women with these antibodies have increased clotting risks.

Seizure Disorders

Approximately 1 million women of reproductive age in the United States are diagnosed with some form of epilepsy. As shown in Tables 5-8 and 5-9, metabolism and clearance of some CHCs are appreciably altered by some, but not all, of the commonly used anticonvulsants. One mechanism with several antiepileptic drugs is potent induction of cytochrome P450 system enzymes. In turn, this increases contraceptive steroid metabolism, and serum levels of these decrease by as much as half (American College of Obstetricians and Gynecologists, 2013d; Zupanc, 2006).

These metabolic interactions usually do not result in increased seizure activity. One possible exception is combined use of CHCs and monotherapy with the anticonvulsant lamotrigine. Serum anticonvulsant levels are decreased by up to 50 percent, which may increase seizure risks (Gaffield, 2011).

Evidence-based guidelines for use of contraceptives by women with epilepsy are listed in the US MEC. Use of CHCs in epileptic women is rated as category 3, that is, theoretical or proven risks usually outweigh the method advantages. CHCs used concurrently with anticonvulsants may reduce contraceptive or anticonvulsant effectiveness. Thus, epileptic women using cytochrome P450 enhancing anticonvulsants are counseled regarding alternate contraceptive methods if feasible. If not, a COC containing at least 30 μg of ethinyl estradiol should be used. For those using lamotrigine monotherapy, CHCs are not recommended.

Although they are not CHCs, progestin-only containing preparations are also affected by use of anticonvulsants that induce the cytochrome P450 enzyme system. These result in decreased serum progestin levels and lower rates of effective ovulation suppression and pose an unacceptable risk of unplanned pregnancy.

Liver Disease

Both estrogens and progestins have known effects on hepatic function. Cholestasis and cholestatic jaundice, which develop more commonly in pregnancy, are also infrequently induced by CHC use. Because susceptibility is likely due to an inherited gene mutation of bilirubin transport, cholestasis with CHCs is more likely in women affected during a pregnancy. Discontinuing CHCs typically resolves symptoms. Whether these cholestatic effects of CHCs also raise risks for subsequent cholelithiasis and cholecystectomy is unclear. Any increased risk is likely to be small, and the known effects of increasing parity on gallbladder disease must also be considered.

Regarding women with viral hepatitis or cirrhosis, the WHO has provided recommendations (Kapp, 2009b). For women who have active hepatitis, CHCs should not be initiated, but these may be continued in women who experience a flare of their liver disease while already taking CHCs. Use of progestin-only contraception in these women is not restricted. With cirrhosis, mild compensated disease does not limit the use of CHCs or progestin-only methods. However, in those with severe decompensated disease, all hormonal types are avoided.

Neoplastic Diseases

Stimulatory effects of sex steroids on some cancers are a concern. It would appear, however, that overall these hormones do not cause cancer (Hannaford, 2007). A report from the Collaborative Group on Epidemiological Studies of Ovarian Cancer (2008) verified earlier studies that showed a protective effect against endometrial and ovarian cancers (Cancer and Steroid Hormone Study, 1987a,b). However, this protection wanes as duration from pill discontinuance increases (Tworoger, 2007). Reports concerning possible increased risks for premalignant and malignant changes of the liver, cervix, and breast are conflicting and are presented next.

First, hepatic focal nodular hyperplasia and benign hepatic adenomas were previously linked with older higher-dose estrogen-containing COCs. However, studies that evaluated women taking contemporary low-dose COCs reported no such association (Hannaford, 1997; Heinemann, 1998). Similarly, earlier correlations between CHCs and hepatocellular carcinoma have been refuted by the multicenter WHO Study (1989) and by Maheshwari and coworkers (2007). For women with known tumors, COCs may be used in those with focal nodular hyperplasia, but avoided in those with benign hepatic adenoma and hepatocellular carcinoma (Kapp, 2009a).

Second, cervical dysplasia and cervical cancer rates are increased in COC users. These risks increase with duration of use. But, according to the International Collaboration of Epidemiological Studies of Cervical Cancer (2007), if COC use is discontinued, by 10 years the risk becomes comparable with that of never-users. The reasons for this neoplasia risk are speculative and may be related to more frequent human papillomavirus (HPV) exposure because of decreased use of barrier methods. It may also be related to the more frequent cytologic screening that COC users may have. Moreover, COCs may increase persistence of HPV infection and HPV oncogene expression (de Villiers, 2003). Importantly, if cervical dysplasia is treated, the recurrence rate is not increased in CHC users.

Last, breast cancer is stimulated by female sex steroid hormones, but it is still unclear whether CHCs have an adverse effect on tumor growth or development. The Collaborative Group on Hormonal Factors in Breast Cancer (1996) analyzed data from studies that included more than 53,000 women with breast cancer and 100,000 nonaffected women. They found a significant 1.24-fold increased risk for current COC users. This risk decreased to 1.16 for those 1 to 4 years after discontinuing COCs and 1.07 for those at 5 to 9 years. The risks were not influenced by age at first use, duration of use, family history of breast cancer, first use prior to pregnancy, or the dose or type of hormone used. This lack of correlation serves to question any causal role of COCs in breast tumorigenesis.

The Collaborative Group investigators also found that COC-associated breast tumors tended to be to be less aggressive and that cancers were detected at earlier stages. They suggested that the increased cancer diagnosis may have been because of more intensive surveillance among users. In a case-control study—4575 cases and 4682 controls—there was no relationship found with either current or past COC use and breast cancer (Marchbanks, 2002). Finally, women heterozygous for BRCA1 or BRCA2 gene mutations have not been shown to have an increased incidence of breast or ovarian cancer with COC use (Brohet, 2007). With regard to benign breast disease, Vessey and Yeates (2007) reported that COC use apparently lowered the relative risk.

HIV Infections and Antiretroviral Therapy

Women with human immunodeficiency virus (HIV) infection or acquired immunodeficiency syndrome (AIDS) require special considerations regarding contraceptive use. As outlined by the American College of Obstetricians and Gynecologists (2012a), affected women need highly effective contraception that meets several criteria. It must be compatible with highly active antiretroviral therapy (HAART), should provide a low risk for acquiring STDs, and must not increase their risk of transmitting HIV to their partners.

Although CHCs are safe for use in HIV-positive women, their metabolism may be variably altered by some HAART regimens in current use. Details of various HAART regimen interactions with CHCs are available at the University of California, San Francisco HIV InSite website: http://hivinsite.ucsf.edu/insite?page=ar-00-02.

Progestin-Only Contraceptives

Contraceptives that contain only a progestin were developed to obviate the unwanted side effects of estrogens. Progestins can be delivered by several routes that include tablets, injections, intrauterine devices, and subdermal implants.

Progestin-Only Pills

Also called mini-pills, POPs are taken daily. They do not reliably inhibit ovulation, but instead thicken cervical mucus and decidualize and atrophy the endometrium. Because mucus changes do not persist beyond 24 hours, to be maximally effective, a pill is ideally taken at the same time daily. Their use has not achieved widespread popularity because of a much higher incidence of irregular bleeding and a slightly higher pregnancy rate than that seen with CHCs (see Table 5-2).

POPs have minimal if any effect on carbohydrate metabolism and coagulation factors. They do not cause or exacerbate hypertension and thus may be ideal for some women at increased risk for other cardiovascular complications. Such women include those with a history of thrombosis or migraine headaches or smokers older than 35 years. Because they do not impair milk production, POPs are suitable for lactating women. When used in combination with breast feeding, POPs are virtually 100-percent effective for up to 6 months (Betrabet, 1987; Shikary, 1987).

POPs should not be taken by women with unexplained uterine bleeding, breast cancer, hepatic neoplasms, pregnancy, or active severe liver disease (Janssen-Ortho, 2014).

Compliance is essential to POP use. If a pill is taken even 4 hours late, an additional form of contraception must be used for the next 48 hours. This may contribute to another major drawback, which is a higher risk for contraceptive failure compared with CHCs. Moreover, with failure, the proportion of pregnancies that are ectopic is increased (Sivin, 1991). POP effectiveness is also decreased by some medications, and in some instances POPs should be avoided (see Tables 5-8 and 5-9).

Another disadvantage is irregular uterine bleeding. This may be characterized by amenorrhea, intermenstrual bleeding, or prolonged heavy menstrual bleeding. As with other progestin-containing contraceptive methods, functional ovarian cysts develop with a greater frequency, although they usually do not require intervention (Hidalgo, 2006; Inki, 2002).

Injectable Progestins

Formulations

There are three injectable depot progesterone preparations that are used worldwide. This method is popular in the United States and is used by approximately 6 percent of women choosing a contraceptive. Injectable progestins have mechanisms of action similar to those for oral progestins and include increased cervical mucus viscosity, creation of an endometrium unfavorable for implantation, and unpredictable ovulation suppression.

Injectable preparations include depot medroxyprogesterone acetate—marketed as Depo-Provera. A 150-mg dose is given by intramuscular injection every 90 days. A derivative of DMPA is marketed as depo-subQprovera 104, and a 104-mg dose is given subcutaneously every 90 days. Because absorption is slower with subcutaneous injections, the 104-mg dose is equivalent to the 150-mg intramuscular preparation (Jain, 2004). With either method, if the initial dose is given within the first 5 days following menses onset, no back-up contraception is necessary (Haider, 2007). A third injectable depot progestin that is not currently available in the United States is norethindrone enanthate, which is marketed as Norgest, and a 200-mg dose is injected intramuscularly every 2 months.

Injectable progestins have contraceptive efficacy equivalent or better than that of COCs. With perfect use, DMPA has pregnancy rates of 0.3 percent, but typical-use failure rates are as high as 7 percent at 12 months (Kost, 2008; Said, 1986). Depot progesterone does not suppress lactation, and iron-deficiency anemia is less likely in long-term users because of less menstrual bleeding. Progestin injectables should not be taken by women with pregnancy, unexplained uterine bleeding, breast cancer, active or history of thromboembolic disease, cerebrovascular disease, or significant liver disease (Pfizer, 2014).

As with most progestin-only contraceptive methods, DMPA does not significantly affect lipid metabolism, glucose levels, hemostatic factors, liver function, thyroid function, and blood pressure (Dorflinger, 2002). Moreover, these have not been shown to increase the risk for thromboembolism, stroke, or cardiovascular disease (Mantha, 2012; World Health Organization, 1998). Despite this, manufacturer prescribing information often lists thrombosis or thromboembolic conditions as contraindications. However, for individuals with these disorders, US MEC considers progestin-containing methods category 2.

Notable Effects

Patients interested in DMPA use should be familiar with its potential effects and side effects. First, as is typical of progestin-only contraception, DMPA usually causes irregular menstrual-type bleeding. Cromer and coworkers (1994) reported that one fourth of women discontinued its use in the first year because of irregular bleeding. Amenorrhea may develop after extended use, and women are counseled about this benign effect.

Prolonged ovulation suppression may also persist after DMPA injections are stopped. In an earlier study by Gardner and Mishell (1970), one fourth of women did not resume regular menses for up to a year. Accordingly, DMPA may not be the best choice for women who plan to use contraception only briefly before attempting conception.

Bone mineral density can also be significantly diminished because of lowered estrogen levels and is most worrisome in long-term users. This loss is particularly relevant for adolescents because bone density increases most rapidly from ages 10 to 30 years (Sulak, 1999). Additionally, decreased bone mineral density may be a concern for perimenopausal women, who will shortly be entering the menopause, a time of known accelerated bone loss. These concerns prompted the FDA in 2004 to require a black-box warning that DMPA “should be used as a long-term birth control method—longer than 2 years—only if other birth-control methods are inadequate.” There are some mitigating factors that balance this concern. First, although bone loss is greatest during the first 2 years, it subsequently slows appreciably. Second, most bone lost during contraceptive use is restored within 5 years after its discontinuance (Clark, 2006; Harel, 2010). In sum, the American College of Obstetricians and Gynecologists (2014c) has concluded that concerns of bone density loss should not prevent or limit use of this contraceptive method.

Of potential cancer risks, cervical carcinoma in situ rates are possibly increased with DMPA use. However, risks for cervical cancer or for hepatic neoplasms are not higher with this method (Thomas, 1995). Advantageously, ovarian and endometrial cancers are decreased (Kaunitz, 1996; World Health Organization, 1991). In addition, Skegg and colleagues (1995) pooled the results of the New Zealand and WHO case-control studies that included almost 1800 women with breast cancer. Compared with 14,000 controls, DMPA contraceptive use was associated with a twofold cancer risk in the first 5 years of use. However, the overall risk was not increased.

Of other effects, some women report breast tenderness with DMPA use. Depression has also been reported, but a causal link is unproven. Finally, although weight gain is often attributed to depot progestins, not all studies have shown this (Bahamondes, 2001; Mainwaring, 1995; Moore, 1995; Taneepanichskul, 1998). Beksinska and coworkers (2010) reported that adolescents who used intramuscular DMPA gained 2.3 kg more weight during a 4- to 5-year interval compared with weight gained by adolescents who used COCs. Subcutaneous DMPA has also been shown to cause modest weight gain in most women (Westhoff, 2007b). Because women who gain weight in the first 6 months of use are more likely to have long-term progressive weight gain, Le and colleagues (2009) suggest that these women may benefit from early counseling.

There are two types of contraceptive methods that are considered as moderately effective. One type includes barrier methods, which are designed to prevent functional sperm from reaching and fertilizing the ovum. The other category consists of fertility awareness methods. Perhaps more so than with other contraceptive methods, moderately effective methods have the highest success rates when used by couples who are dedicated to their use.

Barrier Methods

These methods include vaginal diaphragms and male and female condoms. As shown in Table 5-2, the reported pregnancy rate for these methods varies from 2 to 6 percent in the first year of use and is highly dependent on correct and consistent use.

Male Condom

Most condoms are made from latex rubber, and various sizes are manufactured to accommodate male anatomy. Less commonly, polyurethane or lamb cecum is used. Condoms provide effective contraception, and their failure rate when used by strongly motivated couples has been as low as 3 or 4 per 100 couple-years of exposure (Vessey, 1982). Generally, and especially in the first year of use, the failure rate is much higher.

The efficacy of condoms is enhanced appreciably with a reservoir tip. Lubricants should be water based because oil-based products destroy latex condoms and diaphragms (Waldron, 1989). Key steps to ensure maximal condom effectiveness include: (1) used with every coitus, (2) placed before penis and vagina contact, (3) withdrawn while penis still erect, (4) its base held during withdrawal, and (5) used with spermicide.

A distinct advantage of condoms is that, when used properly, they provide considerable—not absolute—protection against many STDs. Condoms also help prevent premalignant cervical changes, probably by blocking HPV transmission (Winer, 2006).

For latex-sensitive individuals, alternative condoms are available. Condoms made from lamb intestines—natural skin or lambskin condoms—are effective, but they do not provide protection against STDs. Nonallergenic condoms are made with a synthetic thermoplastic elastomer, such as polyurethane, which is also used in some surgical gloves. These are effective against STDs but have significantly higher breakage and slippage rates compared with those of latex condoms (Gallo, 2006). In a randomized trial of 901 couples, Steiner and associates (2003) documented breakage and slippage with 8.4 percent of polyurethane condoms compared with only 3.2 percent of latex condoms. They also reported that 6-month typical pregnancy probabilities were 9.0 percent with polyurethane condoms but only 5.4 percent with latex ones.

Female Condom—Vaginal Pouch

Manufactured by many companies under different names, female condoms prevent pregnancy and STDs. One brand available in the United States is the FC2 Female Condom—a polyurethane cylindrical sheath with a flexible polyurethane ring at each end (Fig. 5-12). The open ring remains outside the vagina, and the closed internal ring is fitted behind the symphysis and beneath the cervix like a diaphragm (Fig. 5-13). It should not be used with a male condom because together they may slip, tear, or become displaced. In vitro tests show the female condom to be impermeable to HIV, cytomegalovirus, and hepatitis B virus. As shown in Table 5-2, the pregnancy rate is higher than with the male condom.

Diaphragm plus Spermicide

The diaphragm consists of a circular, flexible rubber dome of various diameters supported by a circumferential metal spring (Fig. 5-14). When used in combination with spermicidal jelly or cream, it can be very effective. The spermicide is applied to the cervical surface centrally in the cup and along the rim. The device is then placed in the vagina so that the cervix, vaginal fornices, and anterior vaginal wall are partitioned effectively from the remainder of the vagina and the penis. At the same time, the centrally placed spermicidal agent is held against the cervix by the diaphragm. When appropriately positioned, the rim is lodged deep into the posterior vaginal fornix. Superiorly, the rim lies in close proximity to the inner surface of the symphysis immediately below the urethra (Fig. 5-15). If the diaphragm is too small, it will not remain in place. If too large, it will be uncomfortable when positioned. Because the variables of size and spring flexibility must be specified, the diaphragm is available only by prescription. Because of the requirement for proper placement, the diaphragm may not be an effective choice for women with significant pelvic organ prolapse. The malpositioned uterus can cause unstable diaphragm positioning that results in expulsion.

For use, the diaphragm and spermicidal agent can be inserted well before intercourse, but if more than 2 hours elapse, additional spermicide is placed in the upper vagina for maximum protection. Spermicide is similarly placed before each episode of coitus. The diaphragm is not removed for at least 6 hours after intercourse. Because toxic shock syndrome has been described following its use, the diaphragm is not left in place for longer than 24 hours.

Proper diaphragm use requires a high level of motivation. Vessey and coworkers (1982) reported a pregnancy rate of only 1.9 to 2.4 per 100 woman-years for compliant users. In a small study, Bounds and colleagues (1995) reported a much higher failure rate of 12.3 per 100 woman years. The unintended pregnancy rate is lower in women older than 35 years compared with younger women.

Cervical Cap

This reusable, washable, silicone barrier device surrounds the cervix to block sperm passage and is combined with a spermicide. Marketed in the United States, FemCap is currently available in 22-, 26-, and 30-mm diameters to accommodate differing cervical sizes. It may be inserted any time prior to intercourse and must be left in place for at least 8 hours thereafter. Spermicide dosing and redosing mirrors that with a diaphragm. Other caps formerly manufactured in the United States that might still be in use include Prentif, Vimule, Dumas, and Lea Shield.

Fertility Awareness-based Methods

This form of contraception is defined by the WHO (2007) as a method that involves identification of the fertile days of the menstrual cycle (Fig. 5-16). The couple may then avoid intercourse or use a barrier method during those days. The comparative efficacy of fertility-based awareness methods remains unknown (Grimes, 2004). Clearly, proper instruction is critical, and complex charting is involved. These charts, as well as detailed advice, are available from the National Fertility Awareness and Natural Family Planning Service for the United Kingdom at: http://www.fertilityuk.org and from the Natural Family Site at: http://www.bygpub.com/natural.

Spermicides

These contraceptives are marketed variously as creams, jellies, suppositories, aerosol foams, and film (Fig. 5-17). They are used widely in the United States, and most are available without a prescription. Probable users include women who find other methods unacceptable. They are useful especially for women who need temporary protection, for example, during the first week after starting CHC or while nursing.

Spermicidal agents provide a physical barrier to sperm penetration and a chemical spermicidal action. The active ingredient is nonoxynol-9 or octoxynol-9. Importantly, spermicides must be deposited high in the vagina in contact with the cervix shortly before intercourse. Their duration of maximal effectiveness is usually no more than 1 hour. Thereafter, they must be reinserted before repeat intercourse. Douching, if practiced, is avoided for at least 6 hours after intercourse.

High pregnancy rates are primarily attributable to inconsistent use rather than to method failure. Even if inserted regularly and correctly, however, foam preparations are reported to have a failure rate of 5 to 12 pregnancies per 100 woman-years of use (Trussell, 1990). If pregnancy does occur with use, spermicides are not teratogenic (Briggs, 2015).

Spermicides that primarily contain nonoxynol-9 do not provide protection against STDs. In randomized trials, Roddy and colleagues (1998) compared nonoxynol-9 with and without condom use and found no additional protective effects against chlamydial or HIV infection or gonorrhea. Long-term use of nonoxynol-9 was reported to have minimal effects on vaginal flora (Schreiber, 2006).

There is currently much interest in combined spermicide-microbicide agents. These have the advantage of protecting against STDs, including HIV (Weber, 2005). Those in the surfactant class have dual action—they destroy the sperm membrane and they also disrupt the outer envelopes or membranes of viral and bacterial pathogens. Second-generation microbicides also fortify natural defenses. Third-generation microbicides work as topical antiretroviral agents.

Contraceptive Sponge

The contraceptive sponge Today was reintroduced into the United States in 2005. Sold over the counter, it consists of a nonoxynol-9–impregnated polyurethane disc that can be inserted for up to 24 hours prior to intercourse (Fig. 5-18). The disc is moistened and placed directly against the cervix. While in place, the sponge provides contraception regardless of the number of coital episodes. For efficacy, it remains in place for 6 hours after intercourse, and to lower irritation and infection risks, it remains no longer than 30 hours (Mayer Laboratories, 2009). Although perhaps more convenient, the sponge is less effective than the diaphragm or condom.

First popularized by the “morning-after pill” in the 1970s, emergency contraception (EC) is now widely available in other forms. These methods are appropriate for women presenting for contraceptive care following consensual but unprotected sexual intercourse or following sexual assault. There are several methods that, if used correctly, will substantially decrease the likelihood of an unwanted pregnancy in these women. According to the American College of Obstetricians and Gynecologists (2015), methods currently available include sex steroid-containing compounds, antiprogesterone compounds, and the copper-containing IUD (Table 5-11). Importantly, because duration of use is short, women with conditions that might normally contraindicate hormonal forms may be given these for EC.

Information regarding EC is made available to health care providers or patients by several 24-hour sources:

• American Congress of Obstetricians and Gynecologists: www.acog.org

• Emergency Contraception Hotline and website: 1–888-NOT-2-LATE (888–668–2528) and www.not-2-late.com

• Reproductive Health Technologies Project: www.rhtp.org/contraception/emergency/

• Pastillas Anticonceptivas de Emergencia: www.en3dias.org.mx.

Hormone-based Options

Mechanisms of Action

Hormonal contraceptives have different mechanisms of action depending on which day of the menstrual cycle intercourse occurs and which day the tablets are given (Croxatto, 2003). One major mode is inhibition or delay of ovulation (Marions, 2004). Other suggested mechanisms include endometrial changes that prevent implantation, interference with sperm transport or penetration, and impaired corpus luteum function. Despite these effects, every method used for postcoital contraception will have failures. Pregnancies that develop despite emergency hormonal contraception appear unaffected by this prophylaxis. Moreover, emergency hormonal contraception is not a form of medical abortion. Rather, this method prevents ovulation or implantation. It cannot disrupt a zygote that has implanted.

Except perhaps the copper IUD, other EC methods generally will not prevent pregnancy resulting from subsequent episodes of intercourse during the same cycle. Accordingly, use of a barrier technique is recommended until the next menses. When menstruation is delayed 3 weeks past its expected onset, the likelihood of pregnancy is increased and appropriate testing is instituted.

Estrogen-Progestin Combinations

Also known as the Yuzpe method, these COC-containing regimens shown in Table 5-11 have been approved by the FDA for use as EC (Yuzpe, 1974). Although more effective the sooner they are taken after unprotected intercourse, pills should be taken within 72 hours of intercourse, but may be given up to 120 hours. Initial dosing is followed 12 hours later by a second dose.

Efficacy is defined by the number of pregnancies observed after treatment divided by the estimated number that would have occurred with no treatment. This prevented fraction ranges widely between reports and averages approximately 75 percent with COC regimens (American College of Obstetricians and Gynecologists, 2015).

Nausea and vomiting are common with COC regimens because of their high estrogen doses (Trussell, 1998a). An oral antiemetic taken at least 1 hour before each dose may reduce these bothersome symptoms. In randomized trials, a 1-hour pretreatment dose of either 50-mg meclizine or 10-mg metoclopramide was effective (Ragan, 2003; Raymond, 2000). If vomiting occurs within 2 hours of a dose, a replacement dose is given.

Progestin-only Regimens

A progestin-only method of EC is marketed as Plan B and Plan B One-Step. Plan B consists of two tablets, each containing 0.75 mg of levonorgestrel. The first dose is taken within 72 hours of unprotected coitus but may be given as late as 120 hours, and the second dose is taken 12 hours later (see Table 5-11). Ngai and associates (2005) also showed that a 24-hour interval between dosing is effective. Plan B One-Step is a single, 1.5-mg levonorgestrel dose, which is taken ideally with 72 hours or up to 120 hours following intercourse.

Most studies, including a multicenter WHO trial, indicate that the progestin-only regimens are more effective than COC regimens to prevent pregnancy (von Hertzen, 2002). The American College of Obstetricians and Gynecologists (2015) cites an approximate 50-percent decreased pregnancy rate with levonorgestrel compared with COCs. Finally, Ellertson and colleagues (2003) reported a 55-percent pregnancy prevention rate even if Plan B was taken as late as 4 to 5 days after unprotected intercourse.

Antiprogestins and Selective Progestin-receptor Modulators

These agents, described in Chapter 9, have contraceptive activity because they prevent progesterone-mediated endometrial preparation for implantation. There are several mechanisms by which antiprogesterone compounds achieve this. One mechanism is by progesterone-receptor modulation, and two compounds are available.

First, mifepristone (RU 486)—Mifeprex—is a progesterone antagonist. It either delays ovulation or impairs secretory endometrium development. Cheng and colleagues (2012) in their Cochrane review noted that mifepristone in single doses of 25 or 50 mg was superior to other hormonal EC regimens. Mifepristone also had few side effects. In the United States, mifepristone is not used for EC because of its high cost and because it is not manufactured or marketed in an appropriate dose for EC.

A second drug, a selective progesterone-receptor modulator (SPRM), was FDA approved in 2010 for postcoital contraception. Ulipristal acetate—Ella—is taken as a single 30-mg tablet up to 120 hours after unprotected intercourse (Brache, 2010). Side effects include nausea and delay of subsequent menses.

Copper-containing Intrauterine Devices

Insertion of a copper-containing IUD is an effective postcoital contraceptive method. Fasoli and coworkers (1989) summarized nine studies that included results from 879 women who chose this as a sole method of postcoital contraception. The only pregnancy reported aborted spontaneously. Trussell and Stewart (1998b) reported that when the IUD was inserted up to 5 days after unprotected coitus, the failure rate was 1 percent. A secondary advantage is that this method also puts in place an effective 10-year method of contraception.