CHAPTER 38: Contraception

Nearly half of all pregnancies each year in the United States are unintended (Finer, 2016). These may follow contraceptive method failure or stem from lack of contraceptive use. In 2011 to 2013, 7 percent of sexually active fertile women in the United States not pursuing pregnancy did not use any birth control method (Daniels, 2015).

For those seeking contraception, effective contraceptive methods are available and variably selected (Table 38-1). With these methods, estimated failure rates of perfect and typical use during the first year differ widely. To reflect these failure rates, the World Health Organization (WHO) has grouped methods into efficacy tiers (see Table 38-1). Implants and intrauterine devices are found in the top tier. They effectively lower unintended pregnancy rates and are considered long-acting reversible contraception (LARC). The American College of Obstetricians and Gynecologists (2017c) recognizes these tiers, recommends counseling on all options, and encourages highly effective LARC for appropriate candidates.

No contraceptive method is completely without side effects, but contraception usually poses less risk than pregnancy. However, some disorders or medications can raise the risks from certain contraceptives. The World Health Organization (2015) has provided and updated evidence-based guidelines, termed Medical Eligibility Criteria, for the use of all highly effective reversible contraceptive methods by women with various health conditions. Individual countries have subsequently modified these guidelines. The United States Medical Eligibility Criteria (US MEC) was updated in 2016 by the Centers for Disease Control and Prevention (Curtis, 2016b). These documents are available at: http://www.cdc.gov/reproductivehealth/UnintendedPregnancy/Contraception_Guidance.htm.

In the United States MEC, reversible contraceptive methods are organized into six groups by their similarity: combination hormonal contraceptives (CHCs), progestin-only pills (POPs), depot medroxyprogesterone acetate (DMPA), implants, levonorgestrel-releasing intrauterine system (LNG-IUS), and copper intrauterine devices (Cu-IUDs). For a given health condition, each method is categorized 1 through 4 (Table 38-2). The score describes the 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, and (4) method poses an unacceptably high health risk.

Alternatively, depending on the underlying disorder or patient desire, male or female sterilization may be a preferred or recommended permanent contraceptive method (American College of Obstetricians and Gynecologists, 2017b). These options are fully discussed in Chapter 39.

Globally, 14 percent of reproductive-aged women use intrauterine contraception, and in the United States, 10 percent of contracepting women use this method (Buhling, 2014; Daniels, 2015). The five intrauterine devices (IUDs) currently approved for use in the United States are chemically active and continually elute either copper or levonorgestrel. These all have a T-shaped frame of polyethylene that is compounded with barium to render them radiopaque. Of devices, Mirena and Liletta both measure 32 × 32 mm and contain a 52-mg levonorgestrel-releasing cylinder reservoir in the stem of the T (Fig. 38-1). Two trailing strings that are tan (Mirena) or blue (Liletta) are attached to the distal stem to aid eventual device removal. Skyla—known as Jaydess in some countries—contains 13.5 mg of levonorgestrel. It has smaller dimensions—28 × 30 mm—and was sized to more appropriately fit a nulliparous uterus (Gemzell-Danielsson, 2012). Kyleena has the same dimensions but contains 19.5 mg of the same progestin. Two trailing strings off Skyla are tan, and those off Kyleena are blue. Skyla and Kyleena can be differentiated from Mirena and Liletta visually and sonographically by a silver ring near the junction of their stem and arms. Mirena and Kyleena are currently approved for 5 years of use following insertion, whereas Skyla and Liletta are approved for 3 years.

The third device, the T380A IUD named ParaGard, is wound with copper, and two strings extend from the stem base. Originally blue, the strings are now white. It is currently approved for 10 years of use following insertion (Teva Women’s Health, 2014).

In addition to these five currently marketed IUDs, women may retain discontinued brands. A Lippes Loop has two “S” shapes stacked one on the other. The Dalkon Shield has a crab form, whereas a Copper 7 mirrors that number. Progestasert is an early T-shaped progestin-releasing IUD. Last, various metal-eluting ring devices are common in Asia.

Contraceptive Action

All these IUDs are effective. Failure rates are well below 1 percent and similar overall to those of tubal sterilization (Thonneau, 2008; Trussell, 2011b). Their mechanisms have not been precisely defined, but prevention of fertilization is now favored.

With the LNG-IUS, long-term progestin release leads to endometrial atrophy, which hinders normal implantation (Silverberg, 1986). Moreover, progestins create scant viscous cervical mucus that obstructs sperm motility (Apter, 2014; Moraes, 2016). Within the uterus, an intense local endometrial inflammatory response is induced, especially by copper-containing devices. Cellular and humoral components of this inflammation are expressed in endometrial tissue and in fluid filling the uterine cavity and fallopian tubes. These lead to decreased sperm and egg viability (Ortiz, 2007). Also, in the unlikely event that fertilization does occur, the same inflammatory actions are directed against the blastocyst. Also, with the Cu-IUD specifically, copper levels rise in the cervical mucus of users and act to decrease sperm motility and viability (Jecht, 1973). The above effects are considered primary for contraception because ovulation inhibition is inconsistent with the LNG-IUS and lacking with the Cu-IUD (Nilsson, 1984).

Method-Specific Adverse Effects

Ectopic Pregnancy

Contraindications to IUD use are few and shown in Table 38-2. In the past, IUDs were perceived to increase the risk of ectopic pregnancy, but this has since been clarified. Specifically, IUDs provide effective contraception and lower the absolute number of ectopic pregnancies by half compared with the rate in noncontracepting women (World Health Organization, 1985, 1987). But, the IUD mechanisms of action are more effective in preventing intrauterine implantation. Thus, if an IUD fails, a higher proportion of pregnancies are likely to be ectopic (Backman, 2004; Furlong, 2002).

Lost Device

Expulsion of an IUD from the uterus is most common during the first month. Thus, women are examined approximately 1 month following IUD insertion, usually after menses, to identify the tails trailing from the cervix. Following this, a woman is instructed to palpate the strings each month after menses. Regardless of IUD type, the cumulative 3-year expulsion rate approximates 10 percent (Madden, 2014; Simonatto, 2016). This rate is higher in those ≤25 years (Jatlaoui, 2017).

If the tail of an IUD cannot be visualized, the device may have been expelled, may have perforated the uterus, or may be malpositioned. Alternatively, the device may be normally positioned with its tails folded within the endocervical canal or uterine cavity. To investigate, after excluding pregnancy, a cytological brush can be twirled within the endocervical canal to entangle the strings and bring them gently into the vagina. If unsuccessful, the uterine cavity is probed gently with a Randall stone clamp or with a specialized rod with a terminal hook to snare the strings or device.

One should not assume that a device has been expelled unless it was seen. Thus, if tails are not visible and the device is not felt by gentle probing of the uterine cavity, transvaginal sonography can be used to ascertain if the device lies within the uterus. Although traditional sonography will document IUD position adequately in most cases, three-dimensional sonography offers improved visualization, especially with the LNG-IUS (Moschos, 2011). If sonography is inconclusive or if no device is seen, then a plain radiograph of the abdominopelvis is taken. Computed tomography (CT) scanning or, less commonly, magnetic resonance (MR) imaging is an alternative (Boortz, 2012). It is safe to perform MR imaging at 1.5 and 3 Tesla (T) with an IUD in place (Ciet, 2015).

Perforation

During uterine sounding or IUD insertion, the uterus may be perforated, which is identified by the tool traveling farther than the expected uterine length based on initial bimanual examination. Rates approximate 1 per 1000 insertions, and risks include puerperal insertion, lactation, provider inexperience, and extremes of uterine flexion (Harrison-Woolrych, 2003; Heinemann, 2015). Although devices may migrate spontaneously into and through the uterine wall, most perforations occur, or at least begin, at the time of insertion (Ferguson, 2016).

With acute perforation, the fundus is the more common site, and bleeding is typically minimal due to myometrial contraction around the puncture hole. If no brisk or persistent bleeding is noted from the os following instrument or device removal, then patient observation alone is reasonable. Rarely, acute lateral perforations may lacerate the uterine artery, and subsequent brisk bleeding may prompt laparoscopy or laparotomy for control. Following any perforation, although this is not firmly evidence-based, a single dose of broad-spectrum antibiotic may mitigate infection.

With chronic perforation, a device can penetrate the muscular uterine wall to varying degrees. A patient may be asymptomatic but abdominal pain, uterine bleeding, or missing strings can be clues (Kaislasuo, 2013). Those with a predominantly intrauterine location are usually managed by hysteroscopic IUD removal. In contrast, devices that have nearly or completely perforated through the uterine wall are more easily removed laparoscopically. Notably, an extrauterine Cu-IUD frequently induces an intense local inflammatory reaction and adhesions (Kho, 2014). Laparotomy may be necessary, and bowel preparation is considered. Sigmoid and bladder perforations and small-bowel obstruction have been reported remote from insertion (Sano, 2017; Xu, 2015; Zeino, 2011).

Menstrual Changes

Dysmenorrhea and bleeding irregularities can complicate IUD use (Aoun, 2014; Grunloh, 2013). These can be treated with some degree of success by nonsteroidal antiinflammatory drugs (NSAIDs) or tranexamic acid, which is an antifibrinolytic (Godfrey, 2013; Madden, 2012; Sørdal, 2013). Of the two IUDs, heavy bleeding more often complicates Cu-IUD use and may cause iron-deficiency anemia, for which oral iron salts are given. With the LNG-IUS, irregular spotting for up to 6 months after placement often gives way to progressive amenorrhea, which is reported by 30 percent of women after 2 years and by 60 percent after 12 years (Ronnerdag, 1999). This is frequently associated with improved dysmenorrhea.

Infection

The risk of upper genital tract device-related infection is greatest during the first months following IUD insertion (Farley, 1992; Turok, 2016). Pathogens include Neisseria gonorrhoeae, Chlamydia trachomatis, and vaginal flora. Women at risk for sexually transmitted diseases (STDs) should be screened either before or at the time of IUD insertion (Centers for Disease Control and Prevention, 2015; Sufrin, 2012). That said, device insertion need not be delayed while awaiting sexually transmitted disease or Pap test results in asymptomatic women (Birgisson, 2015). If these bacteria are subsequently found and the patient is without symptoms, then the IUD may remain and treatment prescribed as detailed in Chapter 65 (Chlamydial Infections). Importantly, routine antimicrobial prophylaxis before insertion is not recommended (Grimes, 2012; Walsh, 1998). And the American Heart Association does not recommend bacterial endocarditis prophylaxis with insertion (Nishimura, 2014).

After the first month, infection risk is not increased in IUD users who would otherwise be at low risk of sexually transmitted infections. Correspondingly, IUDs appear to cause little, if any, increase in infertility rates in these low-risk patients (Hubacher, 2001). The American College of Obstetricians and Gynecologists (2015c, 2016a) recommends that women at low risk for STDs, including adolescents, be considered good candidates for IUDs. The IUD is also safe and effective in women infected with human immunodeficiency virus (HIV) and may be used in others who are immunosuppressed (Centers for Disease Control and Prevention, 2015; Tepper, 2016a).

If infection does develop, it may take several forms and typically requires broad-spectrum antimicrobials. Pelvic inflammatory disease (PID) without abscess is treated with antibiotics on an outpatient or inpatient basis, depending on infection severity. There are theoretical concerns that a coexistent IUD may worsen the infection or delay resolution. A provider may choose to remove an IUD in this setting, although some evidence supports allowing a device to remain during treatment in those hospitalized with mild or moderate PID (Centers for Disease Control and Prevention, 2015; Tepper, 2013). If infection fails to improve during 48 to 72 hours of treatment, the device is removed. Tuboovarian abscess can complicate PID and is treated aggressively with intravenous broad-spectrum antibiotics and IUD removal. Last, septic abortion mandates immediate uterine evacuation and antibiotics.

Actinomyces israelii is a gram-positive, slow-growing, anaerobic indigenous vaginal bacterium that rarely causes suppurative infection. Some have found it more frequently in the vaginal flora or on the Pap smears of IUD users (Curtis, 1981; Kim, 2014). Current recommendations advise that an asymptomatic woman may retain her IUD and does not require antibiotic treatment (American College of Obstetricians and Gynecologists, 2017c; Lippes, 1999; Westhoff, 2007a). However, if signs or symptoms of infection develop in a woman who harbors Actinomyces species, then the device is removed and antibiotics are given. Early findings with infection include fever, weight loss, abdominal pain, and abnormal uterine bleeding or discharge. Actinomyces species are sensitive to antibiotics with gram-positive coverage, notably the penicillins.

Pregnancy with an IUD

For women who conceive while using an IUD, ectopic pregnancy should be excluded. With intrauterine pregnancy, if the tail is seen, it should be grasped and the IUD removed by gentle outward traction. This action reduces complications such as abortion, chorioamnionitis, and preterm birth (Fulkerson Schaeffer, 2017; Kim, 2010). Specifically, in one cohort, a 54-percent abortion rate and 17-percent preterm delivery rate was noted if the device remained in situ. More favorably, rates of 25 percent and 4 percent, respectively, resulted from prompt Cu-IUD removal (Tatum, 1976). Few data guide management with the LNG-IUS, and most practice extrapolates from copper devices.

If the tail is not visible, attempts to locate and remove the device may result in abortion. Although not our practice, some case reports and small series describe sonography or hysteroscopy to assist in difficult device removals (Pérez-Medina, 2014; Schiesser, 2004). After fetal viability is reached, it is unclear whether it is better to remove an IUD whose strings are visible and accessible or to leave it in place. Fetal malformation rates are not greater with a device left in situ (Tatum, 1976; Vessey, 1979).

Second-trimester miscarriage with an IUD in place is more likely to be infected (Vessey, 1974). Sepsis may be fulminant and fatal. Pregnant women with a device in utero who demonstrate any evidence of pelvic infection are treated with broad-spectrum antibiotics and prompt uterine evacuation. Because of these risks, a woman should be given the option of early pregnancy termination if the device cannot be removed early in pregnancy. Last, in women who give birth with a device in place, appropriate steps should be taken at delivery to identify and remove the IUD.

IUD Insertion

Timing

To reduce expulsion rates and perforation risks, IUD insertion has traditionally followed complete uterine involution—at least 6 weeks after delivery. Women delivered at Parkland Hospital are seen 3 weeks postpartum, and IUDs are inserted 6 weeks postpartum or sooner if involution is complete.

Alternatively, immediately after miscarriage, surgical abortion, or delivery, an IUD may be inserted in the absence of overt infection (Lopez, 2015a; Okusanya, 2014). Also, “immediate” insertion 1 week after mifepristone and completed medical abortion has been described (Sääv, 2012; Shimoni, 2011). The risk of IUD expulsion is slightly higher if it is placed immediately following any of these recent pregnancies (Whitaker, 2017). However, in studies, the number of women in immediate-placement groups who ultimately receive and retain an IUD is greater than in groups scheduled for traditionally timed placement, some of whose members do not return for insertion (Bednarek, 2011; Chen, 2010).

With immediate insertion, techniques depend on uterine size. After first-trimester evacuation, the IUD can be placed using the manufacturer’s standard instructions. If the uterine cavity is larger, the IUD can be placed using ring forceps with sonographic guidance (Drey, 2009; Fox, 2011).

Immediately following vaginal or cesarean delivery, an IUD can be placed by a hand, by its inserter tube, or by ring forceps (Levi, 2015; Xu, 1996). With any of these methods, the arms of the IUD need not be folded into the inserter tube prior to insertion. During cesarean delivery placement, the hand or inserter travels through the unsutured open hysterotomy to deposit the device at the fundus. A second hand cupping the outer fundus can provide back pressure and stabilize the uterus during insertion. Strings are then gently directed toward the cervix. For instrumented insertion following vaginal delivery, the clinician resterilizes the vulva and changes gloves after placental delivery but before perineal repairs. The anterior lip of the floppy cervix is held with ring forceps. A second ring forceps grasp the IUD stem and guides it through the uterine cavity to the fundus. For manual insertion following vaginal delivery, the provider secures the IUD between the index and middle fingers to deposit the device. In either case, back pressure against the fundus by an abdominal hand can guide positioning (Stuart, 2017; The ACQUIRE Project, 2008).

For placement not related to pregnancy, insertion near the end of normal menstruation, when the cervix is usually softer and somewhat more dilated, may be easier and also helps exclude early pregnancy. But, insertion is not limited to this time. For the woman who is sure she is not pregnant and does not want to be pregnant, insertion is done at any time.

Technique

Before insertion, contraindications are sought. Candidates are counseled, and written consent obtained. An oral NSAID, with or without codeine, can be used to allay cramps after insertion (Ngo, 2015). But NSAIDs, misoprostol, or even paracervical blockade do not consistently decrease discomfort experienced during device placement (Bednarek, 2015; Hubacher, 2006; Mody, 2012; Pergialiotis, 2014). Of topical lidocaine products, 2-percent gel is ineffective, but a newer gel and a spray both show promise (Aksoy, 2016; Lopez, 2015b; Tornblom-Paulander, 2015). Bimanual pelvic examination delineates uterine position and size. Abnormalities are evaluated, as they may contraindicate insertion. Mucopurulent cervicitis or significant vaginitis is appropriately treated and resolved before IUD insertion.

The cervical surface is cleansed with an antiseptic solution, and sterile instruments and a sterile IUD are used. A tenaculum is placed on the cervical lip, and the canal and uterine cavity are straightened by applying gentle outward traction. The cavity is then probed by a uterine sound to identify its direction and depth. Specific steps of ParaGard and Mirena insertion are shown in Figures 38-2 and 38-3 and outlined in their respective package inserts.

Following insertion, only the threads should be visible trailing from the cervix. These are trimmed to allow 3 to 4 cm to protrude into the vagina, and their length is recorded. If improper device positioning is suspected, placement should be confirmed, using sonography if necessary. If the IUD is not positioned completely within the uterus, it is removed and replaced with a new device. An expelled or partially expelled device should not be reinserted.

Etonogestrel Implant

Contraception can be provided by thin, pliable progestin-containing cylinders that are implanted subdermally and release hormone over many years. One of these, Nexplanon is a single-rod implant with 68 mg of etonogestrel covered by an ethylene vinyl acetate copolymer cover. The implant is placed subdermally on the medial surface of the upper arm 8 to 10 cm from the elbow in the biceps groove and is aligned with the long axis of the arm. It may be used as contraception for 3 years, removed, and then replaced at the same site or in the opposite arm (Merck, 2016a).

Nexplanon is radiopaque, and its inserter device is designed to assist with subdermal positioning and avert deeper placement. This device replaced Implanon, which is not radiopaque. A misplaced Implanon may be identified with sonography using a 10- to 15-MHz linear array transducer (Shulman, 2006). In some cases, MR imaging may be required if supplemental information is needed despite sonography (Correia, 2012). Both implants are similarly shaped and pharmacologically identical. They are highly effective, and the mechanism of action for progestin-only products is described later (Progestin-Only Contraceptives) (Croxatto, 1998; Mommers, 2012). Implanon is safe and still approved by the Food and Drug Administration (FDA), but it is no longer distributed by the manufacturer.

Levonorgestrel Implants

The first progestin implants contained levonorgestrel (LNG), and systems are still available outside the United States. Jadelle, originally named Norplant-2, provides LNG and contraception for 5 years through two subdermally implanted Silastic rods. After this time, rods may be removed and if desired, new rods inserted at the same site (Bayer Group, 2015). Jadelle is approved by the FDA, however, it is not marketed or distributed in the United States. Sino-implant II is a two-rod system with the same amount (150 mg) of LNG and same mechanism of action as Jadelle but provides 4 years of contraception. Sino-implant II is manufactured in China and approved for use by 20 countries in Asia and Africa (FHI 360, 2012).

Like the etonogestrel implant, these systems are placed subdermally on the inner arm approximately 8 cm from the elbow and have similar removal steps. Implants vary regarding their insertion technique, and manufacturer instructions should be consulted. Both implant systems are highly effective (Sivin, 1998; Steiner, 2010).

The forerunner of these implants was the Norplant System, which provided LNG in six Silastic rods implanted subdermally. The manufacturer stopped distributing the system in 2002.

Few data compare the LNG and etonogestrel implants. In one, efficacy and discontinuation rates by 2.5 years of use were similar (Bahamondes, 2015).

Method-Specific Adverse Effects

Unscheduled bleeding is common with progestin-only methods and described in Progestin-Only Contraceptives. Device-specific adverse effects derive mainly from malpositioning. First, branches of the medial antebrachial cutaneous nerve can be injured during implant or needle insertion that is too deep or during exploration for a lost implant. Clinically, numbness and paresthesia over the anteromedial aspect of the forearm are noted (Wechselberger, 2006). Second, nonpalpable devices are not uncommon and require radiological imaging for localization. As an adjunct, one group adopted the hook-wire tagging method used in breast tumor surgery to allow deep-lying implants to be marked prior to extraction (Nouri, 2013). If imaging fails to locate an implant, etonogestrel blood level determination can help verify that the implant is indeed in situ. This special assay must be coordinated with the manufacturer (1–877–467–5266).

Implant Insertion

Timing

For those not currently using hormonal contraception, the etonogestrel implant is ideally inserted within 5 days of menses onset. With LNG-releasing implants, contraception is established within 24 hours if inserted within the first 7 days of the menstrual cycle (Sivin, 1997; Steiner, 2010). For transitioning methods, an implant is placed on the day of the first placebo combination oral contraceptive (COC) pill; on the day that the next depot-medroxyprogesterone injection would be due; or within 24 hours of taking the last POP (Merck, 2016a). In women certain that they are not pregnant, insertion at other times of the cycle is followed by alternative contraception for 7 days. Related to pregnancy, an implant may be inserted before discharge following delivery or abortion (Sothornwit, 2017).

Nexplanon Insertion Technique

With the patient lying down, her nondominant arm, forearm, and hand are outstretched on the bed with the inner aspects of each exposed upward, and the elbow is flexed. The insertion site is marked with a sterile pen 8 to 10 cm proximal to the medial condyle of the humerus. A second mark is placed 4 cm proximally and delineates the final path of the implant. The Nexplanon is inserted using sterile technique. The area is cleansed aseptically, and a 1-percent lidocaine anesthetic track is injected beneath the skin along the planned insertion path. The implant is then placed as shown in Figure 38-4. After placement, both patient and provider should palpate and identify both ends of the 4-cm implant. To minimize bruising at the site, a pressure bandage is created around the arm and is removed the following day.

With implant extraction, the removal site is first cleansed with antiseptic. The proximal end of the implant is depressed with a finger to allow the distal end to bulge up toward the skin. After anesthetizing the skin over this bulge, the skin is incised 2 mm toward the elbow along the long axis of the arm. The proximal butt of the implant is then pushed toward this incision. Once visible, the distal end of the implant is grasped with a hemostat and removed. If present, superficial adhesions surrounding an implant may be dissected away with hemostat tips placed into the incision.

Actions and Side Effects

Progestin-only contraceptives include the implants just described, injectables, and pills. As their primary contraceptive action, these progestins suppress luteinizing hormone (LH) and in turn block ovulation. As other effects, cervical mucus is thickened to retard sperm passage, and atrophy renders the endometrium unfavorable for implantation. Fertility is restored rapidly following cessation of progestin-only contraception. An exception is DMPA, as described in Barrier Methods (Mansour, 2011).

For all progestin-only methods, irregular or heavy uterine bleeding is a distinct disadvantage. It is the most frequently reported adverse event leading to method discontinuation. Often, counseling and reassurance is sufficient. Troublesome bleeding may be improved by one to two cycles of combination oral contraceptives, by a 1- to 3-week course of estrogen alone, or by a short course of NSAIDs combined with the established method (Abdel-Aleem, 2013). Fortunately, with prolonged use, progestins induce endometrial atrophy, which leads to sustained amenorrhea. For the well-counseled patient, this is often an advantage.

Most progestin-only contraceptive methods do not significantly affect lipid metabolism, glucose levels, hemostatic factors, liver function, thyroid function, or blood pressure (Dorflinger, 2002). However, the increased low-density lipoprotein (LDL) cholesterol and decreased high-density lipoprotein (HDL) cholesterol levels seen with DMPA may be less desirable for women with cardiac or vascular disease risks (Kongsayreepong, 1993).

Progestin-only methods do not impair milk production and are an excellent choice for lactating women. There are no increased risks of genital tract, liver, or breast neoplasia (Samson, 2016; Wilailak, 2012; World Health Organization, 1991a,b, 1992). Weight gain and bone fracture are not prominent side effects of this contraceptive group, except for depot progesterone, discussed in Barrier Methods (Lopez, 2012a, 2013a). Functional ovarian cysts develop with a greater frequency in women using progestin-only agents, although they do not usually necessitate intervention (European Society of Human Reproduction and Embryology, 2001; Hidalgo, 2006; Nahum, 2015). Last, an association between depression and DMPA or POPs is unclear (Civic, 2000; Pagano, 2016; Svendal, 2012; Westhoff, 1995). Women with depression may be prescribed these methods, but surveillance following initiation is reasonable.

Progestin Contraindications

These methods are ideal for most women, but contraindications and cautions are associated with a few conditions listed in Table 38-2. Current breast cancer and pregnancy are the only two absolute contraindications. In a few instances, manufacturer restrictions differ from the US MEC. First, manufacturer prescribing information lists thrombosis or thromboembolic disorders as contraindications (Merck, 2016a; Pfizer, 2015a,b). However, for individuals with these disorders, US MEC considers progestin-containing methods category 2. Moreover, evidence does not link progestin-only methods with thromboembolism, stroke, or cardiovascular disease (Mantha, 2012; Tepper, 2016b; World Health Organization, 1998). Second, for many progestin products, manufacturers note prior ectopic pregnancy as a contraindication. This is secondary to progesterone’s effect of slowing fallopian tube motility and thereby delaying fertilized egg transport to the endometrial cavity. That said, effective contraception lowers pregnancy rates overall. Thus, for those with prior ectopic pregnancy, US MEC considers progestin injectables and implants category 1, and progestin-only pills are category 2.

These currently are available in forms that contain both estrogen and progestin or contain only progestin. Progestin-only injectables and pills are considered very effective, yet second-tier agents, due to the need for increased patient compliance. Similarly, products containing both estrogen and progestin, often termed combination hormonal contraception (CHC), are considered in this tier. These may be supplied as pills, transvaginal rings, or transdermal patches.

Combination Hormonal Contraceptives Mechanism of Action

Actions of combination hormonal contraceptives are multiple, but the most important effect is suppression of hypothalamic gonadotropin-releasing factors. This in turn blocks pituitary secretion of follicle-stimulating hormone (FSH) and luteinizing hormone (LH) and thereby inhibits ovulation. The progestin component of CHCs provides ovulation prevention by suppressing LH; it thickens cervical mucus and thereby retards sperm passage; and it renders the endometrium unfavorable for implantation. Estrogen blocks ovulation by suppressing FSH release. To promote cycle control, estrogen stabilizes the endometrium, which prevents intermenstrual bleeding—also known as breakthrough bleeding. The net effect is an extremely effective yet highly reversible method (Mansour, 2011).

Combination Oral Contraceptive Pills

Composition

These pills are the most frequently used reversible birth control method in the United States. In a 2006 to 2010 survey, 16 percent of contracepting women in the United States were using these (Daniels, 2015). COCs are marketed in a wide variety of estrogen and progestin combinations. Most are available as generics, and the FDA (2016) confirms the bioequivalence of COC generics. The American College of Obstetricians and Gynecologists (2015a) supports the use of either branded or generic preparations.

Pharmacologically, ethinyl estradiol is the most common estrogen present in COC formulations in the United States. Less frequently, mestranol or estradiol valerate is used. Unwanted effects most often attributed to the estrogen component include breast tenderness, weight gain, nausea, and headache.

COCs also contain one of several progestins that are structurally related to progesterone, testosterone, or spironolactone. Thus, these progestins bind variably to progesterone, androgen, glucocorticoid, and mineralocorticoid receptors. These affinities explain many pill-related side effects and are often used to compare one progestin with another.

Most progestins used in COCs are related to testosterone and may impart androgenic side affects such as acne and adverse HDL and LDL levels. To avoid these effects, antiandrogenic progestins have been introduced and include dienogest and nomegestrol acetate. The latter is used in a COC approved outside the United States. Despite these pharmacological differences, the true advantage of one progestin over another is less apparent clinically (Lawrie, 2011; Moreau, 2007).

Another progestin, drospirenone, is structurally similar to spironolactone. The doses in currently marketed COCs have effects similar to 25 mg of this diuretic (Seeger, 2007). Drospirenone displays antiandrogenic activity, provides an antialdosterone action to minimize water retention, and has antimineralocorticoid properties that may, in theory, cause potassium retention and hyperkalemia (Krattenmacher, 2000). Thus, it is avoided in women with renal or adrenal insufficiency or with hepatic dysfunction. Moreover, serum potassium level monitoring is recommended in the first month for patients chronically treated concomitantly with any drug associated with potassium retention. These include NSAIDs, angiotensin-converting enzyme (ACE) inhibitors, angiotensin II antagonists, heparin, aldosterone antagonists, and potassium-sparing diuretics (Bayer HealthCare Pharmaceuticals, 2015).

Since the development of COCs, their estrogen and progestin content has dropped remarkably to minimize adverse effects. Currently, the lowest acceptable dose is limited by the ability to prevent pregnancy and to avoid unacceptable breakthrough bleeding. Thus, the daily estrogen content varies from 10 to 50 μg of ethinyl estradiol, and most contain 35 μg or less.

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.

With COCs termed monophasic pills, the progestin dose remains constant throughout the cycle. In others, the dose frequently is varied, and term biphasic, triphasic, or quadriphasic pill is used depending on the number of dose changes within the cycle. In some formulations, the estrogen dose also varies. In general, phasic pills were developed to reduce the total progestin content per cycle without sacrificing contraceptive efficacy or cycle control. The theoretical advantage of a lower total progesterone dose per cycle, however, has not been borne out clinically (Moreau, 2007). Cycle control also appears to be comparable among mono- through triphasic pills (van Vliet, 2011a,b,c).

Administration

Hormones are taken daily for a specified time (21 to 81 days) and then replaced by placebo for a specified time (4 to 7 days), which is called the “pill-free interval.” During these pill-free days, withdrawal bleeding is expected.

With the trend toward lower estrogen doses to minimize side effects, follicular development and ovulation may occur. To counter this, the active-pill duration in some formulations is extended to 24 days. In comparison, these 24/4 regimens perform similarly to higher-estrogen-dose 21/7 regimens (Anttila, 2011; Marr, 2012).

Alternatively, longer durations of active hormone, designed to minimize the number of withdrawal episodes, have similar efficacy and safety profiles as more traditional administration (Edelman, 2014). These extended-cycle products produce a 13-week cycle, that is, 12 weeks of hormone use, followed by a week for withdrawal menses. The product Amethyst provides continuous active hormone pills for 365 days each year. Such extended or continuous regimens may be especially suited for women with significant menstrual symptoms (Mendoza, 2014).

For general initiation, women ideally begin COCs on the first day of a menstrual cycle. In such cases, a supplementary contraceptive method is unnecessary. With the more traditional “Sunday start,” women begin pills on the first Sunday that follows menses onset, and an additional method is needed for 1 week to prevent conception. If menses begin on a Sunday, then pills are begun that day and no supplemental method is required. Alternatively, with the “quick start” method, COCs are started on any day, commonly the day prescribed, regardless of cycle timing. An additional method is used during the first week (Westhoff, 2002, 2007b). If the woman is unknowingly already pregnant during quick start initiation, COCs are not teratogenic (Lammer, 1986; Rothman, 1978; Savolainen, 1981). However, a missed menses following COC initiation should prompt pregnancy testing. Similar same-day initiation can be implemented with the contraceptive vaginal ring or patch (Murthy, 2005; Schafer, 2006).

For maximum efficiency, pills are best taken at the same time each day. If one dose is missed, the missed pill is taken immediately; the scheduled dose for that day is taken on time; and then daily pills are continued. If two or more doses are missed, the most recent missed pill is taken immediately; the scheduled dose for that day is taken on time; and an effective barrier technique used for 7 days while daily pills are then continued (Curtis, 2016a). If withdrawal bleeding fails to occur during the pill-free interval, a woman should continue her pills but seek attention to exclude pregnancy.

With initiation of COCs, spotting or bleeding is common. It does not reflect contraceptive failure and typically resolves within one to three cycles. If unscheduled bleeding persists, those with bleeding during the first part of a pill pack may benefit from an increase in the estrogen dose, whereas those with bleeding during the second part may improve with a higher progestin dose (Nelson, 2011).

Method-Specific Effects

Altered Drug Efficacy

Some drugs decrease COC effectiveness, and choosing another contraceptive method is preferable. However, if a COC is selected for concurrent use in these instances, a preparation containing a minimum of 30 μg ethinyl estradiol is ideally chosen. Conversely, some COCs interfere with the actions of certain drugs (see Table 38-2).

In obese women, COCs are effective (Lopez, 2016). Some studies point to lowered hormone bioavailability, but overall efficacy remains high (Nakajima, 2016; Westhoff, 2010; Yamazaki, 2015). With the transdermal patch method, however, evidence is more robust that obesity may alter pharmacokinetics and lower efficacy, as discussed in Transdermal Patch.

Metabolic Changes

Combination oral contraceptives alter lipid synthesis and in general raise serum levels of triglycerides and of total cholesterol, HDL, and very-low density lipoprotein (VLDL) cholesterol. Estrogen lowers LDL cholesterol concentrations. Oral contraceptives are not atherogenic, and their effect on lipids is clinically inconsequential for most women (Wallach, 2000). In women with dyslipidemias, limited data suggest that COCs increase the risk for myocardial infarction and minimally so for venous thromboembolism or stroke (Dragoman, 2016). For those with multiple additional risk factors for vascular disease, alternative contraceptive methods are recommended.

With COCs, protein metabolism is affected, and estrogens boost hepatic production of various globulins. First, fibrinogen and many of the clotting factor levels rise in direct proportion to the estrogen dose and may lead to thrombosis. Angiotensinogen production is also augmented by COCs, and its conversion by renin to angiotensin I may be associated with “pill-induced hypertension,” discussed subsequently. Last, COCs elevate sex hormone-binding globulin (SHBG) levels, which in turn lower concentrations of bioavailable testosterone and lessen androgenic side effects.

Regarding carbohydrate metabolism, current low-dose formulations have minimal effects in women who do not have diabetes (Lopez, 2014). And, the risk of developing diabetes is not increased (Kim, 2002). For diabetic women, COCs may be used in nonsmokers with disease duration <20 years and without associated vascular disease, nephropathy, retinopathy, or neuropathy (Curtis, 2016b).

Of other metabolic changes, thyroid-binding globulin and thyroid-stimulating hormone (TSH) levels are elevated, but free plasma thyroxine (FT₄) levels are unchanged (Raps, 2014). Studies have not supported a connection between COCs and weight gain (Gallo, 2014).

Cardiovascular Effects

Despite increased plasma angiotensinogen (renin substrate) levels, women using low-dose COC formulations rarely develop clinically significant hypertension (Chasan-Taber, 1996). However, it is common practice for patients to return 8 to 12 weeks after COC initiation for evaluation of blood pressure and other symptoms.

During initial contraception selection, a history of gestational hypertension does not preclude subsequent COC use. Among women with well-controlled hypertension, COC use is linked to greater risks than nonusers for stroke, acute myocardial infarction, and peripheral arterial disease, and in these women, COCs are considered US MEC category 3 (Curtis, 2016b). Severe forms of hypertension, especially those with end-organ involvement, preclude COC use.

For nonsmoking women younger than 35, the risk of stroke is extremely low (World Health Organization, 1996). COCs are associated with a small increased risk for ischemic stroke (Chan, 2004; Lidegaard, 2012). Rates increase significantly for women who have hypertension, who smoke, or who have migraine headaches with visual aura or other focal neurological changes and use COCs (MacClellan, 2007; Tepper, 2016c). The evidence for stroke risk in migraineurs without aura is less clear (Etminan, 2005; Schürks, 2009). COC initiation may be considered for women with preexisting migraines without aura if they are otherwise healthy, younger, normotensive nonsmokers. For women with prior stroke, COCs should not be considered due to risks for repeat events.

For women with prior myocardial infarction, COCs should not be considered. Also, in women with multiple cardiovascular risk factors, which include smoking, hypertension, older age, and diabetes, the risk for myocardial infarction outweighs the benefits of this method. However, for those without these risks, low-dose oral contraceptives are not associated with an increased risk of myocardial infarction (Margolis, 2007; World Health Organization, 1997).

The risks for deep-vein thrombosis and pulmonary embolism are increased in women who use COCs (Stadel, 1981). These clearly are estrogen-dose related, and rates have substantively declined with lower-dose formulations containing 10 to 35 μg of ethinyl estradiol. The general-population risk of venous thromboembolism (VTE) is 4 to 5 events per 100,000 woman-years. The incidence of VTE with COC use increases three- to fivefold compared with nonusers (Shaw, 2013; van Hylckama Vlieg, 2009). Obesity raises the VTE risk, which is compounded by COCs (Horton, 2016; Suchon, 2016). Accordingly, in an obese woman, COCs are considered a US MEC category 2. VTEs are significantly increased in women older than 35 years who smoke, and COCs are not recommended. Those most at risk for VTE include women with thrombophilias (ESHRE Capri Workshop Group, 2013). Moreover, COC use during the month before a major operative procedure appears to double the risk for postoperative VTE (Robinson, 1991). Thus, the American College of Obstetricians and Gynecologists (2016d) recommends balancing the risks of VTE and the degree of postoperative immobility with the risk of unintended pregnancy during the 4 to 6 weeks required to reverse the thrombogenic effects of COCs before surgery. In the early puerperium, VTE risks are also increased, and COCs are not recommended for women within the first 4 weeks after delivery.

Certain progestins within COC are linked with greater rates of thromboembolism. A slightly higher VTE risk with drospirenone-containing COCs has been shown in two studies. In response, an assessment of benefits and VTE risks in users of these pills has been emphasized (Food and Drug Administration, 2012; Jick, 2011; Parkin, 2011). Desogestrel and gestodene are also implicated and carry similarly elevated risks (Stegeman, 2013; Vinogradova, 2015).

Neoplasia

Most studies indicate that COCs overall are not associated with an increased risk of cancer (Cibula, 2010). In fact, a protective effect against ovarian and endometrial cancer has been shown (Collaborative Group on Epidemiological Studies of Ovarian Cancer, 2008; Tsilidis, 2011). As an exception, the relative risk of cervical dysplasia and cervical cancer is higher in current COC users, but this declines after use is discontinued. Following 10 or more years, risk returns to that of never users (International Collaboration of Epidemiological Studies of Cervical Cancer, 2007). It is unclear whether COCs contribute to breast cancer development. Major studies show no risk or a small risk among current users, which drops with time following cessation (Collaborative Group on Hormonal Factors in Breast Cancer, 1996; Hannaford, 2007; Marchbanks, 2002).

Although COC use in the past was linked to development of hepatic focal nodular hyperplasia and benign hepatic adenoma, large studies do not support this (Heinemann, 1998). Moreover, no evidence supports concern for greater risk of hepatocellular cancer (Maheshwari, 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, 2009b). Rates of colorectal cancer appear to be reduced in ever users (Bosetti, 2009; Luan, 2015).

Other Effects

Cholestasis and cholestatic jaundice are uncommon, but they resolve when COCs are discontinued. In women who have active hepatitis, COCs should not be initiated, but these may be continued in women who experience a flare of their liver disease while already taking COCs. Use of progestin-only contraception in these women is not restricted. Moreover, there is no reason to withhold COCs from women who have recovered. Mild compensated cirrhosis does not limit the use of COCs or progestin-only methods. But in those with severe decompensated disease, all hormonal methods are avoided (Kapp, 2009a).

Chloasma, which is hyperpigmentation of the face and forehead, is more likely in women who demonstrated such a change during pregnancy (Chap. 4, Breasts). This is less common with low-dose estrogen formulations. Although previously used for treating functional ovarian cysts, low-dose COC formulations have been shown to have no effects related to cyst resolution or prevention (European Society of Human Reproduction and Embryology, 2001; Grimes, 2014).

Many noncontraceptive benefits are associated with COC use (American College of Obstetricians and Gynecologists, 2016c). And indeed, COCs may be used for these effects, even in those without contraceptive needs. Dysmenorrhea and heavy menstrual bleeding lessen with COC use. Another action is to improve androgenic conditions such as acne and hirsutism. For women with premenstrual dysphoric disorder (PMDD), several studies have shown symptom improvement in those who use the drospirenone-containing COC Yaz (Lopez, 2012b; Pearlstein, 2005; Yonkers, 2005).

Transdermal Patch

The Ortho Evra patch contains ethinyl estradiol and the progestin norelgestromin. It has an inner layer containing an adhesive and hormone matrix, and a water-resistant outer layer. Thus, women can wear the patch in bathtubs, showers, swimming pools, saunas, and whirlpools without decreased efficacy. The patch may be applied to buttocks, upper outer arm, lower abdomen, or upper torso, but the breasts are avoided. Because the hormones are combined with the adhesive, improper skin adherence will lower hormone absorption and efficacy. Therefore, if a patch is so poorly adhered that it requires reinforcement with tape, it should be replaced.

Initiation of the patch is the same as for COCs, and a new patch is applied weekly for 3 weeks, followed by a patch-free week to allow withdrawal bleeding. Although a patch is ideally worn no longer than 7 days, hormone levels remain in an effective range for up to 9 days. This affords a 2-day window for patch-change delays (Abrams, 2001).

In general, the transdermal patch and vaginal ring produce metabolic changes, side effects, and efficacy rates comparable to those with COC pills. However, the patch has been associated with a higher thromboembolism risk in some but not all studies (Cole, 2007; Jick, 2010; Lidegaard, 2011). In response, the Food and Drug Administration (2015b) approved labeling for the patch to state that the risk for VTE may be increased compared with other COCs, and relative risk estimates range from 1.2 to 2.2. Obesity—90 kg or greater—may be associated with a higher risk for patch contraceptive failure (Janssen Pharmaceuticals, 2015; Zieman, 2002). Finally, application-site reaction and breast tenderness are more frequent during initial cycles in patch wearers (Urdl, 2005).

Transvaginal Ring

The NuvaRing is yet another form of combination hormonal contraception and is a flexible intravaginal ring. The ring is constructed of ethinyl vinyl acetate, and it measures 54 mm in diameter and 4 mm in cross section (Fig. 38-5). During insertion, the ring is compressed and threaded into the vagina, but no specific final orientation within the vagina is required. Its core releases ethinyl estradiol and the progestin etonogestrel, which are absorbed across the vaginal epithelium. Before being dispensed, the rings are refrigerated, and once dispensed, their shelf life is 4 months. The ring is placed within 5 days of menses onset and, after 3 weeks of use, is removed for 1 week to allow withdrawal bleeding. Contraception will still be afforded if a ring is left in place for a fourth week (Merck, 2016b).

Patient satisfaction is high with this method, although vaginitis, ring-related events, and leukorrhea are more common (Lopez, 2013b; Oddsson, 2005). Despite this, no deleterious affect on lower reproductive tract or endometrial epithelia has been found (Lete, 2013; Veres, 2004). A ring may be used concurrently with vaginal medications or with a tampon (Haring, 2003; Verhoeven, 2004a,b). Approximately 70 percent of partners feel the ring during intercourse (Dieben, 2002). If this is bothersome, the ring may be removed for intercourse but should be replaced within 3 hours to maintain efficacy.

Injectable Progestin Contraceptives

Both intramuscular depot medroxyprogesterone acetate—Depo-Provera (DMPA)—150 mg every 3 months, and norethisterone enanthate, 200 mg every 2 months, are injectable progestin contraceptives used worldwide. Of the two, DMPA is available in the United States. DMPA is injected into the deltoid or gluteus muscle, but massage is avoided to ensure that the drug is released slowly. Alternatively, a subcutaneous version, depo-subQ provera 104, is also available and is injected into the subcutaneous tissue of the anterior thigh or abdomen every 3 months.

DMPA is effective, and as with other progestin-only methods, contraception is provided by ovulation inhibition, greater cervical mucus viscosity, and creation of an endometrium unfavorable for ovum implantation. Initial injection is given within the first 5 days following menses onset. Serum levels sufficient for contraception are observed by 24 hours. Thus, no additional contraceptive method is required for initiation within this window. Alternatively, limited data support a “quick wtart,” or initiation of DMPA regardless of cycle day. If so implemented, investigators recommend an initial negative pregnancy test result before injection, a supplemental contraceptive method during the 7 days following injection, and a second pregnancy test after 3 to 6 weeks to identify an early pregnancy (Rickert, 2007; Sneed, 2005). Pregnancies conceived during DMPA use are not associated with a higher risk of fetal malformation (Katz, 1985). For women who present for intramuscular DMPA reinjection more than 13 weeks or for subcutaneous DMPA reinjection more than 14 weeks after the prior dose, the manufacturer recommends exclusion of pregnancy before reinjection (Pfizer, 2015a,b).

Actions and Side Effects

Injected progestins offer the convenience of a 3-month dosing schedule, contraceptive efficacy comparable with or better than COCs, and minimal to no lactation impairment. Iron-deficiency anemia is less likely in long-term users because of amenorrhea, which develops in up to 50 percent after 1 year and in 80 percent after 5 years.

Similar to other progestin-only contraceptive, irregular menstrual bleeding is common, and a fourth of women discontinued DMPA in the first year because of this (Cromer, 1994). Unique to DMPA, prolonged anovulation can follow discontinuation, which results in delayed fertility resumption. After injections are stopped, a fourth of patients do not resume regular menses for up to 1 year (Gardner, 1970). Accordingly, DMPA may not be ideal for women who plan to use birth control only briefly before attempting conception.

As with other progestins, DMPA has not been associated with cardiovascular events or stroke in otherwise healthy women. However, in those with severe hypertension, a higher risk of stroke has been found in DMPA users (World Health Organization, 1998). Moreover, the US MEC authors express concerns regarding hypoestrogenic effects and reduced HDL levels from DMPA in women with vascular disease or multiple risks for cardiovascular disease.

Weight gain is generally attributed to DMPA, and these increases are comparable between the two depot forms (Bahamondes, 2001; Vickery, 2013; Westhoff, 2007c). In long-term users, loss of bone mineral density is also a potential problem (Petitti, 2000; Scholes, 1999). In 2004, the FDA added a black box warning to DMPA labeling, which notes that this concern is probably most relevant for adolescents, who are building bone mass, and perimenopausal women, who will soon have increased bone loss during menopause. That said, World Health Organization (1998) and American College of Obstetricians and Gynecologists (2016b) believe that DMPA should not be restricted in those high-risk groups. And, it seems prudent to reevaluate overall risks and benefits during extended use. It is somewhat reassuring that bone loss appears to be reversible after discontinuation of therapy, although reversal is still not complete after 18 to 24 months (Clark, 2006; Scholes, 2002).

Progestin-Only Pills

So-called mini-pills are progestin-only contraceptives that are taken daily. These contraceptives have not achieved widespread popularity and are used by only 0.4 percent of reproductive-aged American women (Hall, 2012). Unlike COCs, they do not reliably inhibit ovulation. Rather, their effectiveness depends more on cervical mucus thickening and endometrial atrophy. Because mucus changes are not sustained longer than 24 hours, mini-pills should be taken at the same time every day to be maximally effective. If a progestin-only pill is taken even 4 hours late, a supplemental form of contraception must be used for the next 48 hours. Progestin-only pills are contraindicated in women with known breast cancer or pregnancy. Other cautions are listed in Table 38-2.

Male Condom

For many years, male and female condoms, vaginal diaphragms, and periodic abstinence have been used for contraception with variable success (see Table 38-2). When used properly, condoms provide considerable but not absolute protection against a broad range of sexually transmitted diseases, including HIV (Eaton, 2014). Contraceptive efficacy of the male condom is enhanced appreciably by a reservoir tip and probably by the addition of a spermicide. Such agents, as well as those used for lubrication, should be water-based because oil-based products degrade latex condoms and diaphragms.

For individuals sensitive to latex, condoms made from lamb intestines are effective, but they do not provide infection protection. Fortunately, nonallergenic condoms have been developed that are made of polyurethane or of synthetic elastomers. Polyurethane condoms are effective against STDs but have a higher breakage and slippage rate compared with latex condoms (Gallo, 2012a).

Female Condom

The only female condom available in the United States is marketed as the FC2 Female Condom. It is a synthetic nitrile sheath with one flexible polyurethane ring at each end. Its open ring remains outside the vagina, whereas its closed internal ring is fitted under the symphysis like a diaphragm (Fig. 38-6). The female condom can be used with both water-based and oil-based lubricants. Male condoms should not be used concurrently because simultaneous use may cause friction that leads to condom slipping, tearing, and displacement. Following use, the female condom outer ring should be twisted to seal the condom so that no semen spills. As an added value, the female condom may offer some protection against STDs (Minnis, 2005).

Diaphragm Plus Spermicide

The diaphragm consists of a circular latex dome of variable diameter supported by a circumferential latex-covered metal spring. It is effective when used in combination with spermicidal jelly or cream. The spermicide is applied into the dome cup and along the device rim. The diaphragm is then positioned so that the cup faces the cervix and that the cervix, vaginal fornices, and anterior vaginal wall are partitioned effectively from the remainder of the vagina and the penis. In this fashion, the centrally placed spermicide is held against the cervix. When appropriately positioned, one rim is lodged deep in the posterior vaginal fornix, and the opposite rim fits behind the inner surface of the symphysis and immediately below the urethra (Fig. 38-7). If a diaphragm is too small, it will not remain in place. If it is too large, it is uncomfortable when forced into position. A coexistent cystocele or uterine prolapse typically leads to instability and expulsion. Because size and spring flexibility must be individualized, the diaphragm is fitted by providers and available only by prescription.

For use, the diaphragm and spermicide can be inserted hours before intercourse. If more than 6 hours elapse, the diaphragm can remain but additional spermicide is placed in the upper vagina for maximum protection. Spermicide is reapplied before each subsequent coital episode. The diaphragm is not removed for at least 6 hours after intercourse. Because toxic shock syndrome has been described following its use, it may be worthwhile to remove the diaphragm at 6 hours, or at least the next morning, to minimize this rare event. Diaphragm use is associated with a slightly greater rate of urinary infections, presumably from urethral irritation by the ring under the symphysis.

Cervical Cap

FemCap is currently the only available cervical cap in the United States. Made of silicone rubber, it has a sailor-cap shape with a dome that covers the cervix and a flared brim, which allows the cap to be held in place by the muscular walls of the upper vagina. Available in 22-, 26-, and 30-mm sizes, it is used with a spermicide applied once at insertion to both sides of the dome cup. For contraception, it should remain in place for 6 hours following coitus and may remain for up to 48 hours. Even with proper fitting and correct use, pregnancy rates with this method are higher that with the diaphragm (Gallo, 2012b; Mauck, 1999).

These family planning methods attempt to identify the fertile days each cycle and advise sexual abstinence during these days. Their major drawback is their limited efficacy, which is shown in Table 38-1. Common forms of these fertility awareness-based (FAB) methods include Standard Days, Temperature Rhythm, Cervical Mucus, and Symptothermal Methods. Some smartphone applications aim to assist these practices (Fehring, 2013).

The Standard Days Method counsels women to avoid unprotected intercourse during cycle days 8 through 19. For successful use, women must have regular monthly cycles of 26 to 32 days. Those who use this method can mark a calendar or can use Cycle-Beads, which is a ring of counting beads, to keep track of their days.

The Temperature Rhythm Method relies on a sustained 0.4°F rise in the basal body temperature, which usually precedes ovulation. For maximum efficacy, the woman must abstain from intercourse from the first day of menses through the third day after the temperature increase.

The Cervical Mucus Method, also called the Two-Day Method or Billings Method, relies on awareness of vaginal “dryness” and “wetness.” These reflect changes in the amount and quality of cervical mucus at different times in the menstrual cycle. With the Billings Method, abstinence is required from the beginning of menses until 4 days after slippery mucus is identified. With the Two-Day Method, intercourse is considered safe if a woman did not note mucus on the day of planned intercourse or the day prior.

The Symptothermal Method combines changes in cervical mucus—onset of fertile period; changes in basal body temperature—end of fertile period; and calculations to estimate the time of ovulation. This method is more complex to learn and apply, but it does not appreciably improve efficacy.

These contraceptives are marketed variously as creams, jellies, suppositories, films, and aerosol foam. Most can be purchased without a prescription. They are considered a less effective method (see Table 38-1). If pregnancy does occur, they are not teratogenic (Briggs, 2015).

Typically, spermicides function by providing a physical barrier to sperm penetration and a chemical spermicidal action. The active ingredient is nonoxynol-9 or octoxynol-9. Although these are spermicidal, they do not provide STD protection. Ideally, 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, and thereafter, they must be reinserted before repeat intercourse. Douching is avoided for at least 6 hours after coitus.

Contraceptive Sponge

The Today contraceptive sponge is an over-the-counter, one-size-fits-all device. The nonoxynol-9–impregnated polyurethane disc is 2.5 cm thick and 5.5 cm wide and has a dimple on one side and satin loop on the other (Fig. 38-8). The sponge can be inserted up to 24 hours prior to intercourse, and while in place, it provides contraception regardless of coital frequency. It should remain in place for 6 hours after intercourse. Pregnancy is prevented primarily by the spermicide nonoxynol-9 and to a lesser extent by covering the cervix and absorbing semen.

Although the sponge is possibly more convenient than the diaphragm or condom, it is less effective than either (Kuyoh, 2013). Most common causes for method discontinuance are pregnancy, irritation, discomfort, or vaginitis (Beckman, 1989). Although toxic shock syndrome has been reported with the contraceptive sponge, it is rare, and evidence suggests that the sponge may actually limit production of the responsible staphylococcal exotoxin (Remington, 1987). Still, it is recommended that the sponge not be used during menses or the puerperium.

Following unprotected sexual intercourse, many women present for contraceptive care. Several emergency contraception (EC) regimens substantially lower the likelihood of an unwanted pregnancy when used correctly. Current methods include COCs, progestins, progesterone antagonists, and copper-containing IUDs (Table 38-3). Overall, the IUD is most effective, and ulipristal acetate is the most efficient oral regimen (American College of Obstetricians and Gynecologists, 2017a). Patients can obtain information regarding emergency contraception by calling 1–888-NOT-2-LATE or accessing The Emergency Contraception Website: http://not-2-late.com.

Hormonal Emergency Contraception

Except for allergy to a particular component, no conditions in the US MEC contraindicate hormonal EC methods. With progestin-only regimens, levonorgestrel is taken as a single, one-time 1.5-mg dose (Arowojolu, 2002). This is now recommended instead of two 0.75-mg doses separated by 12 or 24 hours (Ngai, 2005). Dosing begins ideally within 72 hours of unprotected coitus but may be given up to 120 hours. Notably, the single-dose regimen is available over-the-counter without a prescription to all reproductive-aged women (Food and Drug Administration, 2013, 2015a).

One progesterone-receptor modulator currently available for EC is ulipristal acetate and is marketed as Ella. It is taken as a single 30-mg tablet up to 120 hours after unprotected intercourse (Brache, 2010; Watson, 2010).

Also known as the Yuzpe method, this older EC method provides a minimum of 100 μg of ethinyl estradiol and 0.5 mg of levonorgestrel in each of two doses. As shown in Table 38-3, a sufficient dose may be achieved by two or more pills. The first dose is taken ideally within 72 hours of intercourse but may be given up to 120 hours. The initial dose is followed 12 hours later by a second equivalent dose.

The major mechanism with all hormonal regimens is inhibition or delay of ovulation. Of oral methods, failure rates are lowest with ulipristal (1 to 2 percent) and greatest with the Yuzpe method (2 to 3.5 percent) (Cleland, 2014). If EC fails to prevent pregnancy or is mistimed, no associations with major congenital malformation or pregnancy complications have been noted with these hormonal methods (Jatlaoui, 2016; Levy, 2014).

With EC administration, nausea and vomiting can be an important side effect (American College of Obstetricians and Gynecologists, 2015b; Gemzell-Danielsson, 2013). Accordingly, an oral antiemetic may be prescribed at least 1 hour before each dose (Rodriguez, 2013). If a woman vomits within 2 hours of a dose, the dose is repeated.

Copper-Containing Intrauterine Devices

For women who are candidates, Cu-IUD insertion is the most effective emergency contraceptive method and provides an effective 10-year method of contraception (Cheng, 2012). If an IUD is placed up to 5 days after unprotected coitus, the failure rate approximates only 0.1 percent (Cleland, 2012; Wu, 2010).

For mothers who are nursing exclusively, ovulation during the first 10 weeks after delivery is unlikely. Nursing, however, is not a reliable method of family planning for women whose infants are on a daytime-only feeding schedule. Moreover, waiting for first menses involves a risk of pregnancy, because ovulation usually antedates menstruation. Certainly, after the first menses, contraception is essential unless the woman desires pregnancy.

As shown in Table 38-2, all methods may be suitable for nursing mothers after the initial weeks, during which thromboembolism risks are still great. With all hormonal methods, very small quantities are excreted in breast milk, but no adverse effects on infants have been reported (Phillips, 2015; World Health Organization, 1988). Although not robust, some older studies link decreased infant weight gain or milk volume with early initiation of combination oral contraceptives before 6 weeks postpartum (Lopez, 2015c; Tepper, 2016a).