Implant contraception

  Implant contraception

  Subdermal contraceptive implants offer women long-acting controlled release of progestins. Since the 1980s, these implants have been approved in more than 60 countries and used by 11 million women worldwide. Their high efficacy, along with their ease of use, makes them a good contraceptive option for several groups of women: those who require progestin-only methods because they should not use estrogen, teens who find adherence to a contraceptive regimen difficult, and healthy adults who desire long-term protection. The levonorgestrel (LNG) implant, no longer marketed in the United States, garnered a million American users but was sometimes difficult to insert and remove. A highly effective and long-lasting single-rod etonogestrel (ENG) implant (Implanon) makes implant contraception available again in the United States.

Benefits of Implantable Contraception

 

  Implants offer a variety of benefits to women. They provide long-term, effective prevention of pregnancy without the need for daily or weekly action, as with pills, rings or patches, or for regular injections. Implant contraception is cost-effective compared with short-acting methods or an unplanned pregnancy.The sustained administration of a low dose of progestin and maintenance of stable serum levels provide high efficacy, a long duration of action, a short fertility recovery time, and no reported serious cardiovascular effects.
  Research and development of progestin-only subdermal implants began in the 1960s, but initial research with implants containing very low doses of progestins found that they were unsuccessful in preventing ectopic pregnancies. Development of the LNG implant, a 6-capsule implantable system using the potent progestin LNG, followed. In 1991 it became the first FDA-approved contraceptive implant. A million U.S. women chose the LNG implant as their contraceptive, and it proved highly effective: over a 7-year duration of use, only about 1% of users became pregnant. Despite low rates of pregnancy and few serious side effects, limited supplies of the Silastic components and negative media coverage led to the LNG implant's withdrawal from distribution in 2002, which left no implant alternative for American women.
  The 15-year experience with the LNG implant instigated further development and improvements in implant design. A 2-rod LNG system was FDA approved in 1998 but never marketed in the United States, although it has many users in Europe and Asia. The FDA has reviewed a contraceptive implant containing 68 mg of etonogestrel (ENG), the active metabolite of desogestrel, in a single rod made of ethylene vinyl acetate (EVA). This 1-implant contraceptive, the most effective hormonal method of birth control yet developed, is now used by more than 2 million women in Europe and Asia. In Australia, a quarter of all women practicing contraception in 2004 used this implant.

Progestin-Only Contraception

 

   Levonorgestrel binds with high affinity to the progesterone, androgen, mineralocorticoid, and glucocorticoid receptors, but not to estrogen receptors. Etonogestrel has shown no estrogenic, antiinflammatory, or mineralocorticoid activity, but it has shown weak androgenic and anabolic activity and strong antiestrogenic activity. Unlike LNG, which is bound mainly to sex hormone–binding globulin, ENG is bound mainly to albumin, which is not significantly affected by varying endogenous or exogenous estradiol concentrations. The overall safety of ENG has been demonstrated through studies of combined estrogen-progestin OCs, the combined vaginal ring, and progestin-only OCs, all of which use desogestrel, a close relative of etonorgestrel, as a component.

Mechanisms of Action

 

  Progestin-containing implants have two primary mechanisms of action: inhibition of ovulation and restriction of sperm penetration through cervical mucus. The LNG implant disrupts follicular growth and inhibits the ovulatory process by exerting negative feedback on the hypothalamic-pituitary axis, causing a variety of changes that range from anovulation to insufficient luteal function. A small number of women using LNG implants will have quiescent ovaries, but most will begin to ovulate as blood concentrations of LNG gradually fall. The ENG implant suppresses ovulation by altering the hypothalamic-pituitary-ovarian axis and down-regulating the luteinizing hormone surge, which is required to support the production, growth, and maturation of ovarian follicles. Some women , especially those with higher endogenous estrogen, form ovarian cysts while using implants. These are not pathologic, are rarely painful and do not require treatment.
  Even if follicles grow during use of progestin implants, oocytes are not fertilized. If the follicle ruptures, the abnormalities of the ovulatory process appear to prevent release of a viable egg. Antiestrogenic actions of the progestins affect the cervical mucus, making it viscous, scanty, and impenetrable by sperm. These ovarian and cervical mechanisms of action provide high contraceptive efficacy and occur prior to fertilization. No signs of embryonic development have been found among implant users, a finding that indicates that progestin implants have no abortifacient properties.


   The high efficacy of Single-Rod Etonogestrel Implant- new device results from the capacity of a low dose of ENG to suppress ovulation and from its long duration of action. After subdermal insertion, users need do nothing more to have nearly complete protection from pregnancy for up to 3 years. The one rod is a great improvement over the old six-capsule LNG implant system, both in time and in ease of insertion. In the U.S. and European trials of the ENG implant system, which began in 1995, average insertion time was 1 minute and removal time was 3 minutes—much faster than with the LNG implant system.
  Its convenience is enhanced by other design features as well. The inserter is preloaded and disposable, and because only one rod is implanted, there is no chance that insertion of additional capsules will move previously placed capsules out of position. In addition, EVA, the plastic from which it is fabricated, makes it is less likely than older Silastic implants to form a fibrous sheath that can make removal more difficult. These differences likely simplify insertion and removal, which for patients means little discomfort, an unobtrusive implant, and almost no scarring. For clinicians it means faster and simpler insertion and removal procedures.

Pharmacology

 

  The ENG implant (Implanon) consists of one nonbiodegradable rod of 40% EVA and 60% ENG (40 mm × 2 mm) covered with a rate-controlling EVA membrane 0.06 mm thick. The rod contains 68 mg ENG, initially absorbed by the body at a rate of 60 mg per day and slowly declining to 30 mg per day after 2 years of use. The high initial rate of absorption is probably due to a significant amount of ENG released from the uncovered ends of the implant. Peak serum concentrations (266 pg/mL) of ENG are achieved within 1 day after insertion, suppressing ovulation. Like other contraceptive steroids, serum levels of ENG are reduced in women taking liver enzyme-inducing drugs such as carbamazepine, griseofulvin, phenytoin, and rifampicin but are not affected by antibiotics.
   Steady release of ENG into the circulation avoids first-pass effects on the liver. Bioavailability of ENG remains nearly 100% throughout 2 years of use. The elimination half-life of ENG is 25 hours, compared with 42 hours for LNG. After implant removal, serum ENG concentrations become undetectable within 1 week. Return of ovulation occurs in 94% of women within 3 to 6 weeks after the method is discontinued.

Efficacy

 

   The efficacy of the single-rod ENG implant was established as greater than 99% in clinical trials of 1100 women for a total of 26,700 cycles of use. Of these, 835 women completed 2 to 3 years and an additional 526 used the ENG implant for 3 years or longer. Neither intrauterine nor ectopic pregnancies were observed during these trials. Among the subjects were 365 women whose body weight was 154 pounds (70 kg) or more, none of whom became pregnant.

Metabolic Effects

 

  Published studies regarding the ENG implant indicate that metabolic effects are minimal and unlikely to be clinically significant. The ENG implant does not appear to have any clinically meaningful effect on lipid metabolism, carbohydrate metabolism, liver function, hemostatic factors, blood pressure, thyroid, or adrenal function.

Safety

  Overall, implants, including the ENG implant, are regarded as safe, with rates of adverse events (including death, neoplastic disease, cardiovascular events, anemia, hypertension, bone density changes, diabetes, gall bladder disease, thrombocytopenia, and pelvic inflammatory disease) comparable with those in women not using implants. The ENG implant reduced or eliminated menstrual pain in 88% of women previously experiencing dysmenorrhea; pain increased in only 2% of the ENG implant users. In a study comparing 42 pairs of infants and lactating mothers using the ENG implant with 38 pairs using intrauterine devices, there were no significant differences between groups in milk volume, milk constituents, timing or amount of supplementary food, or infant growth rates.
   Low-dose progestin contraceptives have few contraindications. They may be less effective in obese women and in those using drugs that stimulate the liver's cytochrome metabolism of steroids, such as some antibiotics (e.g., rifampin) and some anticonvulsants (e.g., phenytoin).

Insertion and Removal

 

Although the ENG implant is designed to facilitate rapid and simple insertion and removal, clinicians should first be trained. Insertion is less complex than with the six-capsule LNG implant, and average insertion time for the ENG implant is 1 to 2 minutes. The disposable applicator comes preloaded. The tip of the needle has two cutting edges, each with a different slope . The extreme tip is more angulated than the remainder and is sharp to facilitate initial penetration of the skin. The upper portion is less angulated and unsharpened to reduce the risk of reperforating the skin during advancement of the implant after initial penetration.

  For easy removal, the implant should be placed subdermally on the inner aspect of the nondominant arm, 6 to 8 cm above the elbow. After disinfecting the insertion site, a small amount (0.1 mL) of local anesthetic can be injected with a tuberculin syringe at the site of skin penetration, but this is not usually necessary because the 19-gauge insertion trocar is very sharp. After insertion, the implant might not be visible but should be palpable.
  For women who either have not been using a contraceptive method or have been using a nonhormonal method, insertion should occur between days 1 and 5 of menses. For women changing from a combination or progestin-only OC or from intrauterine contraception, the ENG implant can be inserted during a hormone-free week. For women changing from injectable contraception, insertion should occur on the day the next injection is scheduled. No backup is necessary if timing of insertion occurs as detailed in the product information. If the ENG implant is the contraceptive method selected following an abortion or delivery, it can be inserted immediately; no additional contraceptive method is required. In all cases, pregnancy should be excluded before insertion, although there is no evidence that hormonal contraceptives cause birth defects.
   The ENG implant can be removed at any time at the woman's discretion, but if left in place it remains effective for 3 years. Removal requires making a 2- to 3-mm incision at the distal tip of the implant and pushing the other end of the rod until it pops out. Mean removal time is commonly less than 5 minutes (2.6 to 5.4 minutes). Pain, swelling, redness, and hematoma have been reported during insertion and removal. Because return to ovulation is rapid following removal, women still desiring contraception should begin another method immediately or have a new rod inserted through the removal incision.

Disadvantages

 

  Although progestin-only contraceptives offer a safe and effective method of preventing pregnancy, they have some drawbacks. Implants require a minor surgical procedure by a trained clinician for insertion and removal. Cost-effectiveness of the method depends upon long-term use; early discontinuation negates this benefit. Lack of protection against sexually transmitted infections (STIs) is a disadvantage of the ENG implants as well as of all hormonal contraceptive methods.
  Side effects associated with the ENG implant include menstrual irregularities (infrequent bleeding (26.9%), amenorrhea (18.6%), prolonged bleeding (15.1%), frequent bleeding (7.4%)), weight gain (20.7%), acne (15.3%), breast pain (9.1%), and headache (8.5%), but these symptoms rarely provoked discontinuation.  Women using any of the progestin-only methods notice changes in bleeding patterns.  A comparative study of bleeding patterns in single-capsule ENG implant users and six-capsule LNG implant users found a statistically significant decrease in mean number of bleeding or spotting days in the ENG users (15.9 to 19.3 days versus 19.4 to 21.6 days; P = .0169). Because total uterine blood loss is reduced, users of progestin-only contraceptives (as well as OCs) are less likely to be anemic. However, the study also found that users of ENG implants had more variable bleeding patterns than users of the LNG implants. Table 1 -10 shows the differences in bleeding between implants, but it is impossible to predict which of these patterns a woman is likely to experience. Despite side effects and dependence on clinicians to insert implants, most women using implantable contraception are satisfied with the method and cite its long duration of use, convenience, and high efficacy.  

TABLE 1-10 - COMPARISON OF VAGINAL BLEEDING PATTERNS IN ENG AND LNG USERS

Bleeding Pattern[*]	ENG Implant (%) (n = 169)	LNG Implant (%) (n = 163)	Statistically Significant
Amenorrhea	21	5	Yes
Infrequent bleeding	27	22	Yes
Frequent bleeding	6	4	No
Prolonged bleeding	12	9	No

ENG, etonogestrel; LNG, levonorgestrel.

*	90-day reference periods.

Discontinuation Rates

 

   Discontinuation rates for the ENG implant have varied by area of use, ranging from 30.2% in Europe and Canada to 0.9% in Southeast Asia.  Bleeding irregularities are cited as the most common reason for discontinuation of the ENG implant. A meta-analysis of 13 studies published between 1989 and 1992 found that among 1716 women using the ENG implant, 5.3% discontinued in months 1 through 6, 6.4% discontinued in months 7 through 12, 4.1% discontinued in months 13 through 18, and 2.8% discontinued in months 19 through 24. Overall, 82% of women continued to use the ENG implant for up to 24 months.

Counseling

  Counseling women to expect bleeding irregularities reduces discontinuation of this system. Prospective users should receive complete information about bleeding irregularities so they can make informed decisions regarding the side effects they are willing to accept in order to benefit from high contraceptive efficacy. Preinsertion counseling and postinsertion follow-up are essential for continued use of implants. Satisfaction with the method increases with proper counseling and minimizes costly removals.
  Sexually active women are exposed to the risk of pregnancy as well as to the risk of STIs, such as human immunodeficiency virus, hepatitis B, human papillomavirus, Chlamydia trachomatis, syphilis, and gonorrhea, whose sequelae may be life threatening. Implantable contraceptives neither increase the risk of nor offer protection against STIs. Women counseled about contraception should also be informed about the risks of STIs and advised that use of condoms concomitantly with an effective method of pregnancy prevention is the best means of protection against unintended pregnancy and STIs. It seems likely that the ENG implant, like OCs and DMPA, reduces the risk of pelvic upper tract infection.
   Clinical experience with the ENG implant has demonstrated that method effectiveness and satisfaction are closely associated with patient education and provider training. In the first 18 months after the ENG implant was introduced in Australia in May 2001, an unexpectedly high number of adverse incidents were reported, and 100 unintended pregnancies occurred. Almost all of these events were traced to improper insertion by untrained clinicians or to poor patient selection, timing, and counseling. Policies that adequately document the process, procedure, and patient consent were initiated by the Royal Australian College of General Practitioners and have corrected the problems.
Intrauterine Hormonal Contraception
  Both medicated and nonmedicated intrauterine devices have been used for contraception. Those containing copper or synthetic progestins are more effective and have fewer side effects than simple plastic or stainless steel devices, and they have largely replaced the earlier varieties. Much information about intrauterine contraception is outdated and irrelevant because it is based on studies of these old contraceptives.
   The LNG-20 (Mirena), manufactured by Leiras-Schering AG in Finland, releases in vitro 20 mg of levonorgestrel per day. The vertical arm of this T-shaped device has a collar that contains 52 mg levonorgestrel dispersed in polydimethylsiloxane. The LNG is released initially at a rate of 20 mg per day in vivo, progressively declining and reaching half of the initial rate after 5 years. The levonorgestrel intrauterine delivery system (IUS) is approved for 5 years, but it lasts up to 10 years, and reduces menstrual blood loss and pelvic infection rates.  Indeed, the levonorgestrel IUS is about as effective as endometrial ablation for the treatment of menorrhagia. The local progestin effect directed to the endometrium can be used in patients taking tamoxifen, patients with dysmenorrhea, and postmenopausal women receiving estrogen therapy.  Smaller devices releasing 5 mg or 10 mg levonorgestrel have been developed in Europe for use for at least 5 years in postmenopausal women.
   The progestin-releasing IUS adds the endometrial action of the progestin to the foreign body reaction. The endometrium becomes decidualized with atrophy of the glands. The progestin IUS probably has two mechanisms of action: inhibition of implantation and inhibition of sperm capacitation, penetration, and survival. The levonorgestrel IUS produces serum concentrations of the progestin about half those of Norplant so that ovarian follicular development and ovulation are also partially inhibited; after the first year, cycles are ovulatory in 50% to 75% of women, regardless of their bleeding patterns. Finally, the progestin IUS thickens the cervical mucus, creating a barrier to sperm penetration. The progestin IUS decreases menstrual blood loss (about 40% to 50%) and dysmenorrhea; with the levonorgestrel IUS, bleeding can be reduced by 90% 1 year after insertion. About 50% of women become amenorrheic 1 year after insertion of the levonorgestrel IUS.  Average hemoglobin and iron levels increase over time compared with preinsertion values.

Efficacy

  The copper-releasing intrauterine contraceptive (IUC) (TCu-380A or Paragard) is approved for use in the United States for 10 years. However, the TCu-380A has been demonstrated to maintain its efficacy over at least 12 years of use. The TCu-200 is approved for 4 years and the Nova T for 5 years. The levonorgestrel IUS can be used for at least 7 years and probably 10 years. The levonorgestrel device that releases 15 to 20 mg levonorgestrel per day is as effective as the new copper IUCs.
  The nonmedicated IUCs never have to be replaced. The deposition of calcium salts on the IUC can produce a structure that is irritating to the endometrium. If bleeding increases after a nonmedicated IUC has been in place for some time, it is worth replacing it. Some clinicians recommend replacing all older IUCs with the new, more effective copper IUCs. This makes sense because newer IUCs are more effective.
  The risk of ectopic pregnancy does not increase with increasing duration of use with the TCu-380A or the levonorgestrel IUS.  In a 7-year prospective study, not a single ectopic pregnancy was encountered with the levonorgestrel IUD, and in a 5-year study, only one.  In 8000 woman-years of experience in randomized multicenter trials, there has been only a single ectopic pregnancy reported with the TCu-380A (which is one-tenth the rate with the Lippes Loop or TCu-200). Therefore, the risk of ectopic pregnancy during the use of the copper IUS or the levonorgestrel IUC is much lower compared with rates in noncontraceptive users; however, if pregnancy occurs, the likelihood of an ectopic pregnancy is high.
  The protection against ectopic pregnancy provided by the TCu-380A and the levonorgestrel IUD makes these IUDs acceptable choices for contraception in women with previous ectopic pregnancies ( Table 1-11 ).

TABLE 1-11 - ECTOPIC PREGNANCY RATES PER 1000 WOMAN-YEARS

Method	Rate
No contraceptive, all ages	3.00-4.50
Levonorgestrel IUS	0.20
TCu-380A IUC	0.20

IUC, intrauterine contraceptive; IUS, intrauterine delivery system.

Side Effects

 

  With effective patient screening and good insertion technique, the copper and medicated IUCs are not associated with an increased risk of infertility after their removal. Even if IUCs are removed for problems, subsequent fertility rates are normal.
   The symptoms most often responsible for IUC discontinuation are increased uterine bleeding and increased menstrual pain. Within 1 year, 5% to 15% of women discontinue IUC use because of these problems. Smaller copper and progestin IUDs have reduced the incidence of pain and bleeding considerably, but a careful menstrual history is still important in helping a woman consider an IUC. Women with prolonged, heavy menstrual bleeding or significant dysmenorrhea might not be able to tolerate copper IUCs but might benefit from a progestin IUC. Because bleeding and cramping are most severe in the first few months after IUC insertion, treatment with an NSAID (which inhibits prostaglandin synthesis) during the first several menstrual periods can reduce bleeding and cramping and help a patient through this difficult time. Even persistent heavy menses can be effectively treated with NSAIDs. NSAID treatment should begin at the onset of menses and be maintained for 3 days. A copper IUC is available in China that also releases a small amount of indomethacin; this device is associated with markedly less bleeding.
It is not unusual to have a few days of intermenstrual spotting or light bleeding. Although aggravating, this does not cause significant blood loss. Such bleeding deserves the usual evaluation for cervical or endometrial pathology. These changes can be objectionable for women who are prevented from having intercourse while bleeding.
Following insertion of a modern copper IUC, menstrual blood loss increases by about 55%, and this level of bleeding continues for the duration of IUC use. This is associated with a slight (1 to 2 days) prolongation of menstruation. Over a year's time, this amount of blood loss does not result in changes indicative of iron deficiency (e.g., serum ferritin). With longer use, however, ferritin levels are lower, suggesting a depletion of iron stores. Assessment for iron depletion and anemia should be considered in long-term users and in women susceptible to iron deficiency anemia. In populations with a high prevalence of anemia, these changes occur more rapidly, and iron supplementation is recommended.
Because of a decidualizing, atrophic impact on the endometrium, amenorrhea can develop over time with the progestin-containing IUS. With the levonorgestrel IUS, 70% of patients are oligomenorrheic and 30% to 40% are amenorrheic within 2 years.  In a group of women who used the levonorgestrel IUS for more than 12 years, 60% were amenorrheic; 12% experienced infrequent, scanty bleeding; and 28% had regular but light bleeding. For some women, the lack of periods is so disconcerting that they request removal. On the other hand, this effect on menstruation is manifested by an increase in blood hemoglobin levels.
The levonorgestrel IUS is very effective when used to treat menorrhagia.  This noncontraceptive benefit is of such a magnitude that this method of treatment achieves comparable results when compared to surgical methods such as endometrial ablation or hysterectomy.  Bleeding is even reduced in the presence of leiomyomas, along with a reduction in myoma size. The levonorgestrel IUS has been used successfully to treat endometriosis, and especially dysmenorrhea associated with endometriosis.
Sufficient progestin reaches the systemic circulation from the levonorgestrel-containing IUD so that androgenic side effects, such as acne and hirsutism, might occur; however, in one study no change could be detected in the circulating levels of sex hormone–binding globulin, and, therefore, clinical effects are unlikely. More extensive clinical studies are needed to assess the impact of this IUS on the lipoprotein profile, but it is unlikely that the low dose of levonorgestrel has an important effect on cardiovascular risk.

Infections

IUC-related bacterial infection is now believed to be due to contamination of the endometrial cavity at the time of insertion. Mishell's classic study indicated that the uterus is routinely contaminated by bacteria at insertion. Infections that occur 3 to 4 months after insertion are believed to be due to acquired STIs, not the direct result of the IUC. The early, insertion-related infections, therefore, are polymicrobial and are derived from the endogenous cervicovaginal flora, with a predominance of anaerobes.
A review of the World Health Organization database derived from all of the WHO IUC clinical trials concluded that the risk of pelvic inflammatory disease was six times higher during the 20 days after the insertion compared with later times during follow-up, but, most importantly, PID was extremely rare beyond the first 20 days after insertion. In nearly 23,000 insertions, however, only 81 cases of PID were diagnosed, and a scarcity of PID was observed in those situations in which STIs are rare. There was no statistically significant difference comparing the copper IUC with the inert Lippes Loop or progestin-containing IUC. These data confirm earlier studies that the risk of infection is highest immediately after insertion and that PID risk does not increase with long-term use.  The problem of infection can be minimized with careful screening and the use of aseptic technique. Even women with type 1 diabetes mellitus do not have an increased risk of infection.
Doxycycline (200 mg) or azithromycin (500 mg) administered orally one hour prior to insertion can provide protection against insertion-associated pelvic infection, but prophylactic antibiotics are probably of little benefit for women at low risk for STIs.
Compared with oral contraception, barrier methods, and hormonal IUCs, there is no reason to think that nonmedicated or copper IUCs can confer protection against STIs, specifically PID. However, the levonorgestrel-releasing IUC has been reported to be associated with a protective effect against pelvic infection, and the copper IUC is associated with lower titers of antichlamydial antibody.  In vitro, copper inhibits chlamydial growth in endometrial cells. Thus, the perception that IUC use is associated with pelvic infection (and infertility) is outmoded. Women who use IUCs must be counseled to use condoms along with the IUC whenever they have intercourse with a partner who could be an STI carrier. Because sexual behavior is the most important modifier of the risk of infection, clinicians should ask prospective IUC users about numbers of partners, their partner's sexual practices, the frequency and age of onset of intercourse, and history of STIs. Women at low risk are unlikely to have pelvic infections while using IUCs.