Hormonal contraceptives—progestogen implants

Levonorgestrel

Subdermal implanted silastic rods containing levonorgestrel (Norplant) have been widely studied and used with the support of the Population Council, but have fallen out of favor in a number of industrialized countries, because of complications. Six rods are inserted subdermally into the arm using a special device; each contains 35 mg of levonorgestrel, which is released into the circulation over 5 years. The total amount of levonorgestrel released daily during the first 500 days averages about 60 micrograms, falling to a plateau of 30 micrograms/day for the remainder of the 5-year period during which the product remains effective. It then has to be removed and replaced. The contraceptive effect of the implant does not seem to be caused primarily by ovulation inhibition, since after about the first year of use the hormone concentrations are too low to have this effect [ , ].

A product similar in approach to Norplant is the biodegradable subdermal capsular implant Capronor, which releases levonorgestrel over 12–18 months. Doses of 12 and 21.6 mg in capsules of differing size have been compared [ ]. Ovulation occurred in all cycles at the lower dose and in a quarter of cycles at the higher dose, so it is likely that the contraceptive effect results essentially from other progestogen-induced changes. Several users had local swelling or itching of the skin at the capsule insertion site, relieved by topical glucocorticoids.

The effectiveness, adverse events, and acceptability of the FDA-approved variant of levonorgestrel capsule implants in the USA over 5 years and the determinants of these outcomes have been studied [ ]. There were three pregnancies, yielding a 5-year cumulative rate of 1.3 per 100 users, an average annual rate of three per 1000 women. Ectopic pregnancy occurred at a rate of 0.6 per 1000 woman-years. There were no pregnancies in women who weighed less than 79 kg. Medical conditions that most often led to removal of the implant were prolonged or irregular menstrual bleeding, followed by headache, weight gain, and mood changes. Weight gain averaged 1 kg/year.

Etonogestrel

Implanon, which became available in 1999, is a subdermal implant that contains 68 mg of etonogestrel, the active metabolite of desogestrel. During the first 2 years after implantation, there is no ovulation; ovulation occurs occasionally in the third year, and it is recommended that the implant be removed after this time. As with Norplant, the principal problem is the occurrence of irregular bleeding [ ], which leads some 25–30% of users to ask that the implant be removed. After some time, amenorrhea occurs in some 20% of users. There was a clinically significant increase in bodyweight in 20% of women carrying the implant [ ]. There may also be some local irritation from the implant, and minor scarring at the implantation site [ , ]. In clinical trials the implant had no effect on coagulation measures, hemostasis, fat metabolism, or hepatic function, but according to the approved product information sheet the possibility of a slight increase in insulin resistance cannot be excluded.

Desogestrel

In clinical trials, implants containing desogestrel were effective, but with bleeding irregularities and ovarian cysts as the primary adverse reactions [ ].

Megestrol

Implants filled with megestrol acetate gave an unexpectedly high incidence of tubal pregnancies, while the absolute number of ectopic pregnancies was clearly greater than that expected for the general population [ ]. This finding reflects the same type of effect as has been described from time to time with oral progestogen-only products.

Norethisterone

A biodegradable delivery system using subcutaneous implants of fused pellets made from norethisterone and pure cholesterol has been tested [ ].

General adverse effects and adverse reactions

Generally speaking, the systemic effects of progestogen implants are very similar to those of low-dose oral progestogen-only contraception, except for variations due to progressive change in the amount of hormone released. Many studies using various metabolic measures have shown no significant pattern of deviation from normal values [ , ], either during use of the implant or after removal [ ].

Adverse reactions to levonorgestrel implants are similar to those observed with progestogen only and combined oral contraceptives. The risks of ectopic pregnancy, other pregnancy complications, and pelvic inflammatory disease are reduced in comparison with those in women who use copper or non-medicated intrauterine devices. The risks of gallbladder disease and frank or borderline hypertension, although small, are respectively about 1.5 and 1.8 times greater in women using levonorgestrel implants than in women not using hormonal contraception. Other serious diseases do not occur significantly more often in users of levonorgestrel implants than in women not using hormonal contraception. The great majority of levonorgestrel implant users have menstrual problems, but serious bleeding problems are not more frequent than in controls. In 16 000 women other health problems reported more frequently by users of levonorgestrel implants than by women not using hormonal contraception included skin conditions, headache, upper limb neuropathies, dizziness, nervousness, malaise, minor visual disturbances, respiratory conditions, arthropathies, weight change, anxiety, and non-clinical depression. Clinical depression is stated to be no more frequent in women using implants compared with those not using hormonal contraception (i.e. intrauterine devices or sterilization), but one must doubt this; certain women with a tendency to depression do experience triggering or aggravation of their condition with hormonal contraceptives, and it is not clear how this could differ between the various products. Problems with removing the device appear to be less frequent with Jadelle than with Norplant.

While the Norplant device has lost favor in some countries because of local intolerance and difficulties with removal, it continues to be widely used elsewhere. The difference in usage of Norplant between countries seems to reflect in part the degree of sophistication of users. Where, as in many western countries, low-dose oral contraceptives are very widely used and have become the standard by which users judge the acceptability of other means of contraception, in contrast the Norplant method has remained popular where expectations are somewhat lower.

Acceptability

As in the case of injectable depot contraceptives [ ] subdermal contraceptive implants are more likely to be used in developing countries, in order to overcome social and adherence problems, and that this can in principle involve risks to which underprivileged populations may be especially subject. All the same, acceptance is strikingly good [ ]. In a survey of data from eight developing countries, information was available on 7977 women starting Norplant, of whom 6625 used intrauterine devices and 1419 had been sterilized; most of the participants were followed for 5 years. All of the methods produced satisfactory degrees of contraception and, with a few exceptions, no characteristic morbidity was detected among Norplant users compared with the other groups. The two principal exceptions concerned gallbladder disease, which was 50% more common in women who used Norplant, and hypertension and borderline hypertension, the incidence of which was markedly raised in current implant users (RR = 1.81; 95% CI = 1.12, 2.92). Other unexpected findings were increased rates of respiratory disease and reduced risks of inflammatory disease of the genital tract in users of Norplant compared with sterilized women and those who used an intrauterine device [ ].

In Burkina Faso, experience in 1660 women over 4 years has been critically reviewed [ ]. There were 247 withdrawals before the fourth year, for various reasons, including cycle disorders (60 withdrawals), unspecified medical reasons (n = 53), personal objections (n = 47), weight gain (n = 14), and contraceptive failure (n = 2). Menstrual disorders, including amenorrhea, spotting, and hypermenorrhea, occurred in 51% of cases. The investigators stressed the need for a good information and sensitization campaign to reduce the number of implant withdrawals before the fourth year of use, since the product often seems to be withdrawn for insufficient reasons.

A study in Senegal produced similar findings over 5 years; the method was effective, but dissatisfaction with cycle control was again a prominent reason for requesting removal of the implant [ ].

A Thai study in 88 asymptomatic, young, HIV-1-positive women immediately after delivery has similarly confirmed the good acceptability of Norplant in a developing country, despite the high incidence of irregular bleeding and some instances of headache and hair loss [ ].

In 10 718 women in China, who were given Norplant no serious or fatal reactions were reported [ ]. When pregnancies did occur, only 3.1 per 100 were ectopic, which is higher than for the general population in Beijing but much lower than in US studies of Norplant (presumably because of lower rates of pelvic inflammatory disease in China). The 5-year continuation rate was 72 per 100 acceptors. Hemoglobin concentrations increased during the first year of use and remained high, presumably because of a reduction in menstrual blood loss. Although early findings suggested a reduction in platelet counts, follow-up work showed that in this population these were already low on admission to the study. Insertion-related complications necessitated removal in only 21 users.

In a study in Greece the method was well accepted by adolescents, although the study was small (13 subjects) [ ]. No significant problems arose during the 24-month follow-up period.

On the other hand, Norplant has been criticized in some Western countries, primarily because of adverse reactions at the implantation site. An overview of adverse events reported to the FDA over 3 years, during which more than 700 000 implants were estimated to be in place, cited reports of 24 women hospitalized for infections at the insertion site and 14 who were either hospitalized or disabled because of difficulties associated with capsule removal [ ]. Fourteen women (two per 100 000) were reportedly hospitalized because of strokes. Three women developed thrombotic thrombocytopenic purpura and six developed thrombocytopenia. Finally, 39 Norplant users developed pseudotumor cerebri (benign intracranial hypertension), an incidence of 5.5 per 100 000; this condition is associated with obesity or recent weight gain, conditions that were present in most of the women for whom data were given. Although the rates of stroke and pseudotumor cerebri were slightly less than the expected rates in the general population of women of reproductive age, one has to take into account the well-known under-reporting of adverse events to agencies such as the FDA; the figures could therefore mean that Norplant users are actually at increased risk.

Reviewing the overall scene in Britain, Hannaford has pointed to the considerable difference in acceptance rates for Norplant in the industrialized world and developing countries and has stressed the fact that the adverse reaction incidence is, according to the best evidence, low [ ].

Rapid rejection of implantable progestogen-based contraceptives by some users, because of discomfort and local complications (for example because of breakage or migration of the device, or less than expert placement and removal), has attracted much attention in past reviews.