Background

After a single act of unprotected intercourse, the risk of pregnancy can be as high as 30% depending on (among other factors) when in the menstrual cycle intercourse occurs.3 The copper intra-uterine contraceptive device can help prevent such pregnancy and has the highest efficacy of any emergency contraceptive currently available (estimated pregnancy rate 0.1-0.15%).2^,3 It can be used up to 5 days after unprotected intercourse, but is less convenient to use than tablets. The most commonly used hormonal emergency contraceptive in the UK has been a combined contraceptive (Schering PC4) comprising four tablets of ethinylestradiol 50µg/norgestrel 500µg (equivalent to levonorgestrel 250µg). Its use is based on the Yuzpe regimen, which consists of two doses of ethinylestradiol 100µg/levonorgestrel 500µg taken 12 hours apart, the first within 72 hours of unprotected intercourse. This regimen reduces the risk of a resulting pregnancy by around 75%.4 Levonelle-2 consists of two doses of levonorgestrel 750µg, which are also taken 12 hours apart, the first within 72 hours of unprotected intercourse.

Pharmacology

Like other hormonal emergency contraceptives, levonorgestrel might work by disrupting ovulation, altering tubal motility, interfering with fertilisation or inhibiting implantation of the fertilised ovum.5 However, no specific mode of action has been defined for levonorgestrel in post-coital contraception.

Following oral administration, levonorgestrel is rapidly and almost completely absorbed, and its serum levels peak after 1.6 hours. Its mean overall elimination half-life is around 9-14.5 hours. Levonorgestrel is excreted as inactive metabolites, equal amounts of which appear in the urine and faeces.

Clinical effectiveness

Evidence on the efficacy of levonorgestrel alone for emergency contraception comes mainly from two randomised controlled trials, which compared use of the drug with the Yuzpe regimen.6^,7

The first trial, an unblinded study, involved 880 women (aged 18-45 years) who had requested post-coital contraception within 48 hours of unprotected intercourse.6 Participants received two doses of either levonorgestrel 750µg (one pill each time) or combined tablets containing ethinylestradiol 50µg and levonorgestrel 250µg (two pills each time); the first dose was taken on admission to the study and the second, 12 hours later at home. Women were followed up at a clinic 3 and 6 weeks after treatment. The main outcome measure was treatment failure (i.e. pregnancy rate). In all, 16 women in the Yuzpe group and 30 in the levonorgestrel group were lost to follow-up, and they were excluded from the efficacy analysis. Altogether, in the 834 women remaining, 15 (3.5%) pregnancies occurred in the Yuzpe group and 12 (2.9%) in the levonorgestrel group. When the 156 women (about 19%) who reported having intercourse after treatment were also excluded, the pregnancy rate was similar for those given levonorgestrel alone (2.4%) and those given the Yuzpe regimen (2.7%).

The second trial had a more robust design than the first. It included double-blinding and an estimate of its power to compare the efficacy of the trial treatments. The study randomised 1998 women (mean age 27 years) who had had unprotected intercourse no more than 72 hours before entering the study.7 Each woman received two doses (the first taken, within 72 hours of unprotected intercourse, under supervision at the clinic and the second dose at home 12 hours later), with both doses comprising either: one levonorgestrel 750µg tablet plus one placebo tablet; or two ethinylestradiol 50µg/levonorgestrel 250µg tablets. The primary outcome measure was pregnancy (analysed as the absolute rate and relative risk of pregnancy). By follow-up, about 1 week after the expected onset of the next menses, 18 women from the Yuzpe group and 25 from the levonorgestrel group had dropped out of the study. Among those who remained, the pregnancy rate was lower in the levonorgestrel group than in the Yuzpe group (1.1% vs. 3.2%, relative risk 0.36; 95% CI 0.18-0.70). The study did not report a standard intention-to-treat analysis that assumed all the women lost to follow-up had become pregnant (admittedly, an unlikely scenario). The longer the delay in starting the emergency contraception after intercourse, the lower the efficacy of both trial treatments.7^,8

A meta-analysis9 of the two published randomised controlled trials6^,7 concluded that levonorgestrel alone appears to be more effective than the Yuzpe regimen in preventing pregnancy (relative risk 0.51; 95% CI 0.31-0.84). However, according to the reviewers10 of this meta-analysis, the women lost to follow-up in the two studies were, mistakenly, not taken into account in calculating the relative efficacy of the two treatments; we have been informed by the World Health Organization10 that the results will be re-analysed.

Unwanted effects

In the larger randomised controlled trial comparing the effectiveness of hormonal emergency contraceptives, unwanted effects (assessed as secondary outcome measures) that were less common with the levonorgestrel than with the Yuzpe regimen included: nausea (23% vs. 50% of women); vomiting (6% vs. 19%); dizziness (11% vs. 17%); and fatigue (17% vs. 29%).7 Unwanted effects similarly frequent with both regimens included: headache (17% in levonorgestrel group vs. 20% in Yuzpe group); breast tenderness (11% vs. 12%); and lower abdominal pain (18% vs. 21%).7 Other, less common, unwanted effects with both regimens consisted mostly of diarrhoea or irregular bleeding/spotting.7 The levonorgestrel regimen was better tolerated than the Yuzpe regimen in the smaller randomised controlled trial.6

Dosage and precautions

The manufacturer of Levonelle-2 recommends that a woman should take one 750µg tablet as soon as possible (and not later than 72 hours) after unprotected intercourse, followed by another tablet 12 hours later. The woman should also be advised that she should return for additional pills if she vomits within 2 hours of taking either dose. After a woman has been given an emergency contraceptive, future contraception needs to be discussed and a follow-up appointment should be offered to assess whether the treatment has been effective.

Progestogen-only pills for routine contraception are contraindicated in some conditions such as unexplained vaginal bleeding and porphyria. However, according to the SPC for Levonelle-2, it is not known whether these contraindications also apply to the use of levonorgestrel alone in emergency contraception. Women with acute porphyria should be made aware that there is an unquantified, but probably low, risk that use of Levonelle-2 may provoke an acute attack. Other conditions, which the manufacturer states as being possible relative contraindications to the use of Levonelle-2, include severe hypertension, diabetes mellitus with associated complications, ischaemic heart disease, stroke or a history of breast cancer. However, the manufacturer also suggests that the risks associated with use of Levonelle-2 in such situations are almost certainly smaller than those associated with pregnancy. Small amounts of levonorgestrel are excreted into breast milk in women who take Levonelle-2 but no adverse effects on milk production or the feeding infant have been documented. According to the manufacturer, Levonelle-2 is not recommended in children and only limited data are available on its use in young women of childbearing potential aged 14 years and over. Nevertheless, it seems plausible that Levonelle-2 could be an effective emergency contraception for girls younger than 14 years who are at risk of pregnancy, so more studies are needed in this population.

Interactions

Drugs that induce hepatic enzymes can accelerate the metabolism, and so reduce the efficacy, of oral contraceptives. These enzyme-inducers include barbiturates, phenytoin, carbamazepine, ritonavir, St John's wort and, particularly, the rifamycins. Such drugs might also reduce the efficacy of hormonal emergency contraceptives when these are taken concurrently; therefore, the dose of hormonal emergency contraceptives should probably be increased by 50% in women taking enzyme-inducing drugs (i.e. in the case of Levonelle-2, the first dose should be doubled to 1.5mg, followed 12 hours later by the usual dose of 750µg). More data are needed to clarify this position.

Getting emergency contraception

Currently, both Schering PC4 and Levonelle-2 are only available as prescription-only medicines (POMs) and supplied through GPs, family planning clinics, sexual health clinics, NHS walk-in clinics and some hospital accident and emergency departments. However, these centres only prescribe hormonal emergency contraceptive pills if the woman presents after, and within 72 hours of, having unprotected intercourse. Some pilot schemes have been established to allow pharmacists to supply hormonal emergency contraceptives under Patient Group Directions. The British Pregnancy Advisory Service (BPAS) is promoting the fact that hormonal emergency contraceptives can be obtained from them "before the emergency" and kept just in case. This service costs about £15 and includes the pills and a consultation with a doctor. Reclassification of Levonelle-2 from a POM to a pharmacy medicine (i.e. for sale from pharmacies under the supervision of the pharmacist) is being considered by the licensing authority, and the outcome is awaited. Providing pharmacists have received suitable training, and the cost of the product is not a disincentive to purchase, the sale of Levonelle-2 through pharmacies would offer an important advance in the provision of emergency contraception.

The cost to the NHS of a course of Levonelle-2 (around £5) is over three times that of Schering PC4 (around £1.60). Schering, who manufacture both drugs, have suggested that Levonelle-2 will eventually replace Schering PC4.11 In light of this, we hope the cost of Levonelle-2 will be reduced.

Conclusion

▼Levonelle-2 (two tablets of levonorgestrel 750µg) is the first oral progestogen-only emergency contraceptive to become available in the UK. It is licensed for use within 72 hours of unprotected intercourse. The levonorgestrel regimen is at least as effective as oral combined hormonal emergency contraception and less likely to cause nausea and vomiting. These findings make Levonelle-2 the hormonal emergency contraceptive of first choice. However, it is more expensive than the longer-established, combined hormonal emergency contraceptive (Schering PC4). We believe the cost of Levonelle-2 should be reduced. The move to make Levonelle-2 available over the counter as a pharmacy medicine is welcome.