Implanon ® failure in an HIV-positive woman on antiretroviral therapy resulting in two ectopic pregnancies

CASE

A 34-year-old nulliparous woman tested positive for HIV-1 infection in June 2005 following her partner's diagnosis. Her nadir CD4 count was 290 cells/mm³ (19%) and she was commenced on an antiretroviral regimen of zidovudine, lamivudine and efavirenz in August 2005. At this time, a discussion took place regarding the potential teratogenicity of efavirenz and the importance of a reliable form of contraception. Both the patient and her partner stated that they had no current plans to start a family. Contraceptive advice was given and she decided to switch from the combined oral contraceptive pill to Implanon^®. This was inserted at a family planning clinic in November 2005 and she became amenorrhoeic soon after. The couple apparently continued to use condoms both as a second method of contraception and also to reduce the likelihood of her acquiring drug-resistant virus from her partner.

During subsequent review appointments at the HIV clinic, the issue of pregnancy was discussed and documented. She was aware of the need to inform us of any plans to conceive in order that her antiretroviral regimen could be amended. At her attendance in January 2008 she reported that her last menstrual period had been in November 2005 when the Implanon was inserted.

In April 2008, she attended the accident and emergency department with severe abdominal pain, increasing in severity over the course of a week. She denied any bleeding per vagina. Her last menstrual period was reportedly seven weeks previously. On examination, she had rebound tenderness in the left iliac fossa. Bimanual vaginal examination revealed exquisite tenderness in the right adnexa and posterior fornix. The contraceptive implant was palpated and noted to be in the correct position. Her serum beta human chorionic gonadotropin (HCG) was elevated at 34,832 IU/L. A transvaginal ultrasound scan showed no evidence of intrauterine pregnancy. Subsequent diagnostic laparoscopy revealed an unruptured ectopic pregnancy in the right fallopian tube and, as such, a right salpingectomy was carried out. She was discharged several days later with advice not to attempt pregnancy for several months.

Unfortunately, this patient's attendance at clinic remained erratic. The first time she attended for HIV review following the ectopic pregnancy was in January 2009. On enquiry, it appeared that she had not attended for replacement of the Implanon device in November 2008 despite resuming regular menses after her previous ectopic pregnancy. She reported her last menstrual period to have been in November 2008. Her partner still apparently always used condoms although she was unable to recall the last time they had sexual intercourse. Despite this, a urine pregnancy test obtained in clinic was positive and the patient was seen by obstetrics and gynaecology. Her serum HCG level was 29,000 IU/L. A transvaginal ultrasound scan revealed an empty uterus and subsequent laparoscopy confirmed an unruptured ectopic pregnancy in the left fallopian tube. She underwent left salpingectomy.

Throughout this period, the patient remained on antiretroviral therapy consisting of tenofovir, emtricitabine and efavirenz. She took no other medications, specifically those that would further induce hepatic enzymes. Levels of etonogestrel were not measured during the time the implant was in place. An assessment has subsequently been carried out by the fertility clinic and our patient is awaiting in vitro fertilization with intracytoplasmic sperm injection.

DISCUSSION

Implanon, approved by the US Food and Drug Administration (FDA) in July 2006 and first introduced in the UK in September 1999, is a single-rod contraceptive implant inserted subdermally in the upper arm. It contains 68 mg of etonogestrel, a synthetic progesterone gradually released into the bloodstream. ¹ It acts primarily by suppressing ovulation and also increases cervical mucus viscosity. ² Based on dose-finding studies it was concluded that 30 µg etonogestrel daily was required to inhibit ovulation. ^1,3 The release rate in vitro of the Implanon device is reportedly 60–70 µg daily in the initial few weeks after insertion, decreasing to 35–45 µg/day at the end of the first year, 30–40 µg/day at the end of the second year and 25–30 µg at the end of the third year. ⁴ The device, therefore, is thought to provide effective contraception for up to three years.

A cohort study of Implanon use in 106 women over 18 months reported no contraceptive failures; however, the discontinuation rate prior to three years was 69.8% (predominantly due to irregular bleeding). ⁵ A subsequent retrospective study three years after the introduction of Implanon in Australia showed a failure rate of 0.1% (218 pregnancies in 204,486 users) ⁶ ; 39% were reportedly due to incorrect insertion; 21% of pregnancies predated insertion; 8% were due to incorrect timing of insertion; 1% were due to device expulsion and 4% were due to hepatic enzyme-inducing medications (predominantly carbamazepine).

It has been widely reported that hepatic enzyme-inducing drugs, especially those affecting cytochrome P450, result in increased clearance of combined contraceptives and, to a lesser extent, progesterone-only contraceptives. Efavirenz is the only component of this patient's antiretroviral drug regimen that is known to be a hepatic enzyme inducer. It is metabolized by cytochrome P450 2B6 and 3A4, and strongly induces the activity of CYP3A4 during long-term administration. ⁷ Other drugs that are substrates of CYP3A4 and whose plasma concentrations are similarly affected include antituberculous agents and anticonvulsants.

There has been one other similar report of Implanon failure in a patient receiving antiretroviral therapy, which was also attributed to efavirenz. ⁸ In this other case, the patient's menstrual cycle resumed two years after implant insertion and soon after this she became pregnant. She subsequently had a ruptured ectopic pregnancy.

On the basis of its pharmacokinetics and reported contraceptive failure, it would seem that Implanon should not be recommended as an efficacious form of contraception for those on certain antiretroviral agents known to induce hepatic enzymes. In the unlikely event that no alternative is available, the patient should be counselled regarding the potential risk of pregnancy and advised to use barrier contraception in addition. It is important to consider replacing the device when the patient resumes regular menses (or after maximum of 2 years) or insert a second device, in order to maintain the concentration of progesterone at a level that will suppress ovulation. The advice, however, from the manufacturers of Implanon (Organon, Roseland, NJ, USA) is that an alternative method of contraception be used in these individuals.

CONCLUSION

This case highlights the importance of addressing the contraceptive needs of HIV-positive women and considering potential side-effects and interactions with other medications. Condoms should be strongly advocated as a form of contraception and also as a barrier method to reduce the risk of transmitting resistant virus. The use of an intrauterine device appears to be the preferred option in patients receiving antiretroviral medication. Implanon should be considered only in those patients on a regimen that does not contain agents known to induce hepatic enzymes.