Immediate versus delayed insertion of an etonogestrel releasing implant at medical abortion-a randomized controlled equivalence trial

Study question: Does a progestin releasing subdermal contraceptive implant affect the efficacy of medical abortion if inserted at the same visit as the progesterone receptor modulator, mifepristone, at medical abortion?

Summary answer: A etonogestrel releasing subdermal implant inserted on the day of mifepristone did not impair the efficacy of the medical abortion compared with routine insertion at 2-4 weeks after the abortion.

What is already known: The etonogestrel releasing subdermal implant is one of the most effective long acting reversible contraceptive methods. The effect of timing of placement on the efficacy of mifepristone and impact on prevention of subsequent unintended pregnancy is not known.

Study design size, duration: This multicentre, randomized controlled, equivalence trial with recruitment between 13 October 2013 and 17 October 2015 included a total of 551 women with pregnancies below 64 days gestation opting for the etonogestrel releasing subdermal implant as postabortion contraception. Women were randomized to either insertion at 1 hour after mifepristone intake (immediate) or at follow-up 2-4 weeks later (delayed insertion). An equivalence design was used due to advantages for women such as fewer visits to the clinic with immediate insertion. The primary outcome was the percentage of women with complete abortion not requiring surgical intervention within 1 month. Secondary outcomes included insertion rates, pregnancy and repeat abortion rates during 6 months follow-up. Analysis was per protocol and by intention to treat.

Participants/materials, setting, methods: Women aged 18 years and older who had requested medical termination of a pregnancy up to 63 days of gestation and opted for an etonogestrel releasing contraceptive implant were recruited in outpatient family planning clinics in six hospitals in Sweden and Scotland.

Main results and the role of chance: Efficacy of medical abortion was 259/275 (94.2%) in the immediate insertion group and 239/249 (96%) in the routine insertion group with a risk difference of 1.8% (95% CI -0.4 to 4.1%), which was within the ±5% margin of equivalence. The insertion rate was 275/277 (98.9%) in the immediate group compared to 187/261 (71.6%) women in the routine group (P < 0.001). At 6 months of follow-up significantly fewer women in the immediate group had become pregnant again (2/277, 0.8%) compared to the routine group (10/261, 3.8%) P = 0.018.

Limitations, reasons for caution: For the main outcome loss to follow-up data was minimized through access to patient records. Efforts were made to reduce loss to follow-up also for secondary outcomes. The results of the sensitivity analysis did not differ from the intention to treat or per protocol analysis.

Wider implications of the findings: Guidelines on postabortion contraception should be amended to include insertion of the etonogestrel releasing implant at the time of mifepristone intake for medical abortion up to and including a gestation of 63 days.

Study funding/competing interests: This study was funded by the Swedish Research Council (2012-2844), Stockholm City County and Karolinska Institutet (ALF). The contraceptive implants were provided by Merck and supplied by MSD Sweden. HKK and KGD have received honorariums for giving lectures for MSD/Merck and have participated in the national (HKK and KGD) and international (KGD) medical advisory boards for MSD/Merck. The other authors have nothing to declare.

Trial registration number: ClinicalTrials number NCT01920022.

Trial registration date: 06 August 2013.

Date of first patient's enrolment: 13 October 2013.