By Rebecca H. Allen, MD, MPH

Associate Professor, Department of Obstetrics and Gynecology, Warren Alpert Medical School of Brown University, Women and Infants Hospital, Providence, RI

Dr. Allen reports she is a Nexplanon trainer for Merck and a Liletta trainer for Actavis, and she has served on advisory boards for Bayer and Pharmanest.

SYNOPSIS: In this randomized, controlled trial, provision of depot medroxyprogesterone acetate on the same day as mifepristone for medical abortion decreased efficacy a small amount and had no effect on repeat pregnancy at six months. Whether this is clinically relevant will depend on individual patient counseling and the woman’s desire for convenient contraceptive coverage.

SOURCE: Raymond EG, Weaver MA, Louie KS, et al. Effects of depot medroxyprogesterone acetate injection timing on medical abortion efficacy and repeat pregnancy: A randomized controlled trial. Obstet Gynecol 2016;128;739-745.

The authors of this multicenter, randomized, controlled trial compared the administration of depot medroxyprogesterone acetate (DMPA) on the same day of mifepristone for medical abortion to DMPA given after the medical abortion was confirmed to be complete. The medical abortion regimen consisted of 200 mg of mifepristone followed 24-48 hours later by 800 mcg of buccal misoprostol. Women were followed for seven months after enrollment and asked to complete a urine pregnancy test before the final contact. Women who were appropriate candidates for medical abortion at the site (gestational age 75 days) did not have recognized nonviable pregnancies, desired DMPA for contraception, and did not plan on using hormonal contraceptives before receiving DMPA were enrolled. The study was powered to detect a surgery rate (to complete the abortion) of no more than 5% higher in the immediate DMPA arm compared to the regular DMPA arm. At six months, evidence of non-pregnancy was defined as a negative pregnancy test at 197 days or later or use of sterilization, intrauterine device (IUD), contraceptive implant, or injectable method at 183 days. Otherwise, a participant who was followed for at least 183 days without a reported pregnancy was considered to have “no evidence of pregnancy.”

A total of 461 women were recruited and randomized, three-quarters of whom were from Mexico and the remainder from the United States. There were 225 women in the immediate DMPA group (five excluded for missing abortion outcomes) and 236 in the regular DMPA group (seven excluded for missing abortion outcomes and three who may have used hormonal contraception prior to DMPA). For the primary outcome, 14 (6.4%) women in the immediate DMPA group required surgery (seven for ongoing pregnancy and seven for other reasons) compared to 12 (5.3%) women in the regular DMPA group (one for ongoing pregnancy and 11 for other reasons) (difference 1.1%; 95% confidence interval, -2.8 to 4.9). The confidence interval approached 5%, and a sensitivity analysis determined that if just one of the five women with unknown abortion outcomes in the immediate DMPA group required surgery, then the upper limit of the 95% confidence interval would have crossed 5% and the investigators could not have concluded that giving DMPA immediately was noninferior to giving it at the regular time after medical abortion completion. In addition, there were more ongoing pregnancies (positive fetal heart beat) in the immediate DMPA arm (8, 3.6%) compared to the regular DMPA arm (2, 0.9%). There was no difference between the two groups in the proportion of women who had evidence of non-pregnancy through six months (93% immediate group vs. 93% regular group, P = 0.64). Of the reported pregnancies by six months, there were five (2.3%) in the immediate DMPA group and seven (3.2%) in the regular DMPA group (P = 0.64). Only half of the participants were still using DMPA at six months.

COMMENTARY

Medical abortion with mifepristone followed by misoprostol accounts for approximately 36% of all abortions before nine weeks’ gestation.1 In March 2016, the FDA updated the mifepristone label to reflect the evidence-based regimen that already was being used (200 mg mifepristone followed by 800 mcg buccal misoprostol) and also expanded medication abortion up to 70 days’ gestation. The success of this medical abortion regimen is 98% at 49 days, 97% at 50-56 days, 95% at 57-63 days, and 93% at 64-70 days’ gestation.2 The rate of surgical intervention for ongoing live pregnancies also is related to gestational age: 0.3% at 49 days, 0.8% at 50-56 days, 2% at 57-63 days, and 3.1% at 64-70 days’ gestation. Because of the small risk of ongoing pregnancy and the fact that misoprostol is associated with congenital anomalies, most facilities in the United States require follow-up for medical abortion in the form of an in-person ultrasound or serum B-HCG follow-up at a laboratory.

Initiating contraception following medical abortion is important for patients. Typically, the contraceptive pill, patch, and ring can be started the day after misoprostol use. Other methods that are administered in the clinic require that the patient return for DMPA injection, implant placement, or IUD insertion when the abortion is complete. The authors of this study attempted to determine whether immediate administration of DMPA would affect the efficacy of medical abortion, given that mifepristone is an anti-progestin, and whether it would decrease repeat pregnancy rates at six months. The authors previously had shown that immediate placement of the etonogestrel implant on the day of mifepristone had no effect on medical abortion efficacy.3 The results from this study, however, were not so clear cut. Based on the data, there is a possibility that immediate DMPA decreases medical abortion efficacy slightly. This may be because DMPA is a higher dose of progestin than the etonogestrel implant. Overall, however, the authors reported that the success rate was still more than 93% for women in the immediate DMPA arm. In addition, women preferred receiving the DMPA on the same day as the mifepristone rather than returning to the clinic.

I wish the authors would have reported outcomes by gestational age. About 47% of the subjects were 49 days’ gestation, 40% were between 50-63 days, and 13% were 64 days (up to 75 days). It may be that they believed they didn’t have enough subjects to report on the group that was 64 days, but this would be important information for counseling patients regarding this option. For example, if failure rates were much higher at greater gestational ages with immediate DMPA, then the patient may opt to defer DMPA until the follow-up visit. It is also interesting that the maximum gestational age was 75 days when the usual cutoff is 70 days.

When comparing the contraceptive continuation rates of this study to their etonogestrel implant study, the authors noted that only 50% of women were still using DMPA at six months. This was much lower than the proportion who continued the etonogestrel implant (> 91%) at six months. In this respect, which method the woman chose was more important than the timing of administration. Facilitating access to contraceptive implants and IUDs should be a priority anytime a woman undergoes an abortion.

Editors note: The bottom line from this study is that DMPA reduced the success of medical abortion with a clinically important real increase in ongoing pregnancies. We should not be mislead by the confusion that this increase did not cross the predefined non-inferiority margin. The coadministration of a massive dose of DMPA, a progesterone-receptor agonist, with mifepristone, a progesterone-receptor antagonist makes no pharmacological sense. The further evidence that neither DMPA initiation or continuation was increased by early administration begs the question of “Why take this additional risk”? Although the same day use of the low-dose etonogestrel implant makes sense and seems to work, same day DMPA is a bad idea.

REFERENCES

  1. Jones RK, Jerman J. Abortion incidence and service availability in the United States, 2011. Perspect Sex Reprod Health 2014;46:3-14.
  2. Mifeprex prescribing information. Available at: http://www.earlyoptionpill.com/wp-content/uploads/2016/03/MIFEPREX-Labeling-and-MG-FINAL_March2016.pdf. Accessed Sept. 27, 2016.
  3. Raymond EG, Weaver MA, Tan YL, et al. Effect of immediate compared with delayed insertion of etonogestrel implants on medical abortion efficacy and repeat pregnancy: A randomized controlled trial. Obstet Gynecol 2016;127:306-312.