Lactogenesis and breastfeeding after immediate vs delayed birth-hospitalization insertion of etonogestrel contraceptive implant: a noninferiority trial.

Background: Initiating a progestin-based contraceptive before the drop in progesterone required to start lactogenesis stage II could theoretically affect lactation. Previous studies have shown that initiating progestin-based contraception in the postnatal period before birth-hospitalization discharge has no detrimental effects on breastfeeding initiation or continuation compared with outpatient interval initiation. However, there are currently no breastfeeding data on the impact of initiating the etonogestrel contraceptive implant in the early postnatal period immediately in the delivery room.

Objective: This study examined the effect of delivery room vs delayed birth-hospitalization contraceptive etonogestrel implant insertion on breastfeeding outcomes.

Study Design: This was a noninferiority randomized controlled trial to determine if time to lactogenesis stage II (initiation of copious milk secretion) differs by timing of etonogestrel implant insertion during the birth-hospitalization. We randomly assigned pregnant people to insertion at 0 to 2 hours (delivery room) vs 24 to 48 hours (delayed) postdelivery. Participants intended to breastfeed, desired a contraceptive implant for postpartum contraception, were fluent in English or Spanish, and had no allergy or contraindication to the etonogestrel implant. We collected demographic information and breastfeeding intentions at enrollment. Onset of lactogenesis stage II was assessed daily using a validated tool. The noninferiority margin for the mean difference in time to lactogenesis stage II was defined as 12 hours in a per-protocol analysis. Additional electronic surveys collected data on breastfeeding and contraceptive continuation at 2 and 4 weeks, and 3, 6, and 12 months.

Results: We enrolled and randomized 95 participants; 77 participants were included in the modified intention-to-treat analysis (n=38 in the delivery room group and n=39 in the delayed group) after excluding 18 because of withdrawing consent, changing contraceptive or breastfeeding plans, or failing to provide primary outcome data. A total of 69 participants were included in the as-treated analysis (n=35 delivery room, n=34 delayed); 8 participants who received the etonogestrel implant outside the protocol windows were excluded, and 2 participants from the delivery room group received the etonogestrel implant at 24 to 48 hours and were analyzed with the delayed group. Participants were similar between groups in age, gestational age, and previous breastfeeding experience. Delivery room insertion was noninferior to delayed birth-hospitalization insertion in time to lactogenesis stage II (delivery room [mean±standard deviation], 65±25 hours; delayed, 73±61 hours; mean difference, -9 hours; 95% confidence interval, -27 to 10). Onset of lactogenesis stage II by postpartum day 3 was not significantly different between the groups. Lactation failure occurred in 5.5% (n=2) participants in the delayed group. Ongoing breastfeeding rates did not differ between the groups, with decreasing rates of any/exclusive breastfeeding over the first postpartum year. Most people continued to use the implant at 12 months, which did not differ by group.

Conclusion: Delivery room insertion of the contraceptive etonogestrel implant does not delay the onset of lactogenesis when compared with initiation later in the birth-hospitalization and therefore should be offered routinely as part of person-centered postpartum contraceptive counseling, regardless of breastfeeding intentions.