Can initial vaginal bleeding patterns in etonogestrel implant users predict subsequent bleeding in the first 2 years of use?

Diana Mansour, Ian S. Fraser, Alison Edelman, Carolina S. Vieira, Andrew M. Kaunitz, Tjeerd Korver, Annpey Pong, Jianxin Lin, Arvind K. Shah, Michelle Fox, Hans Rekers, Mitchell D. Creinin

Research output: Contribution to journalArticlepeer-review

14 Scopus citations


Objectives: To evaluate if a simple method for characterizing vaginal bleeding patterns in etonogestrel contraceptive implant users can predict subsequent patterns and bleeding-related discontinuation over the first 2 years of use. Study Design: We reanalyzed phase 3 study bleeding data for non-breastfeeding participants from the United States, Europe, Russia and Chile during the first 2 years of implant use to characterize and correlate bleeding patterns. We used 90-day reference periods with period 1.1 starting at Day 29 and ending at Day 118. We dichotomized bleeding patterns as “favorable” (amenorrhea, infrequent bleeding and normal frequency bleeding without prolonged bleeding) or “unfavorable’ (prolonged and/or frequent bleeding) and tracked user groups based on these bleeding patterns in reference period 1.1 through Year 1 and from Year 1 through Year 2, respectively. Results: We evaluated data from 537 and 428 women with up to 1 and 2 years use, respectively. Of the 325 (60.5%) women with favorable bleeding in reference period 1.1, 275 (84.6%) reported favorable bleeding also in reference period 2, 197 (60.6%) reported favorable bleeding throughout Year 1, and favorable bleeding in 75–85% of reference periods in Year 2. Among 212 (39.5%) women with unfavorable bleeding in reference period 1.1, 118 (55.7%) continued with unfavorable bleeding in reference period 2, while about 40%–50% reported favorable patterns in RP 2, 3 and/or 4. Initial favorable bleeding resulted in lower discontinuation rates than initial unfavorable bleeding in years 1 (3.7% vs 12.7%, p≪.0001) and 2 (2.5% vs 16.5%, p≪.0001). Conclusion: Implant users with favorable bleeding in the first reference period are likely to continue with favorable bleeding over the next 2 years. Initial bleeding patterns predict overall continuation rates in years 1 and 2. Implications Statement When evaluating vaginal bleeding in any 90-day reference period over 2 years of etonogestrel implant use, approximately 80% of women with favorable and 40% with unfavorable bleeding patterns will have favorable bleeding in the next reference periods. These findings can facilitate counseling regarding bleeding for women using the etonogestrel implant.

Original languageEnglish (US)
Pages (from-to)264-268
Number of pages5
Issue number4
StatePublished - Oct 2019


  • Bleeding
  • Continuation
  • Etonogestrel
  • Implant

ASJC Scopus subject areas

  • Reproductive Medicine
  • Obstetrics and Gynecology


Dive into the research topics of 'Can initial vaginal bleeding patterns in etonogestrel implant users predict subsequent bleeding in the first 2 years of use?'. Together they form a unique fingerprint.

Cite this