Elsevier

Contraception

Volume 75, Issue 2, February 2007, Pages 112-118
Contraception

Original research article
Effectiveness of levonorgestrel emergency contraception given before or after ovulation — a pilot study

https://doi.org/10.1016/j.contraception.2006.08.015Get rights and content

Abstract

Background

Although widely used, the mechanisms of action of the levonorgestrel emergency contraceptive pill (LNG ECP) are still unclear. There are increasing data to indicate that LNG is particularly effective as an ECP by interrupting follicular development and ovulation. An important outstanding question is whether it has any effect on fertilization or implantation.

Method

Ninety-nine women participated; they were recruited at the time they presented with a request for emergency contraception. All women took LNG 1.5 mg in a single dose during the clinic consultation. A blood sample was taken immediately prior to ingestion of the ECP for estimation of serum LH, estradiol and progesterone levels to calculate the day of ovulation. The specimens were analyzed in a single batch. Based on these endocrine data, we estimated the timing of ovulation to be within a ±24-h period with an accuracy of around 80%. Women were followed up 4–6 weeks later to ascertain pregnancy status. The effectiveness of ECP when taken before and after ovulation was determined.

Results

Three women became pregnant despite taking the ECP (pregnancy rate, 3.0%). All three women who became pregnant had unprotected intercourse between Days −1 and 0 and took the ECP on Day +2, based on endocrine data. Day 0 was taken as ovulation day. Among 17 women who had intercourse in the fertile period of the cycle and took the ECP after ovulation occurred (on Days +1 to +2), we could have expected three or four pregnancies; three were observed. Among 34 women who had intercourse on Days −5 to −2 of the fertile period and took ECP before or on the day of ovulation, four pregnancies could have been expected, but none were observed. We found major discrepancies between women's self-report of stage of the cycle and the dating calculation based on endocrine data.

Conclusion

These data are supportive of the concept that the LNG ECP has little or no effect on postovulation events but is highly effective when taken before ovulation.

Introduction

Oral levonorgestrel (LNG) has been widely used as an emergency contraceptive pill (ECP) for many years [1]. Large studies have confirmed its high efficacy [2] if taken within 72 h of unprotected intercourse but show that it can still have some effect if taken as late as 120 h after unprotected intercourse. A single dose of 1.5 mg of LNG has the same efficacy as two 0.75-mg doses taken 12 h apart [2].

Although widely used, the mechanisms of action of the LNG ECP are still unclear [3]. It has been postulated that it may interfere with follicular growth and function, ovulation, migration and function of spermatozoa, fertilization, implantation and endometrial function [4]. There are increasing data to indicate that LNG is particularly effective as an ECP by interrupting follicular development and ovulation [5], [6], [7]. An important outstanding question is whether it has any effect on fertilization or implantation. Recent studies in rats and cebus monkeys have convincingly demonstrated the absence of effect of high-dose LNG on postfertilization events [8], [9].

Calculating the efficacy of emergency contraception (EC) is problematic as it is impossible to carry out a randomized placebo-controlled study. Efficacy is generally estimated by calculating the ratio of the number of observed pregnancies to the number of pregnancies expected [10]. The expected number of pregnancies is estimated by multiplying the number of treated women who had unprotected sexual intercourse on each day of the menstrual cycle by external estimates of the probability of conception resulting from unprotected sexual intercourse on that day of the cycle. The expected day of ovulation is presumed to be the 14th day before the next menses, although it has been shown that only a small percentage of women ovulate exactly 14 days before the onset of menses [11]. Hence, there are a number of assumptions that are required, which introduce considerable uncertainty into the calculations.

Reliance on women's self-report of the day of the menstrual cycle on which unprotected sexual intercourse occurred and assurance that this is the only episode of unprotected intercourse to have occurred during that cycle also make calculations of ECP efficacy unreliable [12], [13], [14]. Other unknown factors are the fertility of the couple and whether the cycle was ovulatory [15].

Most previous studies have related efficacy to the length of time between unprotected intercourse and intake of the ECP and have not taken into account the time relationship between intake of the ECP and ovulation. Based on animal studies, it seems likely that the timing of ECP intake in relation to ovulation may be important in both estimating efficacy and determining mode of action.

In this study, the serum levels of progesterone (P4), estradiol (E2) and LH were measured at the time of ECP ingestion in order to provide a more reliable estimate of the time at which unprotected intercourse occurred in relation to ovulation. This has also allowed more accurate estimations of the timing of unprotected intercourse and the pregnancy risk in relation to the timing of ECP intake.

Section snippets

Participants

One hundred ten women were recruited in six family planning clinics in NSW (61 in Newcastle and 16 in Chatswood, Hurstville and Penrith) and in Queensland (21 in Brisbane and 12 in Toowoomba) at the time they presented with a request for EC. Women who agreed to participate in the study were in good health, were of reproductive age (15–43 years old; mean, 23.4 years), had regular periods (menstrual cycle lengths between 21 and 35 days) and reported a single act of unprotected sexual intercourse

Results

One hundred ten women aged 15 to 43 years (mean, 21; SD, 7.6) were enrolled in this study, but 11 women were excluded due to insufficient blood samples for complete hormonal analysis. Therefore, 99 women were investigated.

Sixty women had not been pregnant previously, 14 had one pregnancy, 6 had two pregnancies, 6 had more than three pregnancies and 2 women did not report their reproductive history. Forty-nine women had used ECP previously. Twenty women had used the ECP once, 11 had used it

Discussion

A most important question today is whether the LNG ECP has an effect on postfertilization events. If the LNG ECP is ineffective once fertilization has occurred, it would become an acceptable contraceptive option for many people who consider that human life begins at fertilization.

A problem with efficacy studies of EC is pinpointing the exact time relationship of unprotected intercourse and ECP ingestion to the occurrence of ovulation. Timing is usually based on data reported by the women and

Conclusion

It is clear from this study that ovulation dating based on retrospective menstrual data is unreliable and does not correlate well with the endocrine data. Studies of EC efficacy based on endocrine data are expensive and, therefore, potentially limit the number of participants. However, to obtain more accurate data on both efficacy and mode of action of EC, more extensive studies of this nature using transvaginal ultrasound in conjunction with endocrine data are required in much larger numbers

Acknowledgments

The authors are grateful to the administrators and staff of the six clinics in NSW and Queensland and to Dr. Georgina Luscombe for expert statistical advice.

References (28)

  • J. Trussell et al.

    Effectiveness of the Yuzpe regimen of emergency contraception by cycle day of intercourse: implications for mechanism of action

    Contraception

    (2003)
  • G. Seregely et al.

    Postcoital contraception with 0.75 mg d-norgestrel (Postinor)

    Ther Hung

    (1981)
  • M.E. Ortiz et al.

    Post-coital administration of levonorgestrel does not interfere with post-fertilization events in the New-World monkey Cebus apella

    Hum Reprod

    (2004)
  • G.W. Dixon et al.

    Ethinyl estradiol and conjugated estrogens as postcoital contraceptives

    JAMA

    (1980)
  • Cited by (126)

    • Effect of levonorgestrel emergency contraception on implantation and fertility: A review

      2022, Contraception
      Citation Excerpt :

      The limited window of action for Levonorgestrel EC on ovulation was supported by another 4 studies of lower quality of evidence (Table 1) [23, 25−27] We identified two studies that reported on observed versus expected pregnancy rates after Levonorgestrel EC intake before or after ovulation [28,29]. The studies were of moderate quality.

    • Emergency contraception: Links between providers' counseling choices, prescribing behaviors, and sociopolitical context

      2019, Social Science and Medicine
      Citation Excerpt :

      Like many other contraceptives, EC pills work by delaying ovulation; no evidence suggests that they end a pregnancy or prevent implantation of a fertilized egg. In contrast to mifepristone (the abortion pill) which ends a pregnancy, both levonorgestrel and ulipristal acetate EC are no more effective than placebo in preventing pregnancy after ovulation has occurred (Brache et al., 2010; Li et al., 2016; Noé et al., 2010; Novikova et al., 2007). Although one form of EC is available over the counter (levonorgestrel), the most effective forms of EC require either a medical procedure (the copper IUD) or a prescription (ulipristal acetate pills).

    • Use of Short Acting Reversible Contraception in Adolescents: The Pill, Patch, Ring and Emergency Contraception

      2018, Current Problems in Pediatric and Adolescent Health Care
      Citation Excerpt :

      Overall, if taken within 24 h of unprotected sex 95% of pregnancies are prevented, but the rate decreases to 85% if taken between 25–48 h and to 58% if taken between 49–72 h.52 In a small pilot study of 99 women, Novikova et al. looked at the observed versus expected pregnancy rate when LNG-EC was taken prior to ovulation or on the day of ovulation or later in the cycle. They found that the when LNG-EC was taken prior to ovulation there were zero pregnancies and their model predicted four to five for women who had intercourse prior to ovulation, but when LNG-EC was taken on the day of or after ovulation, there were three pregnancies and their model predicted three to four for those women.53 In a larger study of 393 women, Noé et al. similarly found a lower than expected pregnancy rate when LNG-EC was taken before ovulation but also found a slightly decreased rate compared to expected after ovulation.54

    View all citing articles on Scopus

    Funding for this study came from The Family Planning Foundation, FPA Health, Sydney.

    View full text