Contraception 81 (2010) 414 – 420

Original research article

Contraceptive efficacy of emergency contraception with levonorgestrel

given before or after ovulation☆
Gabriela Noé⁎, Horacio B. Croxatto1 , Ana María Salvatierra, Verónica Reyes, Claudio Villarroel,
Carla Muñoz, Gabriela Morales, Anita Retamales
Instituto Chileno de Medicina Reproductiva (ICMER), 8320152, Santiago, Chile
Received 9 November 2009; revised 10 December 2009; accepted 16 December 2009

Abstract

Background: We aimed to evaluate whether emergency contraception with levonorgestrel (LNG-EC) administered after ovulation is equally
effective to LNG-EC administered before ovulation.
Study design: We studied a cohort of women attending a family planning clinic for EC. From interview, we recorded menstrual history, time
of intercourse and of intake of LNG-EC. On the day of intake of LNG-EC and during 5 days' follow-up, blood samples were taken for
examination of luteinizing hormone, estradiol and progesterone concentrations, and vaginal ultrasound examinations were done for size of
the leading follicle and/or corpus luteum. Thereafter women were not contacted until next menses or pregnancy occurred.
Results: Of 388 women attending for LNG-EC, 122 women had intercourse on fertile cycle days according to ultrasound and endocrine
findings. At the time of LNG-EC intake, 87 women were in Days −5 to −1 and 35 women were in Day 0 (day of ovulation) or beyond. With
the use of the probability of clinical pregnancy reported by Wilcox et al. [N Engl J Med 333 (1995) 1517–1521], expected numbers of
pregnancies among the 87 and 35 women were 13 and 7, respectively, while 0 and 6 pregnancies, respectively, occurred.
Conclusion: We conclude that LNG-EC prevents pregnancy only when taken before fertilization of the ovum has occurred.
© 2010 Elsevier Inc. All rights reserved.

Keywords: Emergency contraception; Levonorgestrel; Contraceptive efficacy

1. Introduction contraception (LNG-EC) given in the follicular phase to

Research [1] on how levonorgestrel emergency contra-
ceptive pills (LNG-ECPs) act to prevent pregnancy has
studied its effects on the ovulatory process, on sperm
physiology in the female genital tract and on endometrial
receptivity to embryo implantation. It is established that
preovulatory administration of LNG interferes with the
ovulatory process [2–5]. A previous study at our clinic
assessed the interference with ovulation of LNG emergency
☆
This study has been supported by grants 1060252 and FB0807 from
Fondo Nacional de Desarrollo Científico y Tecnológico (FONDECYT) and
by grant A 65559 from the World Health Organization, Department of
Reproductive Health and Research (WHO/RHR).
⁎ Corresponding author. Tel.: +56 2 6649393; fax: +56 2 6336204.
E-mail address: gnoe@icmer.org (G. Noé).
1
Current affiliation. Laboratorio de Inmunologia de la Reproduccion,
Universidad de Santiago de Chile y Centro para el Desarrollo de
Nanociencia y Nanotecnologia (CEDENNA).
0010-7824/$ – see front matter © 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.contraception.2009.12.015
women protected from pregnancy by tubal ligation or
nonhormonal IUD. The results showed this interference
depends on how advanced the follicular phase is when
LNG-EC is given and, specifically, whether treatment takes
place before or after the onset of the luteinizing hormone
(LH) surge [6].
Other studies report that LNG could interfere in vivo with
the migration of spermatozoa by altering cervical mucus and
uterine environment [7–9]. This effect could be significant
when coitus precedes ovulation for several hours or days but
may have less impact when coitus occurs close to ovulation.
Direct effects of LNG on spermatozoa in vitro are of dubious
relevance since effects are only observed with high dose
levels [10,11]. Studies on the impact of LNG-EC on
endometrial parameters involved in endometrial receptivity
are not consistent, and current knowledge on cellular and
molecular markers of endometrial receptivity in the human is
insufficient to resolve this controversy [3,12–14]. A recent
study of mifepristone and LNG on embryo attachment to
 G. Noé et al. / Contraception 81 (2010) 414–420
endometrial cells in an in vitro endometrial cell culture
reported that LNG did not interfere with embryo attachment
to endometrial cells, while mifepristone did so [15].
Ethical and logistical constraints prevent performing
experiments in humans that directly address what happens
to the embryo when women receive LNG-EC. Studies in rats
and capuchin monkeys show that LNG given after
fertilization does not prevent pregnancy and that, when
given before ovulation, it interferes with the ovulatory
process as it does in women [16,17]. Based on these
observations, we advanced the hypothesis that LNG-EC
contraceptive failure results from taking LNG-EC when it
can no longer prevent fertilization. In line with this
hypothesis, a study carried out in Australia assessed the
contraceptive effectiveness of LNG-EC given before or after
ovulation. The day of ovulation and day of intercourse were
determined by reproductive hormone levels measured in a
blood sample drawn immediately before taking LNG-EC.
Results of the study indicate that when LNG-EC was given
before ovulation, it was effective in preventing pregnancy,
while it had no contraceptive effect when given after
ovulation, although the findings were based on small
numbers [18]. The current study aims to accrue a sufficient
number of women in whom unprotected intercourse
occurred during the fertile days of the cycle so as to
determine with better statistical power whether or not LNG-
EC can prevent pregnancy when it is administered before
and after ovulation. Here we report the preliminary results of
this ongoing study.
2. Material and methods
2.1. Emergency contraceptive pills
The product used was TACE® produced by Laboratorio
Recalcine SA, Santiago, Chile, consisting of two tablets
containing 0.75 mg of LNG each.
2.2. Subjects
Volunteers were admitted from among women requesting
ECP at the family planning clinic of ICMER.
Inclusion criteria were women 18 to 38 years old, being
within 120 h of an unprotected sexual intercourse, having
regular menstrual cycles of 21 to 35 days and being between
Day 5 and Day 25 of the current cycle as calculated from Day
1 of the last menstrual period (LMP). Exclusion criteria
either from the study or from data analysis were use of
hormonal contraceptives or an intrauterine device during the
current cycles, having a positive pregnancy test, breastfeed-
ing and/or having unprotected intercourse subsequent to
LNG-EC intake in the studied cycle. In addition, women
who had incomplete follow-up or missing data not enabling
determination of day of cycle were excluded from the
analyses of contraceptive efficacy.
415
2.3. Study design
This was a prospective, open-label, single-drug, single-
dose, uncontrolled clinical trial. It was approved by the ethics
committee of the Chilean National Health Service. After
counselling on the use of ECP by a midwife or physician,
each woman received a single doses of 1.5 mg of LNG;
thereafter the woman was informed about the study and
invited to participate. All participants received detailed
information about the study and signed an informed consent.
Women who accepted participation and met the admission
criteria were examined the same day as treatment was given
to determine on which day of the cycle according to
biological indices they received the LNG-EC.
2.4. Data collected on the day of LNG-EC intake
To determine the stage of the cycle, the following
sequence of procedures was performed: (1) A blood sample
was drawn within 1 h of taking the LNG-EC, for LH,
progesterone (P) and estradiol (E2) concentrations. If
pregnancy was suspected, chorionic gonadotropin (hCG)
was measured in this blood sample. (2) A sample of vaginal
fluid secretion and/or cervical mucus was obtained to
examine the presence of spermatozoa. (3) Transvaginal
ultrasonography (TVU) was performed to determine the
presence and size of a leading follicle and/or a corpus luteum
(CL), the characteristics and thickness of the endometrium
and the presence of fluid in the Douglas pouch. (4) Data of
the unprotected intercourse and current cycle were recorded:
day and hour of intercourse; specifics of unprotected
intercourse (e.g., lack of contraceptive, ruptured condom);
use of vaginal washing; the first day of menstrual bleeding of
the current cycle; and length of the preceding menstrual
cycles. A means of contact (phone, e-mail) for the follow-up
was determined as per convenience to the woman.
2.5. Follow-up
Based upon the first TVU evaluation, women were
allocated to one of three groups with different follow-up
schemes. Group I: women with follicles ≤12 mm who were
within the first 10 days following the onset of menstruation;
these women were considered to have had intercourse
outside the fertile period and no immediate follow-up
procedures were done. Group II: women with a dominant
follicle of N12 mm and proliferative endometrium; in these
women follicular growth and eventual rupture and LH, E2
and P concentrations in serum were assessed daily for a
maximum of 5 days after treatment or until follicular rupture
(FR), whichever came first. When possible, after detection of
FR, one or two additional blood samples were taken between
Days 7 and 11 after FR to determine luteal phase P. Group
III: women either with CL or follicles of b12 mm who were
beyond 10 days of last menstruation and exhibited secretory-
type endometrium on ultrasound examination. This group
was provisionally considered to be in the postovulatory
416 G. Noé et al. / Contraception 81 (2010) 414–420
phase. The concentration of LH, E2 and P was determined
every other day for up to 10 days starting on the day after
ECP intake.
Volunteers of all groups were contacted at the presumed
date of their next menstruation to confirm the occurrence of
menstruation. If it did not occur at that time, they were
scheduled for a urinary hCG test on a weekly basis until
confirmation of pregnancy or menstruation occurred.
2.6. Transvaginal ultrasonography
TVU was performed with a real-time scanner, Aloka SS d
620 ultrasound system, and a 5.0-MHz vaginal transducer.
Data obtained from TVU were as follows: (1) Assessment of
endometrium thickness and ultrasonographic characteristics.
Endometrium was classified as follows: Early proliferative
phase: the endometrium is isoechoic to myometrium and
with an endometrial thickness of 4–8 mm. Late proliferative
phase ( preovulatory phase): the endometrium is slightly
more hypechoic centrally surrounded by two hypoechoic
layers (three-layer aspect) and measures 6–10 mm. Secretory
phase: the endometrium is homogeneously echoic and
measures 7–14 mm [19,20]. (2) The measurement of
follicular diameter calculated as the average of the two
perpendicular maximum diameters, measured in the same
plane. (3) The presence of a CL visualized by color Doppler.
(4) The existence of fluid in the Douglas pouch.
2.7. Hormone determinations
Serum LH and hCG were measured by enzyme
immunometric assays provided by Immunometrics UK
(London, UK). Intra- and interassay coefficients of variation
were 2.5% and 7% for LH and 4.2% and 10% for hCG,
respectively. Estradiol and P were measured by radioim-
munoassay using Coat-A-Count® Estradiol and Coat-A-
Count Progesterone from Siemens Medical Solutions
Diagnostics Products (Los Angeles, CA, USA). Intra- and
inter-assay coefficients of variation were 2% and 6.5% for
E2 and 2.7% and 6% for P, respectively.
2.8. Cervical mucus and vaginal fluid examination
Cervical mucus and vaginal fluid were obtained with a
1-mL syringe; the cervical mucus was examined and
characterized according to the WHO manual [21]. Smears
of mucus and vaginal fluid were examined using a
microscope at 400 HPF (40×) to search for spermatozoa.
2.9. Day of unprotected intercourse and day of treatment in
relation to ovulation
For each participating woman, the day of unprotected
intercourse and the day of intake of LNG-EC in relation to
ovulation were calculated retrospectively. We used a chart
developed by our center that is based on data on serial LH, P,
E2 determinations and serial TVU for measuring follicular
size, FR and CL from more than 100 normal ovulatory cycles
(Noé, G., 2009, personal communication). We plotted the
LH, P, E2 and ultrasonographic data for each woman in the
current study on the chart. This enabled us to determine the
day of the reported unprotected intercourse and the day of
LNG-EC treatment in relation to the day of ovulation for
each of the women in the current study. The aforementioned
chart is based on data of cycles in women coming from the
same population as women participating in the current study.
2.10. Statistical analysis
The characteristics of the women and the time period
between intercourse and intake of LNG-EC were examined
in the 388 women enrolled in the study. The efficacy of
LNG-EC in relation to intake before and after the day of
ovulation was analyzed in 337 women whose data were
sufficient to determine whether the intercourse occurred in
the 5 days preceding ovulation or on the day of ovulation,
and the outcome of the treatment in terms of menstruation
or pregnancy.
After the classification of the risk of pregnancy
according to the cycle day on which unprotected intercourse
occurred using TVU and hormonal indices described above,
women having had intercourse during the fertile days were
divided into two subgroups according to whether LNG-EC
administration took place before or after ovulation.
Expected pregnancies for the entire group and each
subgroup were calculated by multiplying the number of
women having unprotected intercourse on each day of the
fertile period by the probability of clinical pregnancy as
reported by Wilcox et al. [22]. The percentage of prevented
pregnancies was calculated by subtracting from 1 the ratio
of expected to observed pregnancies [1−(observed/
expected)⁎100]. Fisher's Exact Test was used for the
analysis of the concordance of contraceptive efficacy of the
entire group as well as that of preovulatory and post-
ovulatory treatment subgroups. This calculation was done
by applying the biological estimation of the cycle stage as
described above and also according to the presumptive day
of ovulation, as done in previous efficacy studies of LNG-
EC that used LMP and the reported length of the preceding
cycle to estimate the day of ovulation [23–28]. The
GraphPad Prism 4 software was used for these analyses
(GraphPad Software, Inc., San Diego, CA, USA).
3. Results
3.1. Characteristics of women
Three hundred eighty-eight women were enrolled of
whom 48%, 35% and 17% were 18–21, 22–27 and 28–38
years old, respectively. As to the unprotected intercourse,
51% of the women reported no contraceptive use, and 43%
and 6% said they had problems with condom use and coitus
interruptus, respectively. The time elapsed from unprotected
intercourse to treatment was less than 24 h in 65% of
women, 24–47 h in 24% and between 48 and 72 h in 9% of
 G. Noé et al. / Contraception 81 (2010) 414–420
cases; only four women came later than 72 h. Spermatozoa
were detected in 46% of women asking for ECP within
24 h; the absence of spermatozoa was not considered an
exclusion criterion for analysis since it depends on several
factors that not always imply the absence of insemination.
In fact, in one woman who became pregnant, spermatozoa
were not detected after an interval examination after coitus
of 72 h.
3.2. Timing of intercourse and treatment relative
to ovulation
Of the 388 enrolled women, 337 had data from
interview, TVU and hormone measurements that enabled
determination of whether the reported day of unprotected
intercourse was during fertile or infertile days of the
menstrual cycle ovulation and thus the analysis of efficacy
of treatment with LNG-EC. For 51 women the data were
incomplete or uncertain for determining whether or not they
were at risk of pregnancy at the time of the unprotected
intercourse (Fig. 1, Table 1).
In 215 of the 337 women referred to above, the
unprotected intercourse occurred during infertile days of
the cycle and the women were not at risk of pregnancy, while
in 122 women the intercourse occurred in one of the six
fertile days of the menstrual cycle. These 6 days are the day
of ovulation and the five preceding days as demonstrated by
Wilcox et al. [22]. Eighty-seven of the 122 women with
intercourse during fertile days had treatment with LNG-EC
within Days −5 to −1, and 35 had treatment on Day 0 or
thereafter, Day 0 being the day of FR (ovulation).
Fig. 1. Flowchart: disposition of women enrolled and followed up by whether intercourse occurred on fertile or infertile days of the cycle and whether LNG-EC
intake was preovulatory or not.
417
Table 1
Number of women and reasons for exclusion from efficacy analysis
Reason Number of women
Loss to follow-up 15
Gynecologicala 13
Incomplete record 8
Woman uncertain of semen insemination 6
Additional sexual intercourse in the study cycle 5
Cycles longer than 35 days 4
a
Gynecological reason included use of hormonal contraceptives during
the current cycle, breastfeeding, having polycystic ovary or a follicle larger
than 30 mm in ultrasonography examination the day of LNG-EC intake.
3.3. Detection of FR and luteal phase in women at risk
of pregnancy
In the 87 women treated before ovulation, FR was
confirmed in 62 by means of TVU and elevated P level
(n=39), or TUV only (n=18), or by luteal phase values of P
only (n=5).
In 57 cases (66%) of the 87 preovulatory treated women,
FR was detected during the 5 days of follow-up after LNG-
EC intake; half of them had a follicle size N18 mm on the day
of treatment. In 15 women (17%), FR was not detected
during follow-up; in most of them, follicle size was b18 mm
on the day of treatment. In the other 15 (17%), FR could not
be determined because the women did not attend the follow-
up visits (Table 2).
Progesterone was measured in 82 of the 122 women at
risk of pregnancy. P was not measured in 40 women because
no blood sample was obtained after treatment or FR. Luteal
phase was defined by P levels of at least 12 nmol/L. Luteal
418 G. Noé et al. / Contraception 81 (2010) 414–420

Table 2
Number of cycles in which FRs and luteal P were detected during follow-up after administration of LNG-EC to women who had intercourse during the
fertile days
Time of treatment Preovulatory (n=87) Pre- and postovulatory (n=122)
Parameter Follicular rupture Progesterone concentration
Ultrasound finding at LNG-EC intake Detected Undetected Unknowna ≥12 nmol/L b12 nmol/L Unknowna
Follicle diameter, ≥18 30 4 9 23 3 19
mm ≥15 to 17.9 21 8 5 16 4 14
≥12 to 14.9 6 2 1 4 0 3
Unknownb 1 1
Corpus luteum _ _ _ 29 2 4
a
Unknown because there was no follow-up after examination on the day of LNG-EC intake and/or after FR.
b
Unknown case in which follicular diameter was not measured; had blood sample but not ultrasonography during follow-up.

phase was detected in 73 (89%) of the 82 women. Among 35
women with a recent CL on the day of treatment; 83% had P
values above 12 nmol/L (Table 2).
3.4. Contraceptive efficacy
For the 122 women who had intercourse during the fertile
days as based on TVU and hormone measurement, the
overall accumulated probability of pregnancies was 20.3,
while altogether six pregnancies were observed. Thus, the
overall contraceptive efficacy of the LNG-EC was 70%.
The pregnancies prevented by LNG-EC given before or
after ovulation were determined in the two subgroups:
among the 87 women who took LNG-EC before ovulation
(Days −1 and −5), 13.2 pregnancies could be expected and
no pregnancy occurred (pb.001) (Table 3). In the group of 33
women who took the LNG-EC on the day of ovulation (Day
0) or thereafter, six pregnancies occurred and 7.1 were
expected (p=1.00) (Table 3).
3.5. Estimates of LNG-EC contraceptive effectiveness using
menstrual cycle history and biological parameters
In large WHO trials, the efficacy of LNG-EC was
estimated according to the time since onset of last menses
and the reported length of previous cycles [23–28]. The
presumptive day of ovulation was calculated by adding the
mean duration of preceding menstrual cycles to the first day
of the last menstruation of the current cycle and subsequently
subtracting 14 days. We applied this calculation to estimate
the day of ovulation in the 337 women included in the
analyses and compared the results to those obtained when the
day of ovulation was estimated from hormonal and
ultrasonographic data (Table 4).
When the time of ovulation was calculated using previous
cycle length, more women were classified as having had
intercourse during the fertile period than when using
hormonal and ultrasonographic data for timing of ovulation.
This, in turn, resulted in a higher number of expected
pregnancies (26.2) and, consequently, in higher efficacy of
LNG-EC (77%) for the calendar method compared to
biological parameters for estimating fertile days (Table 4).
According to estimates using the calendar method, two of the
six observed pregnancies occurred in women who had
unprotected intercourse on infertile days.
With regard to the mechanism of pregnancy prevention
by LNG-EC, the main difference between the two methods
used to determine the ovulation day was observed in the
group of women estimated to have had LNG-EC treatment
on Day 0 (day of ovulation) and thereafter. When the
calendar method was used, the number of expected
pregnancies was significantly higher than the three
pregnancies observed for this period (Table 4). In contrast,
when hormonal and ultrasonographic parameters were used,
six pregnancies occurred among women assigned to
postovulatory treatment. This number of observed pregnan-
cies was not statistically different from the expected number
of pregnancies (Table 4).

Table 3
Pregnancies prevented by LNG-EC administered before (Days −5 to −1) or
after (Day 0 or beyond); Day 0=FR
4. Discussion
Time of LNG-EC Number of Pregnancies expected/
A previous study in our clinic showed that LNG-EC
administration women observed
interferes with the ovulatory process [5]. Since those
Preovulatory 87 13.2/0⁎
observations were made in previously sterilized women,
(Days −5 to −1)
Postovulatory 35 7.1/6⁎⁎ the findings were suggestive but not conclusive in terms of
(Day 0 or +) preventing pregnancy. For some people a potential post-
Fisher's Exact Test: ⁎pN.0002; ⁎⁎p=1.00.
fertilization effect of LNG-EC poses an ethical problem
Pregnancies expected were calculated according to the probabilities because it would compromise embryo viability and it could
estimated by Wilcox et al. [22] for each fertile day of the menstrual cycle. be claimed that LNG-EC acts as an abortifacient in these
 G. Noé et al. / Contraception 81 (2010) 414–420
Table 4
Comparison of the number (n) of women at risk of pregnancy and the contraceptive efficacy (expected/observed) according to the method used to estimate the
day of ovulation in 337 cases analyzed
Method for estimating Sexual intercourse Time of LNG-EC administration
ovulatory day
Without With Preovulatory (Days −5 to −1)
risk, n risk, n
n Pregnancies expected/observed
Hormones and ecography 215 122 87 13.2/0
of actual cycle
LMP and length of 168a 169 91 12.6/1⁎⁎
previous cycles
Italicized values are to emphasize the differences between methods.
Fisher's Exact Test: ⁎p=1.00; ⁎⁎p=.0012; ⁎⁎⁎p=.0085.
a
Two pregnancies occurred in this group.
situations. The present study evaluated whether or not
LNG-EC prevents pregnancy when it is given after
fertilization has occurred.
Of the six fertile days of the menstrual cycle, five of the
days are before ovulation and 1 day includes ovulation and
the immediate period thereafter. It is therefore not unex-
pected that the majority of women taking LNG-EC during
the fertile days of the cycle do so before ovulation. In our
study, of the 122 women taking LNG-EC during the fertile
cycle days, 87 (71%) did so before ovulation. In 62 (71%) of
these 87 women, ovulation was documented by ultrasono-
graphic evidence, FR and/or elevated P concentration.
Despite the evidence of ovulation in these women who
took LNG-EC before ovulation occurred, no pregnancies
occurred among them. This finding dissuades our doubts
with respect to the lack of contraceptive effectiveness of
LNG once the LH surge has been initiated [6].
On the other hand, among the 35 (27%) women of the 122
women who had unprotected intercourse during fertile cycle
days, and who took LNG-EC on the day of ovulation or
immediately thereafter, the number of observed pregnancies
was very similar to the number that was expected; the
difference was not statistically significant. This finding is
compatible with results of studies in rat and monkey [15,16]
which found that LNG-EC does not prevent pregnancy after
fertilization of the oocyte. Furthermore, the contraceptive
effectiveness of LNG-EC given before or after ovulation is
almost identical to that reported by Novikova et al. [18] in a
similar study carried out in Australia. Among 99 women
enrolled in that study, 34 had intercourse during fertile days
and were treated before or on the day of ovulation; four
pregnancies were expected and none was recorded. Seven-
teen had intercourse during fertile days and were treated 1 or
2 days after ovulation; three or four pregnancies were
expected and three became pregnant.
The effectiveness of LNG-EC estimated by means of the
biological parameters of the actual cycle gave results that
differed from those calculated from LMP and historical
menstrual data. Similar discrepancies have been described
previously [29–31]; Espinós et al. [29] estimated in a group
of 303 women that 99 had had unprotected intercourse
between Days −5 and +1 in relation to the day of ovulation
419
Efficacy, %
Postovulatory (Day 0 or +)
n Pregnancies expected/observed
35 7.1/6⁎ 70
78 13.6/3⁎⁎⁎ 77
according to the calendar method as reported by the
women, while according to blood levels of LH, E2 and P,
51 women had intercourse during Days −5 and +1. Espinós
et al. [29] concluded that the hormonal evaluations more
accurately estimated the fertile days of the cycle and the
probability of pregnancy. Similarly, Stirling and Glasier
[30] reported that out of 64 women 19 were not certain
about the date of LMP and 21 women had urinary
pregnanediol concentrations that were inconsistent with
their cycle day. Severi et al. [31] evaluated the risk of
pregnancy after a single unprotected intercourse in one
cycle among 163 women according to anamnestic data of
the cycle and according to TVU findings of the endome-
trium and the ovaries. Altogether, seven pregnancies were
observed; according to the ultrasonographic data, all
pregnancies occurred after unprotected intercourse on the
fertile days of the cycle, while according to the menstrual
history data, only four of the seven pregnancies occurred
after intercourse on the fertile days. From our findings
together with those of other investigators, we conclude that
the effectiveness of LNG-EC estimated from the biological
parameters of the actual cycle gives more reliable results
than those calculated from LMP and historical menstrual
data, and that anamnestic data of the cycle overestimate the
contraceptive efficacy of LNG-EC. In particular, the
calendar method overestimates the efficacy of LNG-EC in
situations where the pills are taken after ovulation, and
thereby giving credence to the idea that LNG-EC may
interfere with embryo development and/or implantation.
Our finding that FR occurs after administration of
LNG-EC is in line with those reported by Croxatto et al.
[6]. In that study, the purpose was to examine the effect of
LNG-EC on follicles at different stages of development,
using follicular diameter as proxy for follicular develop-
ment. It was observed that FR was suppressed in a high
proportion when treatment was administered to women
with follicles with a diameter of b18 mm. In the current
study, FRs occurred in some two thirds of women taking
LNG-EC preovulatory; this suggests that other mechanism
than suppression of ovulation prevents pregnancy in these
women. We postulate that increased cervical mucus
viscosity caused by LNG impedes the migration of
420 G. Noé et al. / Contraception 81 (2010) 414–420
sperm from their reservoir in cervical crypts to the
Fallopian tubes; therefore the time elapsing between
LNG-EC intake and ovulation would be crucial. Such an
effect on sperm migration of lower doses of LNG has been
reported by Kesserü et al. [7]. Evaluating cervical mucus
properties at different times after LNG-EC administration
in women close to FR would clarify this mechanism of
pregnancy prevention.
In conclusion, this study shows that LNG-EC is very
effective in preventing pregnancy when it is administered
before ovulation, but it is ineffective in preventing pregnancy
once fertilization has occurred. LNG-EC is less effective
than regular contraceptives and its use should be restricted to
emergency situations.
Acknowledgments
We thank Olav Meirik for his comments.
References
[1] Croxatto HB, Ortiz ME, Müller AL. Mechanisms of action of
emergency contraception. Steroids 2003;68:1095–8.
[2] Hapangama D, Glasier AF, Baird DT. The effects of peri-ovulatory
administration of levonorgestrel on the menstrual cycle. Contraception
2001;63:123–9.
[3] Durand M, Cravioto MC, Raymond EG, et al. On the mechanism of
action of short-term levonorgestrel administration in emergency
contraception. Contraception 2001;64:227–34.
[4] Marions L, Hultenby K, Lindell I, Sun X, Stabi B, Gemzell-Danielsson
K. Emergency contraception with mifepristone and levonorgestrel:
mechanism of action. Obstet Gynecol 2002;100:65–71.
[5] Marions L, Cekan SZ, Bygdeman M, Gemzell-Danielsson K. Effect of
emergency contraception with levonorgestrel or mifepristone on
ovarian function. Contraception 2004;69:373–7.
[6] Croxatto HB, Brache V, Pavez M, et al. Pituitary-ovarian function
following the standard levonorgestrel emergency contraceptive dose or
a single 0.75 mg dose given on the days preceding ovulation.
Contraception 2004;70:442–50.
[7] Kesserü E, Garmendia F, Westphal M, Parada J. The hormonal and
peripheral effects of d-norgestrel in postcoital contraception. Contra-
ception 1974;10:411–24.
[8] Brache V, Faúndes A, Johansson E, Alvarez F. Anovulation,
inadequate luteal phase and poor sperm penetration in cervical
mucus during prolonged use of Norplant® implants. Contraception
1985;31:261–73.
[9] Croxatto HB, Diaz S, Salvatierra AM, Morales P, Ebensperger C,
Brandeis A. Treatment with Norplant® subdermal implants inhibits
sperm penetration through cervical mucus in vitro. Contraception
1987;36:193–201.
[10] Yeung WS, Chiu PC, Wang CH, Yao YQ, Ho PC. The effects of
levonorgestrel on various sperm functions. Contraception 2002;66:
453–7.
[11] Bahamondes L, Nascimento JAA, Munuce MJ, Fazano F, Faúndes A.
The in vitro effect of levonorgestrel on the acrosome reaction of human
spermatozoa from fertile men. Contraception 2003;68:55–9.
[12] Palomino A, Boric A, Gabler F, Espinoza A, Vega M, Devoto L.
Efecto de levonorgestrel como anticonceptivo de emergencia sobre
receptores de progesterona durante la ventana de implantación. Revista
Cubana de Salud Pública 2003;29(Suppl № 1):38.
[13] Gemzell-Danielsson K, Marions L. Mechanism of action of mifepris-
tone and levonorgestrel when used for emergency contraception. Hum
Reprod Update 2004;10:341–8.
[14] Ugocsai G, Rozsa M, Ugocsai P. Scanning electron microscopic
(SEM) changes of the endometrium in women taking high doses of
levonorgestrel as emergency postcoital contraception. Contraception
2002;66:433–7.
[15] Lalitkumar PG, Lalitkumar S, Meng CX, et al. Mifepristone, but not
levonorgestrel, inhibits human blastocyst attachment to an in vitro
endometrial three-dimensional cell culture model. Hum Reprod 2007;
22:3031–7.
[16] Müller AL, Llados CM, Croxatto HB. Post-coital treatment with
levonorgestrel does not disrupt post-fertilization events in the rat.
Contraception 2003;67:415–9.
[17] Ortiz ME, Ortiz RE, Fuentes M, Parraguez VH, Croxatto HB. Post-
coital administration of levonorgestrel does not interfere with post-
fertilization events in the new-world monkey Cebus apella. Hum
Reprod 2004;19:1352–6.
[18] Novikova N, Weisberg E, Stanczyk FZ, Croxatto HB, Fraser IS.
Effectiveness of levonorgestrel emergency contraception given before
or after ovulation — a pilot study. Contraception 2007;75:112–8.
[19] Heiken JP, Lee JKT. MR imaging of the pelvis. Radiology 1988;166:
11–6.
[20] Zawin M. Imaging of the normal female pelvis. In: Schwartz LB,
editor. Diagnostic imaging for reproductive failure. New York: The
Parthenon Publishing Group Inc.; 1998. p. 37–48.
[21] WHO. Laboratory manual for the examination of human semen and
sperm–cervical mucus interaction. In: WHO Special Programme of
Research, Development and Research Training in Human Reproduc-
tion, editor. Sperm–cervical mucus interaction. 4th ed. Cambridge:
University Press; 1999. p. 51–8.
[22] Wilcox AJ, Weinberg CR, Baird DD. Timing of sexual intercourse in
relation to ovulation. Effects on the probability of conception, survival
of the pregnancy and sex of the baby. N Engl J Med 1995;333:
1517–21.
[23] Ho PC, Kwan MS. A prospective randomized comparison of
levonorgestrel with the Yuzpe regimen in post-coital contraception.
Hum Reprod 1993;8:389–92.
[24] Task Force on Postovulatory Methods of Fertility Regulation.
Randomized controlled trial of levonorgestrel versus the Yuzpe
regimen of combined oral contraceptives for emergency contraception.
Lancet 1998;352:428–33.
[25] von Hertzen H, Piaggio G, Din J, et al. Low dose mifepristone and two
regimens of levonorgestrel for emergency contraception: a WHO
multicentre randomised trial. Lancet 2002;360:1803–10.
[26] Arowojolu AO, Okewole IA, Adekunle AO. Comparative evaluation of
the effectiveness and safety of two regimens of levonorgestrel for
emergency contraception in Nigerians. Contraception 2002;66:269–73.
[27] von Hertzen H, Piaggio G. Levonorgestrel and mifepristone in
emergency contraception. Steroids 2003;68:1107–13.
[28] Hamoda H, Ashok PW, Stalder C, Flett GM, Kennedy E, Templeton
A. A randomized trial of mifepristone (10 mg) and levonorgestrel for
emergency contraception. Obstet Gynecol 2004;104:1307–13.
[29] Espinós JJ, Rodriguez-Espinosa J, Senosiain R, et al. The role of
matching menstrual data with hormonal measurements in evaluating
effectiveness of postcoital contraception. Contraception 1999;60:243–7.
[30] Stirling A, Glasier A. Estimating the efficacy of emergency contraception
— how reliable is the data? Contraception 2002;66:19–22.
[31] Severi FM, Bocchi C, Florio P, Cobellis L, Ignacchiti E, Petraglia F.
Transvaginal ultrasonography in women receiving emergency contra-
ception. Fertil Steril 2003;79:1074–7.