Postcoital treatment with levonorgestrel does not disrupt
A.L. Müller, C.M. Llados, H.B. Croxatto*
Pontificia Universidad Católica a de Chile, Facultad de Ciencias Biológicas, Unidad de Reproducción y Desarrollo, Av. Alameda Bernardo O’Higgins
Received 28 October 2002; received in revised form 8 January 2003; accepted 13 January 2003
Levonorgestrel (LNG), a progestin widely used for regular hormonal contraception, is also used for emergency contraception (EC) to prevent
pregnancy after unprotected intercourse. However, its mode of action in EC is only partially understood. One unresolved question is whether ornot EC prevents pregnancy by interfering with postfertilization events. Here, we report the effects of acute treatment with LNG upon ovulation,fertilization and implantation in the rat. LNG inhibited ovulation totally or partially, depending on the timing of treatment and/or total doseadministered, whereas it had no effect on fertilization or implantation when it was administered shortly before or after mating, or beforeimplantation. It is concluded that acute postcoital administration of LNG at doses several-fold higher than those used for EC in women, which areable to inhibit Ovulation, had no postfertilization effect that impairs fertility in the rat. 2003 Elsevier Inc. All rights reserved. Keywords: Levonorgestrel; Emergency contraception; Ovulation; Fertilization; Implantation;
1. Introduction
On the other hand, animal experimentation allows segregating
groups treated before or after critical events, such as ovulation,
Levonorgestrel (LNG), a progestin widely used for regular
fertilization and implantation in order to define the contribution of
hormonal contraception, is also used for emergency contraception
pre- and postfertilization events to the contraceptive efficacy of the
(EC), alone or combined with estrogen. Hormonal EC is taken after
drug. Even though extrapolation of the results to the human has
unprotected intercourse in order to prevent pregnancy. The mode of
considerable limitations, experiments in animals often shed light on
action of EC has become the subject of heated debate in several
possible mechanisms operating in the human.
Latin American and Caribbean countries. The main question is
It is known that acute treatment with LNG administered during
centered on whether or not EC prevents pregnancy by interfering
the follicular phase can inhibit or delay ovulation in various species,
with postfertilization events. This issue is of importance for many
including the human [1]. Also, acute treatment with LNG
people, considering that a new human life begins at the time
administered within the first 10 h after intercourse decreased the
fertilization is completed. Accordingly, interference with
number of spermatozoa recovered from the uterine cavity in women
postfertilization events would lead to loss of human life.
[21. To our knowledge, no such effect has been described in
When a woman uses EC she does not know whether she takes
animals. Furthermore, its action when given after mating is not well
the pins before or after ovulation and before or after fertilization.
documented. Therefore, we undertook experiments in animals to
For ethical and logistical reasons, it has not been possible to
investigate the possible occurrence of postfertilization effects.
segregate groups of women who take EC after fertilization in order
In this paper, we describe the effects of acute treatment with
to assess its effect on the establishment of pregnancy. Hence, there
LNG given either before or after ovulation or fertilization, or before
is no direct evidence, in favor or against, that acute treatment with
implantation upon fertility parameters in the rat. The main
LNG prevents pregnancy by interference with postfertilization
intervention variable was the time of administration relative to
ovulation, mating and fertilization.
* Corresponding author. Tel.: +56-2-686-2878; fax: +56-2-222-5515. E-mail address: [email protected] (H.B. Croxatto). 0010-7824/03/$ - see front matter 2003 Elsevier Inc. All rights reserved. doi: 10.1016/S0010-7824(03)000021-0number of number of ovulating ovulated ovulating ovulated 8.0 ±1.6 6.0 ±3.0 4.7 ±0.8 9.0 ±2.0 7.1 ±0.8 Diestrus Proestrus Estrus
Fig. 1. Effects of preovulatory administration of levonorgestrel (LNG) or vehicle upon ovulation in the rat. | = timing of each injection of LNG or vehicle; | =
the approximate time of ovulation (line dashed); • = autopsy. T he time scale indicates days of the estrous cycle, with light and dark periods depicted by
horizontal white and black bars, respectively. n = number of animals treated. † Per ovulating rat *p = 0.02.
The measured variables were percentage of animals ovulating,
mean number of ovulated eggs, mean percentage of fertilized eggs
Each rat was injected subcutaneously with LNG 50 ug/kg body
and mean number of implanted embryos.
weight per injection or with vehicle, except in one control group thatwas left undisturbed. This dose is approximately four times higherthan the dose used for EC and its bioavailability is presumed to be
2. Material and methods
higher due to systemic administration. This dose was given one tofour times with 12-h intervals. The timing of treatment relative toovulation, mating, fertilization and implantation is summarized in
Adult Sprague Dawley female rats weighing approximately 200
2.4. Effect of preovulatory treatment upon ovulation
g and adult male rats 5-7 months old were used. All animals werekept in the same room with water and pet chow ad libitum,
Each rat was injected with LNG or vehicle twice per day (8:00
illumination from 7:30 a.m. to 9:30 p.m. and temperature 20-24ºC.
a.m. and 8:00 p.m.) at both diestrus and proestrus [Group 1 (G1)] or
Females were submitted daily to vagina] smears. Estrous cycle
with a single injection at 8:00 p.m. of diestrus (G2) or at 8:00 a.m.
stages were classified according to Turner J 3 J. In order to obtain
of proestrus (G3). These rats were killed at estrus in order to assess
pregnant animals, each proestrous or estrous female was caged
percentage of animals ovulating and mean number of ovulated eggs
individually with two intact males for 1 h or all night. At the end of
this period, the presence of semen, blood or a plug in the vagina
2.5. Effect of treatment before mating and ovulation upon fertilization and
and/or the presence of spermatozoa in the vagina] smear verified
mating. If mating occurred, the day of estrus was designated day 1of pregnancy (P1).
The results of the previous experiment indicated that a single
Animal care and experimental procedures were done according
injection of LNG in proestrus was unlikely to affect ovulation.
with the ethical guidelines of the Institutional Ethics Committee.
Therefore, it was feasible to study effects of preovulatory treatmentupon fertilization and implantation.
Proestrous rats were injected with LNG or vehicle once at 4:00
p.m. Each rat was caged with two males from 9:30 to 10:30 p.m. Ifmating was verified, tile female was isolated and killed on P2 inorder to assess the percentage of fertilized eggs (G4). If mating had
Crystalline LNG was obtained from Norplant implants (Leiras,
not taken place by 10:30 p.m., the female remained overnight with
Turku, Finland) or from Schering AG, Berlin, Germany. LNG was
the males. When mating was verified the following morning, the
dissolved in 100 ul ethanol (Merck, Darmstadt, Germany) and was
female was isolated and killed on P2 in order to assess the
further diluted with 1,2-propanediol (Sigma, St. Louis, MO, USA)
percentage of fertilized eggs or killed on P12 to assess the number
Vehículo number of fertilized number of fertilized implanted implanted 9.2 ± 1.1 (6) 98 ± 1.8 (5) 10.2 ± 0.5 (5)
• 100 (5) 11.2 ± 0.4 (5) 11.0 ± 0.5 (7)*
• 94 ± 5.6 (2) 10.8 ± 0.5 (5) 10.0 ± 0.7 (4)
• 75 ± 13.8 (6) 15.3 ± 0.6 (6) 97 ± 3.3 (6) 12.9 ± 0.8 (8)
• 97 ± 3.3 (6) 11.8 ± 1.6 (4) 12.8 ± 0.8 (4) 11.0 ± 0.8 (5) 10.6 ± 0.7 (5) 11.8 ± 0.8 (4) 10.0 ± 0.9 (5) Pro P1 P2 P3 P4 P5 P12
Fig. 2. Effects of levonorgestrel (LNG) or vehicle upon fertilization and implantation in the rat. | = timing of each injection of LNG or vehicle; — = period of
pairing resulting in mating; A, B, and C = approximate times of ovulation, fertilization and implantation, respectively; • = autopsy. T he time scale indicates
days of the estrous cycle (Pro: proestrus; Est: estrus) and days of pregnancy (P1-12). n = number of animals treated. *Animals not injected with vehicle. None
of the differences was statistically significant.
Failure to mate was infrequent and females that failed to mate were
A separate assessment (not shown) determined that after mating
occurring overnight, fertilization was completed by midmorning inour rat colony. Thus, both of these treatments were given after
2.6. Effect of treatment after mating and before ovulation upon fertilization
Proestrous rats were caged with males from 9:00 to 10:00 p.m. 2.10. Assessment of ovulation, fertilization and
Those that mated were injected with LNG at 11:00 p.m. (G6) or
with LNG or vehicle at 11:00 p.m. and again 12 h later (G7). These
Unmated rats were killed by an overdose of ether at noon of
estrus to assess the number of oocytes present in their oviducts. Mated rats were killed between 3:00 and 4:00 p.m. on P2 or P12 to
2.7. Effect of treatment after ovulation and before mating upon
assess the number of fertilized and unfertilized eggs and the number
of implanted embryos, respectively. At estrus or P2, oviducts were
Females were injected with LNG or vehicle once at 8:00 a.m. on
removed and flushed with saline to count the number of eggs using
the day of estrus and immediately caged with males until 9:00 a.m.
low magnification (250X). When necessary, cumulus-egg
(G8). Those that mated were isolated and killed on P2 or P12.
complexes were dispersed with hyaluronidase (Sigma). In order toassess the occurrence of fertilization, eggs were placed betweenslide and cover slip, and were examined using bright field
2.8. Effect of treatment after ovulation and mating upon fertilization
microscopy (40OX). Eggs were considered fertilized when a sperm
tail was seen in the cytoplasm or in the perivitelline space. At P12,
Females were caged with males from 8:00 to 9:00 a.m. on the
each implantation site was dissected, using a stereomicroscope, to
day of estrus. lf mating was verified, the female was injected
verify the presence or absence of an embryo (60X). Only
immediately with LNG or vehicle and killed on P2 or P12 (G9).
implantation sites containing live embryos are presented in theresults. 2.9. Effect of treatment after fertilization uponimplantation
Proestrous rats were caged with males overnight. Mated females
were injected twice with LNG or vehicle at 11:00 p.m. of P1 and at
Ovulation data are expressed as percentage of rats that ovulated
11:00 a.m. of P2 (G10) or at 11:00 p.m. of P3 and at 11:00 a.m. of
calculated over the total number of animals treated and also as
mean number of ovulated oocytes per ovulating rat ± SE.
All rats that mated had either fertilized eggs or implanted embryos.
In those injected with LNG at 11:00 p.m. of proestrus and 11:00
Thus, only the mean percentage of fertilized eggs and the mean
a.m. of estrus (G7), the mean percentage of fertilized eggs was 94 ±
number implanted embryos is presented per group. One-sided
5.6% and the mean number of implanted embryos was 10.8 ± 0.5
Fisher's exact test or chi-square, as appropriate, was applied to
for LNG and 10.0 ± 0.7 for controls (Fig. 2). None of these
two-way tables of percentage of rats that ovulated. Kruskal-Wallis
differences was statistically significant.
test was used for mean number of ovulated oocytes or mean number
of implanted embryos. Differences were considered statistically
significant at p < 0.05. All statistical analyses were carried out on
3.4. Effect of treatment a after ovulation and before mating upon
Intercooled Stata 6.0 computational program for Windows (Stata
fertilization and implantation (G8)
Corporation, College Station, TX, USA).
The mean percentage of fertilized eggs in rats treated at 8:00 a.m.
of estrus and mated within the next hour was 75 ± 14.8% for LNG
and 97 ± 3.3% for controls. The mean number of implanted
3. Results
embryos was 15.3 ± 0.6 and 12.9 ± 0.8, respectively (Fig. 2). None
of these differences was statistically significant. 3. 1. Effect of preovulatory treatment upon ovulation (G1, G2 and G3)3.5. Effect of treatment a after ovulation and mating upon and
Following injection of LNG or vehicle twice per day at both
diestrus and proestrus (G1), the percentage of ovulating rats was 0%
and 83%, respectively. The mean number of ovulated eggs in the
The mean percentage of fertilized eggs in rats injected
control group was 8.0 ± 1.6 (Fig. 1). The percentage of ovulating
immediately after mating in early estrus was 97 ± 3.3% for
rats among those treated once in the evening of diestrus (G2) or in
LNG and 100% for vehicle, and the mean number of
the morning of proestrus (G3) was 25% and 63%, respectively, with
implanted embryos was 11.8 ± 1.6 for LNG and 12.8 ± 0.8
LNG, and 98% in both control groups. The difference between
for controls (Fig. 2). None of these differences was
experimental and control in G2 is statistically significant (1) =
0.02). The mean number of ovulated eggs was 6.0 ± 3.0 and 9.0 ±
2.0, respectively, for rats treated with LNG and 4.7 ± 0.8 and 7.1 ±
0.8 for controls. These differences were not statistically significant. 3.6. Effect treatment after fertilization upon implantation (G10 and G11)3.2. Effect of treatment before mating and ovulation upon fertilization andimplantation (G4 and G5)
All rats had implanted embryos on P12. The mean number of
implanted embryos in those treated at 11:00 p.m. of P1 and at 11:00
a.m. of P2 (G10) was 11.0 ± 0.8 for LNG and 10.6 ± 0.7 for
All rats in these experiments had either fertilized eggs on P2 or
controls. In those injected with LNG or vehicle at 11:00 p.m. of P3
implanted embryos on P12. The mean percentage of fertilized eggs
and at 11:00 a.m. of P4 (G 11), the mean number of implanted
after injecting LNG or vehicle at 4:00 p.m. of proestrus and mating
embryos was 11.8 ± 0.8 and 10.0 ±10.9, respectively (Fig. 2). None
between 9:30 and 10:30 p.m. (G4), was 100% in both groups.
of the differences was statistically significant.
Among those mated overnight (G5), the mean percentage of
fertilized eggs was 100% in the LNG group and 98 -- 1.8% in tile
controls, and the mean number of implanted embryos was 9.2 ± 1.1
and 10.2 -0.5, respectively (Fig. 2). None of these differences was
4. Discussion
The results presented show that acute treatment with LNG, either
3.3. Effect of treatment after mating and before ovulation upon
shortly before or after mating, before or after fertilization, or before
fertilization and implantation (G6 and G7)
implantation, have no effect on the establishment of pregnancy inthe intact rat (Fig. 2). This excludes that LNG interferes withpostfertilization steps in this species and is in keeping with the fact
All rats had either fertilized eggs on P2 or implanted embryos
on P12. The mean percentage of fertilized eggs in proestrous rats
that daily treatment of castrated rats with LNG is able to sustain
mated between 9:00 and 10:00 p.m. and injected with LNG at
pregnancy [4]. No evidence of antifertility effect attributable to
11:00 p.m. (G6) was 100% and the mean number of implanted
interference with postfertilization events could be demonstrated. In
embryos was 11.2 ± 0.4. The number of implanted embryos in
contrast, preovulatory treatment with LNG was able to inhibit
contemporaneous undisturbed animals (not injected) was 11.0 ± 0.5
ovulation. The suppression of ovulation was heavily dependent upon
the schedule of administration (Fig. 1). Acknowledgments
Following injection of LNG either twice per day, once in the
evening of diestrus or once in the morning of proestrus, the
Support for this study was provided by Fundación
percentage of animals that failed to ovulate was 100%, 75% and
Instituto Milenio de Investigación en Biología
37%, respectively. The effectiveness of a single injection
Fundamental y Aplicada, Cátedra Presidencial en
diminished as treatment became closer to the expected time of the
Ciencias, Fondecyt #898000-8 and RF98024#98.
ovulatory episode. Therefore, LNG inhibited ovulation totally orpartially depending on the timing of treatment and possibly the
References
total (lose administered. Tile dose-effect relationship was notadequately examined. The possibility that ovulation may have been
[1] Croxatto HB, Devoto L, Durand M, et al. Mechanism of action of
delayed cannot he excluded, since we did not perform autopsies on
hormonal preparations used for emergency contraception: a review of
the third day after estrus to test for tile occurrence of delayed
the literature. Contraception 2001;63:111-21.
[2] Kesserü E, Garmendia F, Westphal N, Parada J. The hormonal and
Inhibition of ovulation in the rat after single administration of
peripheral effects of d-norgestrel in postcoital contraception.
LNG has been documented previously. Ovulation was delayed in
rats injected in the afternoon of metaestrus with L NG 50 ì g/kg
[3] Turner CD, editor. General endocrinology. Philadelphia: WB
dissolved in oil, whereas it was totally suppressed by 200 ì g/kg [5].
When LNG dissolved in oil was injected in the afternoon of
[4] Kuhnz W, Beier S. Comparative progestational and androgenic
diestrus, it blocked ovulation in a dose-dependent fashion; the
activity of norgestimate and levonorgestrel in the rat. Contraception
antiovulatory dose was 200-400 ì g/kg [6]. T he most obvious
differences with our experiments are the vehicle used and the time
[5] Beattie CW, Corbin A. The differential effects of diestrous
of administration. In our study, 50 ì g/kg of L NG dissolved in
progestogen administration on proestrous gonadotrophin levels.
1,2-propanediol injected in the evening of diestrus inhibited
ovulation in 75% of the animals and four doses administered 12 h
[6] Upton GV, Corbin A. The relevance of the pharmacological properties
apart beginning in the morning of diestrus (in total = 200 ì g/kg),
of a progestational agent to its clinical effects as a combination oral
contraceptive. Yale J Biol Med 1989;62:445-57.
There is much information on the effects of LNG upon
[7] Hapangama D, Glasier AF, Baird DT. The effects of peri-ovulatory
ovulation in women. In brief, LNG can inhibit ovulation when it is
administration of levonorgestrel on the menstrual cycle.
administered one or more days before follicular rupture [7-9]. It is
known that LNG affects ovulation in animals [5,10] and women
[8] Durand M, Cravioto M, Raymond EG, et al. On the mechanism of
action of short-term levonorgestrel administration in emergency
[11], at least through interference with signaling from the pituitary.
Contraception. Contraception 2001;64:227-34.
Inconsistent effects of LNG upon endometrial markers of
[9] Marions L, Hultenby K, Lindell 1, et al. Emergency contraception
receptivity have been reported in women [8,9]. The effect of LNG
with mifepristone and levonorgestrel: mechanism of action. Obstet
upon markers of receptivity in the rat was not addressed in the
[10] Spona J, Schneider WHF, Bieglmayer C, Schroeder R, Pirker R.
In conclusion, postcoital administration of LNG at doses able
Ovulation inhibition with different doses of levonorgestrel and other
to affect ovulation had no antifertility effect attributable to
progestogens: clinical and experimental investigations. Acta Obstet
interference with postfertilization events in the rat.
[11] Croxatto HB. Mechanisms that explain the contraceptive action of
progestin implants for women. Contraception 2002;65:21-7.
The Golden Pheasant Inn, Plumley Moor Road, Plumley, Restaurant booking recommended, bar reservations not required DISCRETIONARY 10% SERVICE CHARGE MAY BE ADDED TO TABLES OF 8 OR MORE ALL OF THIS IS GIVEN TO OUR HARDWORKING STAFF AND NOT TO THE ‘MANAGEMENT’ THE STARTERS ; TRADITIONAL PRAWN COCKTAIL Succulent North Atlantic Prawns in our homemade seafood sauce CHEF’S HOMEMADE SOUP OF THE