Contraception 84 (2011) 465 – 477
Review article
Fertility after discontinuation of contraception: a comprehensive
review of the literature☆
Diana Mansoura,⁎, Kristina Gemzell-Danielssonb , Pirjo Inkic , Jeffrey T. Jensend
a
Newcastle Sexual Health Services, New Croft Centre, New Croft House, Newcastle upon Tyne NE1 6ND, UK
Division of Obstetrics and Gynaecology, Department of Women's and Children's Health, Karolinska Institutet,
Karolinska University Hospital, SE-171 76 Stockholm, Sweden
c
Bayer HealthCare Pharmaceuticals, D-13342 Berlin, Germany
d
Department of Obstetrics and Gynecology, Oregon Health and Science University, Portland, OR 97239, USA
Received 1 October 2010; revised 29 March 2011; accepted 8 April 2011
b
Abstract
Background: Fear of adverse effects on subsequent fertility following reversible contraceptive use is an important concern for some women.
Study Design: We undertook a comprehensive and objective review of the literature for prospective studies reporting pregnancy rates in
women following contraceptive cessation. One-year pregnancy rates and pregnancy outcomes are summarized.
Results: We identified and analyzed 17 studies according to preset criteria. Typical 1-year pregnancy rates following cessation of oral
contraceptives or the levonorgestrel-releasing intrauterine system (LNG-IUS) ranged between about 79% and 96% and those for copper
intrauterine devices (IUDs) were almost as high, ranging between about 71% and 91%. One-year pregnancy rates following cessation of
contraceptive implants were between 77% and 86%, with one study showing a rate lower than 50%. For injectable contraceptives [(a)
norethisterone enanthate and (b) 5 mg estradiol cypionate and 25 mg medroxyprogesterone (Cyclofem)], only two studies were reported,
with 1-year pregnancy rates following cessation of 73% and 83%, respectively. There was no evidence of increased pregnancy
complications or adverse fetal outcomes following cessation of any of the reversible methods reported.
Conclusions: Overall, 1-year pregnancy rates following cessation of oral contraceptives, contraceptive implants and monthly injections,
copper IUDs and the LNG-IUS are broadly similar to those reported following discontinuation of barrier methods or use of no
contraceptive method.
Crown Copyright © 2011 Published by Elsevier Inc. All rights reserved.
Keywords: Fertility; Pregnancy rate; Contraceptives
1. Introduction
Women's reproductive goals and contraceptive practices
change depending on their life situation. The ability to
control fertility with highly effective reversible contracep☆
Conflicts of interest: Dr Mansour has received support to undertake
research, attend clinical meetings and scientific advisory boards from Bayer
HealthCare Pharmaceuticals; Dr Gemzell-Danielsson has participated on
scientific advisory boards, clinical trials and has been a speaker for Bayer
HealthCare Pharmaceuticals; Dr Pirjo Inki is an employee of Bayer
HealthCare Pharmaceuticals; and Dr Jensen has participated on scientific
advisory boards, received research support and been a speaker for Bayer
HealthCare Pharmaceuticals.
⁎ Corresponding author. Tel.: +44 0 191 229 2862; fax: +44 0 191
229 2979.
E-mail addresses: [email protected],
[email protected] (D. Mansour).
tives has emancipated many women to choose to become
pregnant by choice not chance. An essential feature of any
reversible method of contraception is that it should not
adversely affect future fertility—often an important concern
for women [1]. Any fertility delay or impairment following
the cessation of a given contraceptive method may be
associated with reduced use and poor user satisfaction
especially in young women [2], where misconceptions and
lack of information [2,3] add to their general distrust of
effective contraceptive methods.
In the 1960s, following discontinuation of oral contraceptives, initial reports suggested a syndrome of anovulation
and infertility in some women who previously had regular
menses [4–7]. Clinicians feared that long-term ovulation
suppression with oral contraceptives could lead to a state of
prolonged ovulatory suppression where return to normal
0010-7824/$ – see front matter. Crown Copyright © 2011 Published by Elsevier Inc. All rights reserved.
doi:10.1016/j.contraception.2011.04.002
466
D. Mansour et al. / Contraception 84 (2011) 465–477
menstruation was delayed resulting in some women remaining amenorrheic for more than 1 year. This so-called postpill
amenorrhea syndrome has been subsequently refuted [8].
Nonetheless, the effect of long-term oral contraceptive use
on reproductive function and fertility, particularly as it may
affect adolescent maturation and reproductive development,
continues to be of concern to some clinicians [3].
Fertility is a complex issue that depends on a number of
factors (both male and female) including underlying
reproductive potential, behavioral/lifestyle and environmental factors. The negative association between age and
fecundity is generally well accepted [9,10] and is related to
reduced coital frequency [11,12], a decline in fertility [10,13]
and increased prevalence of underlying gynecological
disease [14]. In addition, smoking in women has been
associated with impaired fertility [15–17], as has obesity
[18] and exposure to a number of environmental toxins
(including pesticides, solvents and chlorinated hydrocarbons) [19–21]. Consequently, studies assessing pregnancy
rates or time to pregnancy following cessation of contraceptive use may be influenced by a number of underlying factors
specific to the population under study.
The aim of this article was to systematically review the
available literature comparing pregnancy rates in women
who wanted to become pregnant after discontinuing various
contraceptive methods. While there have been recent
reviews focusing on this aspect for individual methods
such as oral contraceptives [22] and copper intrauterine
devices (IUDs) [23], there have been no comprehensive
reviews covering most reversible methods of contraception.
This information will be useful when counseling individuals
or couples about the impact of contraceptive use on their
future fertility, especially for those who have not yet
completed their families.
2. Methods
We performed a computerized literature search of Medline® and Embase® using the Ovid interface system to
retrieve relevant articles published up to 30 September 2009.
The combined search terms used were (where $ represents
truncation of a word that allows all possible suffix variations
of a root word to be retrieved; for example, the search
“discontin$” would retrieve documents with “discontinue,”
discontinuing,” “discontinuation” etc.): return to fertility;
return of fertility; time to pregnancy; planned pregnancy;
fecundity; delay in conception; fertility after discontin$; time
to conception; pregnancy delay; and time to ovulation.
Articles were included if they reported 1-year pregnancy
rates following discontinuation of a reversible contraceptive
method because of planned pregnancy. One-year pregnancy
rates were chosen as the endpoint of interest because couples
unable to conceive within 1 year of trying are generally
considered to have impaired fertility or infertile. Infertility
was beyond the scope of this review. The search was
restricted to prospective clinical/observational studies (with
at least 50 patients to obtain reasonably robust rates) and
English language reports.
Studies were selected in a two-stage process. First, the
titles and abstracts from the electronic searches were
scrutinized for articles that reported or were likely to report
pregnancy rates. Full manuscripts of all citations that were
likely to meet the predefined selection criteria were obtained
and scrutinized for inclusion in this review. The reference
lists of primary and pertinent review articles were also
examined to identify cited studies not captured by the
electronic search. We did not specifically search for studies
assessing the return of fertility after stopping “natural”
contraceptive methods (such as lactational amenorrhea,
fertility awareness-based methods or withdrawal)—these
data were only recorded if they were included as a control
group in studies that met the inclusion criteria.
Time to pregnancy rates were also collated for those who
became pregnant within 1 year. As pregnancy rate was the
outcome of interest, studies reporting delivery/outcome rates
only were excluded. Studies published before 1980 were also
excluded, as were those assessing fertility after abortion.
Since the studies reflected data collected from a variety of
settings and time points, no statistical analyses were
performed. Our aim was to summarize the 1-year and time
to pregnancy rates by contraceptive method discontinued
and provide an estimate of variability between studies. The
fertility outcomes in these studies were also reviewed. The
reasons for failure to achieve pregnancy in these studies were
beyond the scope of this review.
3. Results
We identified and extracted information from 17
prospective studies (studies assessing more than one method
are only counted once) reporting pregnancy rates following
cessation of reversible contraceptive use to become pregnant
(see Appendix A). Two studies were excluded because they
included a mixture of both prospective and retrospective data
or included women who tried to conceive following an
interval after discontinuation of a contraceptive method
[24,25]. Of the studies included, 3 assessed oral contraceptives [26–28], 10 evaluated nonhormonal IUDs (including
copper [29–35], plastic [36] or a mixture of device types
[37,38]), 3 assessed the levonorgestrel-releasing intrauterine
system (LNG-IUS) [31,32,34], 5 assessed implants (Norplant or Norplant II [32,36,39,40] and Implanon [41]) and 2
injectable contraceptives [(a) norethisterone enanthate and
(b) 5 mg estradiol cypionate and 25 mg medroxyprogesterone (Cyclofem)] [35,42]. No appropriate data were found on
the vaginal ring, the contraceptive patch, depo-medroxyprogesterone acetate (DMPA) or barrier methods.
Table 1 summarizes the pregnancy rates at 1 year (or 13
cycles) following cessation of various reversible contraceptive methods because of planned pregnancy. The 1-year
D. Mansour et al. / Contraception 84 (2011) 465–477
467
Table 1
One-year pregnancy rates and median times to pregnancy
Contraceptive method
Mean exposure timea
(years)
Mean age at discontinuationa
(years)
One-year pregnancy
rate (%)
Median time to
pregnancyb
Oral contraceptives [26–28]
Copper intrauterine devices [29–35]
Levonorgestrel-releasing
intrauterine system [31,32,34]
Implants [32,36,39–41]
Injectables [35,42]d
1.7–7.2
2.0–3.2
1.9–2.5
26.8–28.1
25.0–30.5
26.8–27.8
79.4–95
71.2–91.1
79.1–96.4
2.5–3 cycles
2–3.7 months
4 months
2.4–4.7
0.6–1
25.8–29.7
24.5–25.6
37.5–85.6c
72.5–82.9
2.9–7.7 monthsc
4.5–5 months
a
Studies that reported this parameter.
For those who became pregnant within 1 year (data estimated from available data).
c
One study reported unusually low 1-year pregnancy rates (37.5% and 48.8% for Norplant and Implanon, respectively) [41]. If this study was excluded, then
the 1-year pregnancy rates ranged between 76.5% and 85.6% and median time to pregnancy of 2.9–4.4 months.
d
The injectable contraceptives pertain to (a) norethisterone enanthate and (b) 5 mg estradiol cypionate and 25 mg medroxyprogesterone (Cyclofem).
b
pregnancy rates reported included a mixture of crude
cumulative pregnancy rates or life-table rates (gross rates
or otherwise not specified). Available data indicate that there
is wide overlap in the reported 1-year pregnancy rates after
contraceptive discontinuation. Typical 1-year pregnancy
rates after stopping oral contraceptives or the LNG-IUS
ranged between about 80% and 95% and those for copper
IUDs were almost as high ranging between about 70% and
90% (increasing to 80% and 99% at 2 years). Implants had
the widest range for 1-year pregnancy rates reported—one
study reported unusually low 1-year pregnancy rates (37.5%
and 48.8% for Norplant and Implanon, respectively) [41]. If
this study was excluded, then the 1-year pregnancy rates for
implants ranged between 76.5% and 85.6%, increasing to
88.3%–95.7% by 2 years.
The median time to pregnancy is also summarized in
Table 1; these data were abstracted from the studies selected
for evaluation and pertain only to those women who
conceived within 1 year of cessation of their contraceptive
method. Median time to pregnancy data following oral
contraceptives use was reportedly based on the number of
cycles in contrast to months for the other methods. Although
implants had the widest range for median time to pregnancy,
this was due to the study with unusually low 1-year
pregnancy rates [41]. If this study was excluded, then
median time to pregnancy reported for implants based on
two studies was 2.9 and 4.4 months [36,39], in line with the
estimates with other reversible methods.
Data for the Lippes loop intrauterine contraceptive device
was excluded from Table 1 because it is no longer marketed.
However, the 1-year pregnancy rate following its removal
was 74.7% (86.7% at 2 years) in one study with a median
time to pregnancy of about 3.8 months [36]. Two studies that
met inclusion criteria but did not specify the type of IUD or
reported data for a variety of IUDs (medicated, nonmedicated, Lippes loop, copper, progesterone-releasing intrauterine system) were also excluded from Table 1 [37,38]. The 1year pregnancy rates for these studies were 92.3% and
86.1%, respectively, with median time to pregnancy of about
1 month in the later study (appropriate data not available in
the former study).
The effect of age at contraceptive discontinuation on
pregnancy rates appears to be mixed, with some studies
showing a decrease in 1-year (or 2-year) pregnancy rates
with increasing age (although not always significantly so)
[26,29,30,32,33,38,39] and others showing no such decrease
[27,34,37,42]. In addition, the effect of parity on pregnancy
rates was inconsistent, with some studies suggesting that
age-adjusted pregnancy rates were significantly higher in
parous women [26], lower (but not significant pb.1) in
nulliparous women [33], unaffected by parity [32] or
significantly lower in parous women [29]. Those studies
that assessed the effect of duration of contraceptive use
suggest that pregnancy rates were not affected [26–29,31–
34,37–39,42]. For oral contraceptives, Cronin et al. [26]
showed that current smokers had a lower pregnancy rate than
nonsmokers, but the type of progestin in the formulation had
no major effect on pregnancy rate. Weight does not appear to
affect pregnancy rates following injectable contraceptive use
[42], nor does duration of amenorrhea or a history of pelvic
inflammatory disease (PID) while using the LNG-IUS [32]
or bleeding pattern before implant removal [39]. However,
most studies had small sample sizes, and this limited the
ability to draw conclusions as to potential interactions
between individual characteristics of subjects (such as age,
parity or obesity) and the contraceptive method.
Pregnancy outcomes were reported in eight studies
(Table 2). None of the studies following discontinuation of
oral contraceptive use that met our inclusion criteria
reported pregnancy outcomes. Another study that evaluated
pregnancy outcomes following use of a variety of IUDs
was excluded from the summary table because the results
combined the outcomes of inert, copper and progesteronereleasing devices [37]. The terms “live delivery” and “term
delivery” were used alternatively between studies and were
not generally defined, making direct comparisons across
studies difficult. Across the available studies, live deliveries
ranged between 84% and 95% and term deliveries between
79% and 93%, with no obvious trend across the various
methods assessed. For copper IUDs, spontaneous and
induced abortions ranged between 6% and 15% and
between 1% and 4%, respectively, and were similar to
468
D. Mansour et al. / Contraception 84 (2011) 465–477
Table 2
Pregnancy outcomes
Contraceptive method
Formulation
(trade name)
Pregnant
(n)
Copper T-200
Nova-T
671
50
Sivin et al. 1992 [32]
Skjeldestad and Bratt 1987 [33]
TCu380Ag
Copper IUDs
66
95
88.4
15
8.4
Belhadji et al. 1986 [34]
TCu380Ag
17
88
12
Copper IUDs
Tadesse 1996 [30]
Andersson et al. 1992 [31]
LNG-IUS
Andersson et al. 1992 [31]
LNG-20 IUS
104
Sivin et al. 1992 [32]
LNG-20 IUS
68
Belhadji et al. 1986 [34]
LNG-20 IUS
22
Implants
Buckshee et al. 1995 [39]
Sivin et al. 1992 [32]
Sivin et al. 1992 [32]
Diaz et al. 1987 [40]
Injectables
Bahamondes et al. 1997 [42]
Norplant II
Norplant II
136
86
Norplant
Norplant
33
75
Cyclofem-monthly
58
Pregnancy outcomes (%)
Live
births
Term
delivery
87.8
84
82
85.6
89
86
Induced
abortion
Ectopic
pregnancy
4
2
3
1.1
0
2.1
5.8
2.9
1
9
0
0
8.5
6
14
89.7
94.8
Spontaneous
abortion
Additional information
88
4.4
8
5.9
2
1
93
79
4
9
4
0
3.4
3.7% preterm (not defined)
2% outcome unknown and
2% still birth
Copper IUDs assessed:
Nova T, MLCu 250
and MLCu 375
11 women still pregnant
at time of analysis
4.8% still pregnant at time
of analysis
2% premature delivery
(not defined)
17 women still pregnant at
time of analysis
1% premature delivery
(not defined)
5% premature delivery
and 5% outcome unknown
1.7% hydatidiform mole
Data summarized pertains only to studies the met the criteria for inclusion in this review (see Methods section).
those reported for the LNG-IUS (6%–14% and 0%–3%,
respectively). For implants, spontaneous and induced
abortions ranged between 4% and 9% and between 2%
and 6%, respectively. Ectopic pregnancies were generally
rare, occurring at 0%–2% across the various methods.
There was no apparent association between the occurrence
of congenital malformations or newborn health concerns in
studies that reported fetal outcomes [32,33,39,40,42].
4. Discussion
This comprehensive review shows that there is wide
overlap in the reported 1-year pregnancy rates after the
cessation of various methods of contraception. The baseline
prevalence of infertility will influence the fertility rates of
women seeking pregnancy following discontinuation of a
contraceptive method. The ranges of the 1-year pregnancy
rates for oral contraceptives, copper IUDs and the LNG-IUS
broadly overlap those reported in retrospective studies of
women wishing to conceive following discontinuation of
barrier methods or using no contraceptive method (85.2%–
94%) [43–46]. Moreover, these results are at least consistent
with 1-year pregnancy rates (92%) reported by Gnoth et al.
[47] in women who proactively used “natural family
planning” to conceive—most of these women had used
fertility awareness as their contraceptive method immediately before trying to conceive. Overall, population surveys
suggest 12-month infertility prevalence rates of 3.5%–16.7%
(median 9%) [48].
Similar results were noted among studies of contraceptive
implants. The 1-year pregnancy rates for past users of
implants was unexpectedly wide (37.5–85.6%) and showed
greater variability than those reported following barrier
method use or no contraceptive method. This anomaly was
attributable to one study in Indonesian women that reported
unusually low 1-year pregnancy rates (37.5% and 48.8% for
Norplant and Implanon, respectively) [41]. Lower than
expected pregnancy rates were noted even up to 2 years
following implant discontinuation. Although factors that
might explain the low pregnancy rates were not discussed
further in the original report, it is possible that there was
underreporting of aborted pregnancies. Another explanation
may be that the motivation to become pregnant was lower in
this population of women with a mean parity of 2.3 than
among women in the comparator studies. Considering that
the low pregnancy rate was seen with both implants and is
not consistent with other reports or with a biologically
D. Mansour et al. / Contraception 84 (2011) 465–477
plausible mechanism, it is unlikely to be a clinically
important difference.
The median time to pregnancy in women who conceived
within 1 year after oral contraceptives (2.5–3 cycles) and
copper IUDs (2–3.7 months) use were at least consistent
with those reported for no contraceptive method (1.5–2.0
months) [43,44,47]. Median time to conception in those who
proactively used natural family methods to conceive within 1
year has been estimated as two cycles [47]. These results are
probably not suggestive of transient persistent ovarian
suppression with oral contraceptives or residual foreign
body reaction following removal of copper IUDs. Moreover,
the LNG-IUS (4 months) and implants (2.9–4.4 months,
excluding the study in Indonesian women with unexpectedly
low pregnancy rates) had median time to pregnancy at least
consistent with copper IUD use. Indeed, following removal
of the LNG-IUS or implants, plasma progestin levels
becomes undetectable within a few days [49–51].
For monthly injectable contraceptives (i.e., norethisterone
enanthate [35] and Cyclofem [42]), the 1-year pregnancy
(72.5% and 82.9%) rates were also generally lower than
those reported following barrier method use or no contraceptive method. A biologically plausible mechanism may
exist, as hormone levels remain elevated in some women
who receive depot injections. Although no studies with depot
medroxyprogesterone met our inclusion criteria, reported 1year pregnancy rates following discontinuation typically
range between 51.6% and 78.2% [24,52,53]. Two-year
pregnancy rates following discontinuation of depot
medroxyprogesterone are typically N90% [24,53].
Median times to pregnancy following injectable contraceptives use (i.e., norethisterone enanthate [35] and
Cyclofem [42]) were slightly longer (4.5 and 5 months)
relative to the other contraceptive methods assessed. In
addition, discontinuation was defined as 90 days [35] or 30
days [42] after the last injection in the two studies,
respectively, which may be considered as an additional
delay should the desire for pregnancy be made shortly after
receiving the injection. The longer median time to
conception following injectable contraceptives use (in
addition to the recognized 90- and 30-day contraceptive
efficacy period, respectively) may reflect prolonged transient
residual contraceptive effects [54,55]. Nonetheless, resumption of ovulation has been reported to occur in 70% of
women within 90 days (100% within 140 days) after the last
norethisterone enanthate injection. For Cyclofem, although
follicular activity usually returns within 28 days after the
injection, luteal function can be suppressed for at least seven
weeks [55]. In comparison, the mean time for return to
ovulation after discontinuing depot medroxyprogesterone
has been reported as 210 days after the last injection [54].
The pharmacokinetics of DMPA and other injectables
suggest a mechanism for prolonged contraceptive effects in
some users beyond the recognized window of contraceptive
efficacy and a basis for slightly lower 1-year pregnancy rates
following use of injectable contraceptives. Indeed, a
469
prospective follow-up assessing ovarian function and return
of fertility following DMPA discontinuation in 188 women
who dropped out of the UpJohn collaborative DepoProvera® clinical study to become pregnant published in
1979 (we excluded from our review studies published before
1980) by Schwallie and Assenzo, reported a pregnancy rate
of 67% at 12 months since last injection (median time to
conception 10 months) [56]. However, there was a high loss
to follow-up or a change of mind regarding becoming
pregnant (39%) in this latter study.
There was heterogeneity in the reporting of various
subject characteristics (i.e., not all the relevant parameters
influencing fertility were consistently reported such as
smoking habits, history of sexually transmitted infections
and prevalence of nulliparity). In general, factors that may
influence fertility rates were accounted for by conducting
subgroup analyses, although not all studies did formal
statistical subgroup analyses or showed evidence of
accounting for confounding factors. For example, age may
be a confounding factor in subgroup comparisons based on
duration of use or parity. However, the mean age across all
the studies included in this review ranged between 24.8 and
30.5 years in those studies that reported this parameter, with
no obvious differences in the ages of participant between the
various methods or studies. In addition, the small sample
sizes of most of the studies limited the ability to draw
conclusions about subgroups. Three studies did not perform
any subgroup analyses [35,36,41].
There were no apparent concerns regarding pregnancy
outcomes following cessation of the various contraceptive
methods in the studies that reported this outcome. Spontaneous
and induced abortions were reported consistently across these
studies, ranging between 3% and 15% and between 0% and
6%, respectively. The incidence of spontaneous abortion
across the studies is consistent with that generally accepted
once pregnancy has been clinically recognized (12%–15%)
[57]. A history of one or more induced abortion was reported
by 15% of the participants in the Nurses’ Health Study II [58],
although it was not clear how many among these were
specifically trying to get pregnant. Importantly, there were no
congenital malformations or newborn health concerns in
studies that reported fetal outcomes [32,33,39,40,42].
This review has a number of limitations that need to be
considered. A meta-analysis of the available data to yield an
overall combined mean return to fertility for each method
reviewed could not be done due to the inherent heterogeneity
between studies in terms of the type of data reported (crude
rates vs. life-table rates). In addition, there was heterogeneity
in the reporting of some patient characteristics that are
known to affect fertility such as the proportion of current
smokers [59] or those with history of sexually transmitted
infections [60]. Noticeably absent was information on the
husband or partner, as well as frequency and timing of
intercourse. Available evidence highlights the need to time
intercourse during the few days just before ovulation (i.e.,
the fertility window) to achieve pregnancy [61,62]. In
470
D. Mansour et al. / Contraception 84 (2011) 465–477
addition, the reasons for not achieving pregnancy were not
usually reported in the studies assessed. Thus, it is possible
that some data may have been influenced by undiagnosed
subfertility and/or a lack of knowledge by women of their
fertility window, and as such, not truly representative of
fertile couples maximizing their conception chances. Nonetheless, taken together, these results are important when
counseling couples about contraceptive reversibility as they
provide reassurance that past use of any contraceptive
method does not reduce long-term subsequent fertility and
that concern for subsequent fertility should not influence
contraceptive decision making. However, a transient delay in
achieving pregnancy may occur following discontinuation of
injectable contraceptives.
The results of this review need to be interpreted in the
context of regional differences in fecundity that may affect
time to pregnancy that include behavioral, environmental
and/or genetic factors [63,64]. Prospective studies of
contraceptive methods randomly assigned and used for a
proscribed interval and thereafter discontinued with the
intention to become pregnant would be required to minimize
the effects of confounding factors and to enable more precise
comparisons of pregnancy rates and time to pregnancy
following contraceptive discontinuation. Since such a study
is not likely, results from this review provide reassuring
information for contraceptive users.
Acknowledgments
Medical writing support for the development of this
manuscript was provided by Richard Glover of InScience
Communications, a Wolters Kluwer business, with the
financial support of Bayer HealthCare Pharmaceuticals.
Appendix A. Fertility after discontinuation of contraception: a comprehensive review of the literature
Contraceptive
method
Oral contraceptives
Cronin et al.
2009 [26]
Formulation
(trade name)
Exposure time
Subjects
(n)
Demography
Pregnancy rate§
Study design
Setting
Notes
Oral
contraceptives
(various)
Mean
7.2 years
2,064
Mean age
28.1 years
1-year
cumulative rate:
Prospective
observational
Austria, Belgium,
Denmark, France,
Germany, The
Netherlands, UK
Progestin type
did not affect
pregnancy rate
65% nulliparous
• 79.4%
[95% confidence
interval (CI)
77.6%–81.1%]
Median time
to pregnancy:
Wiegratz et al.
2006 [27]
Zimmermann
et al. 1999 [28]
30 mcg EE/2 mg
DNG (21/7;
Valette)
30 mcg EE/2 mg
DNG (21/7;
Valette)
Mean
21.5±16.8
cycles
Median:
4–6 cycles
(estimated from graph)
706
348
Mean age:
26.8±4.3 years
Mean: 26.8 years
(for those who
completed 1 year)
• 2.5 cycles
(estimated from graph)
1-year
cumulative rate:
• 86.8% (FAS)
• 94.0%
(of those with
complete data, 613/652)
Median time
to pregnancy:
• 2.5 cycles
(estimated from graph)
1-year rate:
2-year pregnancy
rate 88.3% (95%
CI 86.8%–89.6%)
Prospective
observational
Germany
2-year pregnancy
rate not reported
Prospective
observational
Germany
2-year pregnancy
rate not reported
Prospective
Belgium
TTP data
not presented
D. Mansour et al. / Contraception 84 (2011) 465–477
38.3% smokers
Smokers had a low
pregnancy rate.
• 95% (of those
who completed
study, 173/183)
Median time
to pregnancy:
• 3 cycles
(estimated from graph)
Copper IUDs
Delbarge et al.
2002 [29]
GyneFix
Mean
104.6±93.5
weeks
128
Mean age
30.5±4.5 years
1-year
cumulative rate:
• 88%
2-year pregnancy
rate 99%
471
(continued on next page)
472
Appendix A (continued)
Contraceptive
method
Formulation
(trade name)
Exposure time
Tadesse
1996 [30]
Copper T-200
Median
3.5 years
Subjects
(n)
780
Pregnancy rate§
Study design
Setting
Notes
Mean age 29 years
1-year rate:
Prospective
observational
Ethiopia
TTP data
not presented
Excluded those with
PID, infertility, heavy
menstrual bleeding,
previous ectopic
pregnancy
2-year pregnancy
rate not reported
Prospective
Denmark, Finland,
Hungary, Norway
and Sweden
2-year pregnancy
rate 79.7%
Prospective
observational
Not specified
No TTP data
3.2% nulligravid
• 86%
93.5% multipara
Andersson et al.
1992 [31]
Nova-T
Median 21 months
(range 6–53)
71
3.3% grand
multipara
Median 27 years
(range 19–38)
14.1% history
of PID
Sivin et al.
1992 [32]
Skjeldestad and
Bratt 1987 [33]
TCu380Ag
Copper IUDs
(Nova T, MLCu 250
and MLCu375)
Mean
38.9±2.2 months
56% b24 months
103
101
28.6 (SE 0.4)
years
50% para 0 or 1
History of
PID 16%
Mean age
28.3 years
17% Nulliparous
5% treated
for PID
Belhadji et al.
1986 [34]
TCu380Ag
23.96 (SE 1.86) years
50
Mean age at
insertion 25. 1
(SE 0.53) years
at insertion
Parity 1.7
(SE 0.14)
1-year cumulative
gross life-table
(Tietze) rate:
• 71.2%
Median time
to pregnancy:
• 2 months
(estimated from graph)
1-year life-table rate:
• 77% (all)
• 79% (b30 years)
1 year cumulative
gross life-table
(log-rank) rate
(estimated from graph):
• 81%
Median time
to conception:
• 3 months
(estimated from graph)
1-year
cumulative rate:
• 91.1% (SE 7.6%)
Median time
to conception:
• 3 months
2-year pregnancy 88%
Prospective
observational
Norway
2-year gross
cumulative rate
(estimated from
graph): 93%
Prospective
Not specified
2-year pregnancy
rate not reported
D. Mansour et al. / Contraception 84 (2011) 465–477
Demography
ICMR Task Force
on hormonal
contraception
1986 [35]
Copper T-200
Other IUDs (not LNG-IUS)
Gupta et al.
IUDs (various
1989 [37]
Copper plus
Progestasert 52 and
Lippes loop)
28.2±14.7 months
56% b2 years
110
Mean age
25.0±3.1 years
Parity 1.8±0.9
1-year
cumulative rate:
• 83.6%
Median time
to conception:
• 3.7 months
Prospective
India
2-year pregnancy
rate not reported
91
78% b30 years
1-year
cumulative
rate:
Prospective
Not specified
TTP not specified
nor could a specific
time be estimated
• 92.3%
Lippes C IUDs
31.8±8.6 months
Randic et al.
1985 [38]
IUD (medicated
and nonmedicated;
otherwise,
not specified)
59% b3 years
75
576
Mean age
25.9±3.6 years
0% nullipara
Mean age
27.5 years
1-year
cumulative rate:
• 74.7%
Median time
to conception
• 3.8 months
(estimated from graph)
1-year life-table rate:
Prospective
Indonesia
Prospective
Yugoslavia
2-year pregnancy
91.8% (Weighted
overall rate)
Prospective
Denmark, Finland,
Hungary, Norway
and Sweden
2-year pregnancy
rate 86.6%
Prospective
observational
Not specified
TTP not reported
• 88.6% (b30 years)
• 78.4% (≥30 years)
Weighted overall
rate (86.1%)
Median time
to conception
(estimated from table)
• 1 month
LNG-IUS
Andersson et al.
1992 [31]
LNG-20 IUS
Median 19 months
(range 3–50)
138
Median 27 years
(range 18–36)
46% para
0 or 1
12.4% history
of PID
Sivin et al.
1992 [32]
LNG-20 IUS
Mean
30.2±1.7 months
91
27.8 (SE 0.4)
years
History of
PID 13%
1-year cumulative
gross life-table
(Tietze) rate:
• 79.1%
Median time
to pregnancy:
• 4 months
(estimated from graph)
1-year life-table rate:
• 84% (all)
D. Mansour et al. / Contraception 84 (2011) 465–477
Affandi et al.
1987 [36]
Virtually all
conceived (96.7%)
within 18 months
2-year pregnancy
86.7%
2-year pregnancy
88%
(continued on next page)
473
• 89% (b30 years)
• 73% (≥30 years)
474
Appendix A (continued)
Contraceptive
method
Formulation
(trade name)
Exposure time
Subjects
(n)
Belhadji et al.
1986 [34]
LNG-20 IUS
22.72 (SE 1.42)
years
60
Demography
Pregnancy rate§
Study design
Setting
Notes
Mean age
25.10 (0.48)
years at insertion
Parity 1.7
(SE 0.11)
1-year cumulative rate:
Prospective
Not specified
2-year pregnancy
rate not reported
Prospective
Indonesia
2-year pregnancy
rate 60.0%
Prospective
Indonesia
2-year pregnancy
rate 73.8%
Prospective
India
2-year pregnancy
rate 88.3%
Prospective
observational
Not specified
2-year
pregnancy 92%
TTP not reported
Prospective
observational
Not specified
2 year
pregnancy 87%
TTP not reported
• 96.4% (SE 3.5%)
Implants
Affandi
1999 [41]
Implanon (68 mg
etonogestrel rod)
Mean
35.3±13.1 months
80
Mean age
28.0±3.2 years
Mean parity: 2.3
Affandi
1999 [41]
Norplant (six 36 mg
levonorgestrel rods)
Mean 55.8±17.7
months
80
Mean age
27.9±3.4 years
Mean parity: 2.2
Buckshee et al.
1995 [39]
Norplant II (two 75-mg
levonorgestrel rods)
30.0±11.5 months
159
25.8±3.3 years
All parous
Sivin et al.
1992 [32]
Sivin et al.
1992 [32]
Norplant II (two 75-mg
levonorgestrel rods)
Norplant (six 36-mg
levonorgestrel rods)
Mean 29.1±1.2
months
Mean 33.6±1.9
months
116
62
26.7 (SE 0.3)
years
47% para 0 or 1
28.2 (SE 0.5)
years
48% para 0 or 1
1-year
cumulative rate:
• 48.8%
Median time
to pregnancy
• 4.6 months
(estimated from graph)
1-year
cumulative rate:
• 37.5%
Median time
to pregnancy
• 7.7 months
(estimated from graph)
1-year
cumulative rate:
• 80.3%
Median time to
conception
• 2.9 months
(estimated from graphs)
1-year life-table rate:
• 84% (all)
• 85% (b30 years)
• 80% (≥30 years)
1-year life-table rate:
• 83% (all)
• 86% (b30 years)
• 76% (≥30 years)
D. Mansour et al. / Contraception 84 (2011) 465–477
Median time
to conception:
• 4 months
Affandi et al.
1987 [36]
Diaz et al.
1987 [40]
Norplant (six 36-mg
levonorgestrel rods)
Norplant
(six 36-mg
levonorgestrel rods)
32.4±8.8 months
1–8 years
51
87
Mean age
26.0±3.6 years
0% nullipara
Mean age
29.7±5 years
Prospective
Indonesia
2-year
pregnancy 90.2%
• 76.5%
Median time
to conception
• 4.4 months
(estimated from graph)
1-year life-table rate:
Prospective
Chile
Comparison with
copper T–IUDs
(n=44)
TTP data not presented
(data presented in
3-month intervals for
meaningful comparison)
2-year pregnancy
rate 95.7%
Prospective
Brazil, Chile,
Columbia, Peru
2 year pregnancy
rate not reported
Prospective
India
2-year pregnancy
rate not reported
• 85.6%± 4.1%
Injectables
Bahamondes
et al. 1997 [42]
ICMR
Task Force
on Hormonal
Contraception
1986 [35]
§
Cyclofem-monthly
(5 mg estradiol
cypionate and 25 mg
medroxyprogesterone)
Norethisterone
Enanthate
(NET EN) 200 mg
Mean number
of injections
7.1±4.6
11.9±4.9 months
70
69
Mean age
25.6±4.4 years
1-year
life-table rate
Mean age
24.8±3.3 years
• 82.9%
Median time
to pregnancy
(estimated
from graph):
• 4.5 months
1-year
cumulative rate:
Parity 2.2±1.3
• 72.5%
Median time
to conception:
• 4.5 months
D. Mansour et al. / Contraception 84 (2011) 465–477
1-year cumulative rate:
Median time to pregnancy for those who got pregnant within 1 year. TTP, time to pregnancy; FAS, full analysis set.
475
476
D. Mansour et al. / Contraception 84 (2011) 465–477
References
[1] Sihvo S, Hemminki E, Kosunen E. Contraceptive health risks—
women's perceptions. J Psychosom Obstet Gynaecol 1998;19:117–25.
[2] Perlman SE, Richmond DM, Sabatini MM, Krueger H, Rudy SJ.
Contraception. Myths, facts and methods. J Reprod Med 2001;46:
169–77.
[3] Hitchcock CL, Prior JC. Evidence about extending the duration of oral
contraceptive use to suppress menstruation. Wom Health Iss 2004;14:
201–11.
[4] Horowitz BJ, Solomkin M, Edelstein SW. The oversuppression
syndrome. Obstet Gynecol 1968;31:387–9.
[5] Shearman RP. Amenorrhea after treatment with oral contraceptives.
Lancet 1966;2:1110–1.
[6] Kotz HL, Dodek OI. The oral contraceptives and infertility. Med Ann
Dist Columbia 1966;35:255–8.
[7] Dodek Jr OI, Kotz HL. Syndrome of anovulation following the oral
contraceptives. Am J Obstet Gynecol 1967;98:1065–70.
[8] Archer DF, Thomas RL. The fallacy of the postpill amenorrhea
syndrome. Clin Obstet Gynecol 1981;24:943–50.
[9] Schwartz D, Mayaux MJ. Female fecundity as a function of age: results
of artificial insemination in 2193 nulliparous women with azoospermic
husbands. Federation CECOS. N Engl J Med 1982;306:404–6.
[10] Dunson DB, Baird DD, Colombo B. Increased infertility with age in
men and women. Obstet Gyneco 2004;103:51–6.
[11] James WH. The causes of the decline in fecundability with age. Soc
Biol 1979;26:330–4.
[12] Mulligan T, Moss CR. Sexuality and aging in male veterans: a crosssectional study of interest, ability, and activity. Arch Sex Behav 1991;
20:17–25.
[13] Dunson DB, Colombo B, Baird DD. Changes with age in the level and
duration of fertility in the menstrual cycle. Hum Reprod 2002;17:
1399–403.
[14] DeCherney AH, Berkowitz GS. Female fecundity and age. N Engl J
Med 1982;306:424–6.
[15] Olsen J, Rachootin P, Schiodt AV, Damsbo N. Tobacco use, alcohol
consumption and infertility. Int J Epidemiol 1983;12:179–84.
[16] Howe G, Westhoff C, Vessey M, Yeates D. Effects of age, cigarette
smoking, and other factors on fertility: findings in a large prospective
study. Br Med J (Clin Res Ed) 1985;290:1697–700.
[17] Homan GF, Davies M, Norman R. The impact of lifestyle factors on
reproductive performance in the general population and those
undergoing infertility treatment: a review. Hum Reprod Update 2007;
13:209–23.
[18] Pandey S, Bhattacharya S. Impact of obesity on gynecology. Women's
Health (Lond Engl) 2010;6:107–17.
[19] Sallmen M, Lindbohm ML, Kyyronen P, et al. Reduced fertility among
women exposed to organic solvents. Am J Ind Med 1995;27:699–713.
[20] Smith EM, Hammonds-Ehlers M, Clark MK, Kirchner HL, Fuortes L.
Occupational exposures and risk of female infertility. J Occup Environ
Med 1997;39:138–47.
[21] Buck GM, Vena JE, Schisterman EF, et al. Parental consumption of
contaminated sport fish from Lake Ontario and predicted fecundability.
Epidemiology 2000;11:388–93.
[22] Barnhart KT, Schreiber CA. Return to fertility following discontinuation of oral contraceptives. Fertil Steril 2009;91:659–63.
[23] Skjeldestad FE. The impact of intrauterine devices on subsequent
fertility. Curr Opin Obstet Gynecol 2008;20:275–80.
[24] Pardthaisong T. Return of fertility after use of the injectable contraceptive
Depo Provera: up-dated data analysis. J Biosoc Sci 1984;16:23–34.
[25] Wilson JC. A prospective New Zealand study of fertility after removal
of copper intrauterine contraceptive devices for conception and
because of complications: a four-year study. Am J Obstet Gynecol
1989;160:391–6.
[26] Cronin M, Schellschmidt I, Dinger J. Rate of pregnancy after using
drospirenone and other progestin-containing oral contraceptives.
Obstet Gynecol 2009;114:616–22.
[27] Wiegratz I, Mittmann K, Dietrich H, Zimmermann T, Kuhl H. Fertility
after discontinuation of treatment with an oral contraceptive containing
30 mcg of ethinyl estradiol and 2 mg of dienogest. Fertil Steril 2006;
85:1812–9.
[28] Zimmermann T, Dietrich H, Wisser KH, Munch C. Fertility after
discontinuation of the dienogest-containing oral contraceptive Valette.
First data of an ongoing study. Drugs Today 1999;35:89–95.
[29] Delbarge W, Batar I, Bafort M, et al. Return to fertility in nulliparous
and parous women after removal of the GyneFix intrauterine
contraceptive system. Eur J Contracept Reprod Health Care 2002;7:
24–30.
[30] Tadesse E. Return of fertility after an IUD removal for planned
pregnancy: a six-year prospective study. East Afr Med J 1996;73:
169–71.
[31] Andersson K, Batar I, Rybo G. Return to fertility after removal of a
levonorgestrel-releasing intrauterine device and Nova-T. Contraception 1992;46:575–84.
[32] Sivin I, Stern J, Diaz S, et al. Rates and outcomes of planned pregnancy
after use of Norplant capsules, Norplant II rods, or levonorgestrelreleasing or copper TCu 380Ag intrauterine contraceptive devices. Am
J Obstet Gynecol 1992;166:1208–13.
[33] Skjeldestad FE, Bratt H. Return of fertility after use of IUDs (Nova-T,
MLCu250 and MLCu375). Adv Contracept 1987;3:139–45.
[34] Belhadj H, Sivin I, Diaz S, et al. Recovery of fertility after use of the
levonorgestrel 20 mcg/d or Copper T 380 Ag intrauterine device.
Contraception 1986;34:261–7.
[35] Anonymous. ICMR (Indian Council of Medical Research) Task Force
on Hormonal Contraception. Return of fertility following discontinuation of an injectable contraceptive-norethisterone enanthate (NET
EN) 200mg dose. Contraception 1986;34:573–82.
[36] Affandi B, Santoso SS, Djajadilaga, Hadisaputra W, Moeloek FA,
Prihartono J, et al. Pregnancy after removal of Norplant implants
contraceptive. Contraception 1987;36:203–9.
[37] Gupta BK, Gupta AN, Lyall S. Return of fertility in various types of
IUD users. Int J Fertil 1989;34:123–5.
[38] Randic L, Vlasic S, Matrljan I, Waszak CS. Return to fertility after IUD
removal for planned pregnancy. Contraception 1985;32:253–9.
[39] Buckshee K, Chatterjee P, Dhall GI, et al. Return of fertility following
discontinuation of Norplant(R)-II subdermal implants. ICMR Task
Force on hormonal contraception. Contraception 1995;51:237–42.
[40] Diaz S, Pavez M, Cardenas H, Croxatto HB. Recovery of fertility and
outcome of planned pregnancies after the removal of Norplant
subdermal implants or Copper-T IUDs. Contraception 1987;35:
569–79.
[41] Affandi B. Pregnancy after removal of etonogestrel implant contraceptive (Implanon). Med J Indones 1999;8:62–4.
[42] Bahamondes L, Lavin P, Ojeda G, et al. Return of fertility after
discontinuation of the once-a-month injectable contraceptive Cyclofem. Contraception 1997;55:307–10.
[43] Hassan J, Kulenthran A, Thum YS. The return of fertility after
discontinuation of oral contraception in Malaysian women. Med J
Malaysia 1994;49:348–50.
[44] Ang Eng S, Arshat H. Pregnancy after contraceptive use. Malays J
Reprod Health 1986;4:6–11.
[45] Linn S, Schoenbaum SC, Monson RR, Rosner B, Ryan KJ. Delay in
conception for former ‘pill’ users. JAMA 1982;247:629–32.
[46] Hassan MA, Killick SR. Is previous aberrant reproductive outcome
predictive of subsequently reduced fecundity? Hum Reprod
2005;20:657–64.
[47] Gnoth C, Godehardt D, Godehardt E, Frank-Herrmann P, Freundl G.
Time to pregnancy: results of the German prospective study and
impact on the management of infertility. Hum Reprod 2003;18:
1959–66.
[48] Boivin J, Bunting L, Collins JA, Nygren KG. International
estimates of infertility prevalence and treatment-seeking: potential
need and demand for infertility medical care. Hum Reprod 2007;
22:1506–12.
D. Mansour et al. / Contraception 84 (2011) 465–477
[49] Backman T, Rauramo I, Jaakkola K, et al. Use of the levonorgestrelreleasing intrauterine system and breast cancer. Obstet Gynecol 2006;
107:208.
[50] Croxatto HB, Diaz S, Pavez M, Cardenas H, Larsson M, Johansson
ED. Clearance of levonorgestrel from the circulation following
removal of Norplant subdermal implants. Contraception 1988;38:
509–23.
[51] Huber J, Wenzl R. Pharmacokinetics of Implanon. An integrated
analysis. Contraception 1998;58:85S–90S.
[52] Hassan MAM, Killick SR. Is previous use of hormonal contraception
associated with a detrimental effect on subsequent fecundity? Hum
Reprod 2004;19:344–51.
[53] Depo-Provera®. Physician information. Available at: http://
mediapfizercom/files/products/uspi_depo_provera_contraceptivepdf
(accessed 4 Jan 2010).
[54] Fotherby K, Howard G. Return of fertility in women discontinuing
injectable contraceptives. J Obstet Gynaecol 1986;6(Suppl 2):S110–5.
[55] Fotherby K, Benagiano G, Toppozada HK, et al. A preliminary
pharmacological trial of the monthly injectable contraceptive Cycloprovera. Contraception 1982;25:261–72.
[56] Schwallie PC, Assenzo JR. The effect of depo-medroxyprogesterone
acetate on pituitary and ovarian function, and the return of fertility
following its discontinuation: a review. Contraception 1974;10:
181–202.
477
[57] Simpson JL, Mills JL, Holmes LB, et al. Low fetal loss rates after
ultrasound-proved viability in early pregnancy. JAMA 1987;258:
2555–7.
[58] Michels KB, Xue F, Colditz GA, Willett WC. Induced and
spontaneous abortion and incidence of breast cancer among young
women: a prospective cohort study. Arch Intern Med 2007;167:
814–20.
[59] American Society for Reproductive Medicine. Smoking and infertility.
Fertil Steril 2004;81:1181–6.
[60] Sweet RL. Sexually transmitted diseases. Pelvic inflammatory disease
and infertility in women. Infect Dis Clin North Am 1987;1:199–215.
[61] Stanford JB, White GL, Hatasaka H. Timing intercourse to achieve
pregnancy: current evidence. Obstet Gynecol 2002;100:1333–41.
[62] 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.
[63] Juul S, Karmaus W, Olsen J. Regional differences in waiting time to
pregnancy: pregnancy-based surveys from Denmark, France, Germany, Italy and Sweden. The European Infertility and Subfecundity Study
Group. Hum Reprod 1999;14:1250–4.
[64] Jensen TK, Slama R, Ducot B, et al. Regional differences in waiting
time to pregnancy among fertile couples from four European cities.
Hum Reprod 2001;16:2697–704.