Keywords
venous thromboembolism - contraception - hormone replacement therapy - thrombophilia
- women
Introduction
Venous thromboembolism (VTE) is a common disease with an average incidence of 1 to
2 per 1,000 per year. The annual total burden of VTE has been estimated by modeling
country-specific data from six European countries (i.e., France, Germany, Italy, Spain,
Sweden, and the United Kingdom) with an overall population of 310.4 million inhabitants.[1] The extrapolation of data from these countries revealed approximately 760,000 patients
per year with nonfatal VTE and 370,000 patients per year with VTE-related death, of
which only 7% were diagnosed antemortem. In a meta-analysis of 18 studies involving
7,515 patients with a first unprovoked VTE who had completed at least 3 months of
anticoagulant therapy, the long-term risk for recurrent VTE was substantial.[2] The cumulative risk of recurrent VTE at 10 years after discontinuation of anticoagulation
was 41% in men and 29% in women. For recurrent VTE, a case fatality rate of 4% (95%
confidence interval [CI]: 2–6%) was reported. Moreover, within 5 years after the discontinuation
of anticoagulant therapy a substantial number of patients will develop long-term sequelae
after VTE such as postthrombotic syndrome (≈28%) or chronic thromboembolic pulmonary
hypertension (≈2%).[1]
Data from the Danish National Patient Registry covering all Danish hospitals indicate
that the lifetime risk of VTE is slightly higher in women than in men and that reproductive
factors contribute modestly to the risk increase in women.[3] In particular, women of childbearing age are at increased risk of VTE when compared
to men of the same age, which is mainly attributed to combined hormonal contraceptive
(CHC) use and pregnancy. Moreover, when compared to men, VTE manifests more often
with symptoms of pulmonary embolism (PE) in women.[4]
In this comprehensive review, specific issues of VTE in women will be highlighted.
Combined Hormonal Contraception and Risk of VTE
Combined Hormonal Contraception and Risk of VTE
Traditionally, estrogens are considered as sex hormones that mainly affect the development
and regulation of the female reproductive system. However, estrogens are ubiquitous
and have impact on almost all physiological organs and tissues including the cardiovascular
system. In humans, there are four natural estrogens that exert differential effects
throughout lifetime[5] ([Table 1]). The natural estrogens estrone (E1), estradiol (E2), and estriol (E3) as well as
synthetic estrogens have been introduced for therapeutic use more than 80 years ago.
Ethinyl estradiol (EE) was the first synthetic estrogen that has been applied for
the treatment of dysmenorrhea. Conjugated equine estrogens (CEE), a mixture of estrogen
conjugates found in the urine of mares, were approved for the relief of climacteric
symptoms such as hot flushes in the early 1940s. Approximately 20 years later, the
first combined oral contraceptive pill containing an estrogen in combination with
a gestagen component was introduced in the United States and has been available in
Germany since 1961.
Table 1
Natural and synthetic estrogens
|
Estrogen
|
Synthesis and function
|
|
Natural human estrogens
|
|
Estrone (E1)
|
Present throughout life, primary estrogen in menopause
|
|
Estradiol (E2)
|
Produced by the ovaries, primary estrogen during reproductive years
|
|
Estriol (E3)
|
Produced by the placenta, major estrogen during pregnancy
|
|
Estetrol (E4)
|
Produced by the human fetal liver after the 9th week of gestation, present only during
pregnancy with high levels in the fetus and lower levels in the maternal circulation
|
|
Synthetic estrogens
|
|
Ethinyl estradiol (EE)
|
First synthetic estrogen, introduced for the treatment of dysmenorrhea
|
|
Conjugated equine estrogens (CEE)
|
Estrogen conjugates found in the urine of mares, introduced for the treatment of hot
flushes in postmenopausal women
|
The association of VTE and the use of CHCs became evident shortly after the introduction
of CHCs.[6] The extent of the risk increase mainly depends on the content of estrogen (predominantly
EE), the gestagen component, and the route of administration. Since then, the composition
of CHCs has changed to include lower doses of estrogens and newer gestagen components,
but this has not resolved the problem of thromboembolic complications. A review by
the European Medicines Agency (EMA) in 2013 concluded that the risk of VTE was higher
for recently developed gestagen components in CHCs such as desogestrel or drospirenone
(DRSP) when compared to older gestagens such as levonorgestrel (LNG), norethisterone,
or norgestimate.[7] Consequently, European authorities decided to update the product information of
CHCs and to provide information to women regarding the risk of VTE related to CHC
use and to health care professionals regarding the types of CHC with the lowest VTE
risk. Khialani et al analyzed the CHC prescription pattern in three European countries
(the Netherlands, Denmark, United Kingdom) before and after the European Commission
mandated changes in product information.[8] Among new CHC users in these three countries, the proportion of women initiating
a CHC classified by the EMA as the safest option increased over time. Before the EMA
decision, ≈76% of new users received a CHC containing LNG , norethisterone, or norgestimate;
since 2014 this proportion has increased to ≈84%. In Germany, the proportion of women
receiving a low-risk CHC is substantially lower. A recent data analysis of the largest
public health insurance company (AOK) revealed that the proportion of women with low-risk
CHCs was 28% in 2009 and increased only moderately to 47% in 2019.[9]
Only recently, the “Bundesinstitut für Arzneimittel und Medizinprodukte” (BfArM) revised
its warning (“Rote-Hand-Brief”) concerning the use of CHCs, which is still the most
common method to avoid unwanted pregnancies in Germany.[10] Of note, the term CHC refers to oral as well as nonoral preparations (e.g., transdermal
patches, vaginal ring).
Oral Combined Hormonal Contraceptives
Modern oral preparations usually contain low-dose estrogen (e.g., 20–35 µg of EE)
and a synthetic gestagen (progestin). These “pills” increase the risk of VTE two-
to sixfold in females using hormonal contraceptives when compared to females not using
hormonal contraceptives. However, the absolute risk of VTE in young women without
additional risk factors remains low. In the general population, the overall VTE incidence
of women without hormonal contraception is approximately 2 per 10,000 per year ([Table 2]).[10]
Table 2
Incidence of VTE of the progestin component of traditional combined hormonal contraceptives
(CHCs) according to the latest warning (“Rote-Hand-Brief”) of the German Federal Institute
for Drugs and Medical Devices (Bundesinstitut für Arzneimittel und Medizinprodukte,
BfArM)[10]
|
Group
|
Brand name
|
Annual incidence of VTE
|
|
Women without hormonal contraception
|
–
|
2 per 10,000
|
|
CHC with levonorgestrel
|
e.g., Evaluna, Femigoa, Femikadin, Kleodina, Leios, Levomin, Microgynon, Minisiston,
Miranova, Monostep, Triquilar
|
5–7 per 10,000
|
|
CHC with norethisterone
|
e.g., Eve
|
|
CHC with norgestimate
|
–
|
|
CHC with extended-cycle levonorgestrel
|
e.g., Seasonique
|
5–15 per 10,000
|
|
CHC with etonogestrel
|
e.g., Nuvaring (vaginal ring)
|
6–12 per 10,000
|
|
CHC with norelgestromin
|
e.g., Evra (transdermal patch)
|
|
CHC with dienogest
|
e.g., Dienogenance, Dienovel, Finic, Maxim, Sillaba, Valette
|
8–11 per 10,000
|
|
CHC with drospirenone
|
e.g., Aida, Daylette, Drosfemine, Drospifem, Petibelle, Yasmin
|
9–12 per 10,000
|
|
CHC with gestodene
|
e.g., Femovan
|
|
CHC with desogestrel
|
e.g., Desmin, Desofemine
|
|
CHC with chlormadinon
|
e.g., Belara, Bellissima, Chariva, Enriqa, Madinette, Minette
|
Still unknown
|
|
CHC with nomegestrol[a]
|
e.g., Zoely
|
Abbreviations: CHC, combined hormonal contraceptive; VTE, venous thromboembolism.
a Recent data suggest that CHCs containing estradiol/nomegestrol acetate (E2/NOMAC;
brand name: Zoely), estradiol valerate/dienogest (E2V/DNG; brand name: Qlaira), or
estetrol/drospirenone (E4/DRSP; brand name: Drovelis) are noninferior to CHCs containing
levonorgestrel[16]
[17]
[18]
Estrogens bind to classic estrogen receptors (e.g., Erα, ERβ) that act as transcription
factors and regulate the transcription and expression of procoagulant, anticoagulant,
and fibrinolytic factors, leading to a state of hypercoagulability.[11]
[12] The use of CHCs is accompanied by increased activities of prothrombotic factors
(e.g., fibrinogen, prothrombin, and factors VII, VIII, and X) and reduced activities
of natural coagulation inhibitors (e.g., antithrombin, protein S, and tissue factor
pathway inhibitor). The extent of acquired resistance against activated protein C
(APC resistance) has been shown to correlate with the risk of VTE.[13]
[14] These hormonal effects on the coagulation system are more pronounced in obese women
and last up to 6 to 8 weeks after the cessation of CHC intake.[15]
The higher the estrogen dose is, the higher the risk of VTE. It has been suggested
that the use of newer CHCs containing natural E2, estradiol valerate (E2V), or estetrol
(E4) is associated with a lower risk of thromboembolic complications. In a large prospective
observational study, E2/nomegestrol acetate (E2/NOMAC; brand name: Zoely) was shown
to be noninferior to LNG-containing CHCs.[16] E2V has similar pharmacokinetic and pharmacodynamic properties to that of E2 as
it is converted to E2 in the intestinal mucosa. To overcome inacceptable bleeding
irregularities, E2V has been combined with dienogest (E2V/DNG; brand name: Qlaira).
A large international active surveillance study demonstrated that VTE occurred less
frequently with E2V/DNG in comparison to CHC containing LNG or other CHCs.[17]
Only recently, E4—in combination with DRSP (E4/DRSP; brand name: Drovelis)—has been
introduced as a new estrogen component in CHCs.[18]
[19] E4 is a natural estrogen that is produced in the fetal liver and—due to both agonist
and antagonist estrogenic properties—acts differentially on human tissues and has
only minimal impact on liver gene expression. Phase 2 trials revealed only minimal
impact on hemostasis biomarkers.[20]
[21] The changes of hemostatic parameters (e.g., endogenous thrombin potential, resistance
to APC, prothrombin fragment 1 + 2, sex hormone binding protein) related to the use
of E4/DRSP were at least similar or even less than that of EE/LNG and more pronounced
than that of an EE/DRSP combination. Moreover, E4/DRSP administered in a 24/4 regimen
completely inhibited ovulation and was associated with a favorable bleeding pattern
and good body weight control resulting in high user satisfaction.[18]
Among CHCs, formulations containing LNG, norethisterone, or norgestimate are associated
with the lowest risk of VTE,[22]
[23] whereas combinations with third- or fourth-generation progestins (e.g., DRSP, gestodene,
or desogestrel) exert a higher risk increase ([Table 2]). The risk of VTE is highest within the first year of prescription but remains elevated
even thereafter. In one study, the risk was fivefold increased even after more than
60 months of use compared to nonusers of hormonal contraception (odds ratio [OR]:
5.2; 95% CI: 4.3–6.2),[24] whereas another study reported ORs ranging from 2.4 to 4.6 depending on the progestin
component after at least 4 years of exposure.[25] Of note, the risk is temporarily increased again if CHC intake is resumed after
a period of interruption of several weeks or months.
In general, the absolute risk of VTE in young and healthy women of childbearing age
is rather low (annual incidence of VTE: 2 per 10,000). However, there may be a significant
risk increase if additional VTE risk factors are present (e.g., hereditary thrombophilia,
a positive family history of VTE, aged >35 years, obesity, smoking). Therefore, gynecologists,
hemostaseologists, and experts of vascular medicine, as well as the BfArM, call to
prescribe CHCs with the lowest risk increase and to choose or switch to an estrogen-free
contraceptive method for women with prior VTE, those with comorbidities predisposing
them to VTE (e.g., inflammatory bowel disease), or those with additional VTE risk
factors.[26] The relevant clinical risk factors and recommended consequences are shown in [Table 3].
Table 3
VTE risk factors, clinical relevance, and recommended consequences from the Association
of Scientific Medical Societies in Germany (AWMF)[26]
|
VTE risk factor
|
Risk increase
|
Recommendation
|
|
Age > 35 y
|
Low to moderate
|
Thorough risk assessment and counseling.
Generally, all contraceptive methods are applicable (provided that no additional risk
factors are present)
|
|
BMI > 35 kg/m2
|
Low to moderate
|
Thorough risk assessment and counseling.
If possible, avoid CHCs (especially if aged >35 years and/or smokes >15 cigarettes
per day)
|
|
Smoking
|
Low to moderate
|
Thorough risk assessment and counseling.
If possible, avoid CHCs (especially if aged >35 years and/or has a BMI >35 kg/m2)
|
|
Prolonged immobility, major surgery
|
Moderate to high
|
Thorough risk assessment and counseling.
No new CHC prescription.
Risk-adapted VTE prophylaxis measures.
|
|
Prior VTE
|
Moderate (after VTE related to a nonhormonal risk factor) to high (after unprovoked
or hormone-related VTE)
|
Avoid CHCs.
Consider consultation with a hemostaseologist.
Consider testing for thrombophilia.
|
|
Family history of VTE
|
Moderate (after VTE related to a nonhormonal risk factor) to high (after unprovoked
or hormone-related VTE)
|
Avoid CHCs.
Consider consultation with a hemostaseologist.
Consider testing for thrombophilia
|
|
Asymptomatic thrombophilia or known thrombophilia in a first-degree relative
|
Low to high (dependent on the thrombophilic defect and VTE manifestation in family
members)
|
Reluctant indication of screening for thrombophilia in asymptomatic individuals, in
case of thrombophilia consultation with a hemostaseologist before the prescription
of CHCs.
|
Abbr.: BMI, body mass index; CHC, combined hormonal contraceptive; VTE, venous thromboembolism.
The recently updated AWMF-S3 guideline states that an individual risk assessment should
be mandatory in each case in which a CHC prescription is planned.[26] The woman must be informed about the increased risk of VTE and alternative contraception
methods, and the woman's preference should be included in decision-making. Moreover,
if the woman decides to use a CHC, she should be informed about the leading symptoms
of deep vein thrombosis (DVT) and PE and advised to contact an emergency physician
or her general practitioner as soon as symptoms occur. In Germany, specific patient
information materials can be downloaded from the BfArM website.[10]
Nonoral CHC Formulations
Attention should be given to the fact that transdermal and transvaginal CHCs have
also been related to an increased risk of VTE and therefore provide no benefit in
terms of VTE risk.[27]
[28]
[29]
[30] Analyzing data from four national registries in Denmark, Lidegaard et al calculated
an eightfold increased relative risk of VTE in users of transdermal CHC patches (95%
CI: 3.5–17.7) when compared to nonusers of hormonal contraceptives.[28] The relative risk of a CHC-releasing vaginal ring was 6.5 (95% CI: 4.7–8.9). Thus,
the corresponding incidence rates are 9.7 and 7.8 per 10,000 years of exposure.
Estrogen-Free Contraception Methods
Progestin-only pills (POPs) do not contain any estrogen and, based on the current
state of scientific knowledge, do not increase the risk of VTE.[26]
[31] Common progestins used in POPs are LNG and desogestrel. In addition to the classical
oral application, intrauterine devices (IUDs), injectables for intramuscular application,
and subcutaneous implants that slowly release progestin have been approved. However,
because CHCs have been related to better menstrual control and have desirable CHC
side effects such as a decrease in acne and hirsutism, POPs are prescribed less frequently.[32] Only recently has a DRSP-only pill (brand name: Slinda) been approved by the EMA.
This progestin mono-preparation containing 4 mg DRSP is similar in efficacy to CHCs
and is a POP with a better bleeding profile than traditional POPs (i.e., lower rates
of unscheduled intracyclic bleeding and spotting). Moreover, DRSP has antigonadotropic,
antimineralocorticoidic, and antiandrogenic properties.[33] Whereas the intake of CHCs containing DRSP is associated with a sixfold increased
incidence of VTE, no cases of VTE and no changes in hemostaseological factors (i.e.,
APC resistance, protein C, antithrombin, factor VII, factor VIII, D-dimer levels)
were observed in more than 2,500 women and over 25,000 cycles in DRSP users.[34]
Because POPs do not increase the risk of VTE, they can be even used—among other estrogen-free
contraception methods—by women with a substantially heightened risk of VTE (e.g.,
known thrombophilia or prior VTE).[35] Unfortunately, this is in contrast to the summary of the product characteristics
of most POPs, which still consider acute VTE as a contraindication.[32]
The risk of VTE is also low in users of LNG-releasing IUDs.[28]
[36] Of note, users of injectable depot medroxyprogesterone acetate had a 1.9- to 3.6-fold
increased risk of VTE when compared to nonusers of hormonal contraceptives.[36]
[37]
Alternative estrogen-free contraception methods include copper IUDs and barrier methods
such as cervical caps, diaphragms, condoms and spermicidal foam, sponges, and films.
These methods also do not increase the risk of VTE. However, barrier methods are not
as effective as birth control pills or IUDs in preventing pregnancy.
Hereditary Thrombophilia and Hormonal Contraception
Hereditary Thrombophilia and Hormonal Contraception
Whether CHCs should be avoided by women without prior VTE but with known thrombophilia
or a positive family history of VTE is a controversial issue. Several studies revealed
an increased risk of VTE in women with hereditary thrombophilia who used CHCs.[38]
[39] However, there is consensus that a general screening for hereditary thrombophilia
before the prescription of CHCs is not advisable. Because factor V Leiden (FVL) and
prothrombin G20210A mutations are common in the European population, at least syllabication?
thrombophilia can be detected in 3 to 9% of healthy subjects ([Table 4]). Despite the high prevalence in the general population, the absolute risk of VTE
is low, provided that no other VTE risk factors are present. A recent French study
that included 2,214 relatives from 651 families with known thrombophilia calculated
an annual VTE incidence rate of 0.36% (hazard ratio [HR]: 1.91; 95% CI: 1.30–2.80)
in patients with mild thrombophilia and 0.64% (HR: 3.78; 95% CI: 2.50–5.73) in patients
with severe thrombophilia.[40]
Table 4
Classical hereditary thrombophilia: prevalence and relative risk of VTE[84]
[85]
[86]
|
Prevalence in the general population
|
Relative risk of a first VTE
|
|
Low-risk thrombophilia
|
|
FVL mutation, heterozygous
|
2–7%
|
4–6
|
|
PT mutation G20210A, heterozygous
|
1–2%
|
3–5
|
|
High-risk thrombophilia
|
|
FVL mutation, homozygous
|
0.01–0.02%
|
4–41
|
|
PT mutation G20210A, homozygous
|
Very rare
|
2–4
|
|
FVL and PT mutation, double heterozygous
|
Very rare
|
2–7
|
|
Antithrombin deficiency
|
0.02–0.2%
|
13–59
|
|
Protein C deficiency
|
0.2–0.5%
|
13–42
|
|
Protein S deficiency
|
0.1–0.7%
|
26–56
|
Abbreviations: FVL, factor V Leiden; PT, prothrombin time; VTE, venous thromboembolism.
However, there is a remarkable risk increase in women with thrombophilia who use CHCs.
The risk increase can be roughly estimated with multiplication of relative risks.[41] For example, assuming a basic annual risk of 2 per 10,000, the combined effect of
FVL (fivefold risk increase) and a CHC containing DRSP (sixfold risk increase) results
in an absolute annual risk of 60 per 10,000 or 0.6%. Only recently, a French study
group analyzed the risk of VTE in women with a FVL mutation who were prescribed different
CHCs. The authors demonstrated that the higher VTE risk associated with CHC containing
DRSP or cyproterone acetate, compared with second generation gestagens or norgestimate,
is even greater among FVL carriers.[42] A Dutch study analyzed the joint effects of common prothrombotic genetic factors
and CHC use and found that the risk of VTE was highest when both risk factors were
present.[43] The joint effect of FVL and CHCs containing LNG resulted in an OR of 17.4 (95% CI:
11.4–26.6), whereas the OR was 24.8 (95% CI: 12.3–50.0) when LNG-containing CHCs were
used in women with the prothrombin mutation G20210A. The risk increase was even higher
in women with a genetic factor using gestodene, desogestrel, and cyproterone.
Consequently, women with thrombophilia and a positive family history of VTE should
be counseled to preferably use an estrogen-free method for contraception, especially
if an index patient suffered unprovoked or hormone-related VTE at a younger age. If
the prescription of CHCs seems to be inevitable due to comorbidities, a thorough hemostaseologic
work-up is recommended, and the prescription of a CHC with LNG, norethisterone, or
norgestimate may be considered.
Contraception in Women with Acute VTE
Contraception in Women with Acute VTE
Women with acute CHC-related VTE often discontinue contraception use immediately after
the diagnosis of VTE. Consequently, withdrawal bleeding occurs in the initial phase
of anticoagulant therapy when higher doses of a factor-Xa inhibitor are required (e.g.,
10 mg of apixaban or 15 mg of rivaroxaban twice daily) or vitamin K antagonist (VKA)
therapy is given in addition to therapeutic-dose low-molecular-weight heparin (LMWH).
Thus, the early discontinuation of a CHC may result in heavy menstrual bleeding. Moreover,
the risk of unwanted pregnancies and oral anticoagulant exposure to a fetus will be
increased. Of note, direct factor Xa or thrombin inhibitors as well as traditional
VKAs cross the placental barrier and have the potential to harm a fetus. Beyer-Westendorf
et al collected all available case reports of direct oral anticoagulant (DOAC) exposure
in pregnancies and identified 614 unique cases.[44] Of the 336 pregnancies with available outcome data (55%), 74 (22%) resulted in miscarriage
and 21 (6%) in fetal abnormalities, of which 12 (4%) were adjudicated as major birth
defects potentially related to DOAC exposure. Data from an observational cohort study
including 1,642 pregnancies indicate that the risk for birth defects and fetal loss
of phenprocoumon and other VKAs seems to be time-dependent and increases steeply after
the 5th week of gestation.[45] The overall rate of major birth defects was 7.4% and spontaneous abortions occurred
in 38% of pregnancies with exposure to phenprocoumon, which is the most commonly used
VKA in Germany.
Therefore, the current AWMF-S3 guideline stipulates safe contraception for all women
of childbearing age who require oral anticoagulant therapy.[26] However, there has been controversy regarding the continuation of hormonal therapy
in women diagnosed with VTE. Whereas the World Health Organization (WHO) recommended
the cessation of CHCs at VTE diagnosis, the International Society on Thrombosis and
Haemostasis Scientific and Standardization Committee recommended discontinuing CHCs
before stopping anticoagulant therapy.[46]
[47]
To the best of current knowledge, the prothrombotic effect of CHCs is compensated
by full-therapeutic anticoagulation so that the continued intake of CHCs seems to
be justifiable. In a post hoc subgroup analysis of the EINSTEIN-DVT/-PE studies that
compared VTE recurrence among women with and without hormonal treatment before the
age of 60, there was no risk increase among those who continued hormonal therapy (3.7
vs. 4.7%; HR 0.56; 95% CI: 0.23–1.39).[48] To minimize the risk of VTE under anticoagulant therapy, an estrogen-free contraception
method, such as POPs, or a progestin, or copper IUD, should be preferred ([Table 5]). If a woman and her attending physician decide to continue CHCs while she is being
treated with oral anticoagulants, a switch to a combination containing a low-risk
progestin (e.g., LNG) can be considered. Of note, the current AWMF-S3 guideline recommends
switching to an estrogen-free contraception method at least 6 weeks before the discontinuation
of anticoagulant therapy.[26]
Table 5
AWMF-S3 guideline recommendation concerning contraception in women with VTE treated
with oral anticoagulants and after the termination of therapy[26]
|
Contraception before VTE
|
Contraception during anticoagulant therapy
|
Contraception after the termination of anticoagulant therapy
|
|
None
|
Estrogen-free method (except DMPA);
avoid CHCs (since not applicable after the termination of anticoagulation)
|
Estrogen-free method (except DMPA)
|
|
CHCs
|
Switch to an estrogen-free method or continue CHCs and switch to an estrogen-free
method at latest 6 weeks before the termination of anticoagulant therapy
|
Estrogen-free method (except DMPA)
|
|
Progestin-only contraceptives (oral or IUD, except DMPA)
|
Continue progestin-only contraceptives
|
Estrogen-free method (except DMPA)
|
|
Barrier methods (e.g., condoms)
|
To increase contraceptive efficacy, consider switching to a preferably estrogen-free
method (except DMPA)
|
Estrogen-free method (except DMPA)
|
Abbreviations: DMPA, depot medroxyprogesterone acetate; IUD, intrauterine device;
VTE, venous thromboembolism.
Infertility Treatment
Infertility is common and affects approximately 10% of couples. The number of pregnancies
achieved by fertility treatment (artificial reproduction technologies [ART]) has been
increasing in recent decades. Complications of fertility treatment comprise VTE, ovarian
hyperstimulation syndrome (OHSS), and implantation failure.
The risk of VTE after in vitro fertilization (IVF) was increased during the whole
duration of pregnancy but was highest during the first trimester (HR: 4.22; 95% CI:
2.46–7.26). VTE incidence is significantly higher after successful ART cycles than
after unsuccessful cycles (9.4 vs. 1.3 per 1,000 cycles; OR: 13.94; 95% CI: 1.41–137.4).[49]
In a Swedish population-based cohort study including 902,891 first pregnancies between
1992 and 2012, women who achieved pregnancy after IVF with fresh embryo transfer had
a ninefold increased risk of VTE in the first trimester compared to that in women
with spontaneous pregnancies (HR: 8.96; 95% CI: 6.33–12.67).[50] No significant risk increase was observed for pregnancies after frozen-thawed embryo
transfer. The absolute incidence of VTE was 1.77 per 1,000 pregnancies after IVF with
fresh embryo transfer as compared to 0.14 per 1,000 after natural conception and 0.61
per 1,000 after IVF with frozen-thawed embryo transfer.
Particular attention should be given to women with OHSS which is a recognized complication
of ART after controlled ovarian hyperstimulation with exogenous gonadotropin administration.
OHSS occurs in moderate or severe forms in 2 to 3% of all IVF patients[51]
[52] and is accompanied by high E2 concentrations. A substantially increased incidence
of first-trimester VTE has been reported in women with moderate or severe OHSS (16.8
per 1,000 pregnancies; OR: 99.7; 95% CI: 61.6–161.1) as compared to women with natural
pregnancies.[53] Thromboembolism is considered the most severe complication related to OHSS, and
unusual site thrombosis involving the veins of the upper extremities and the internal
jugular or cerebral veins has been reported.[54]
[55]
[56] In women with severe OHSS, a VTE incidence of 1 to 4% has been reported. Therefore,
guidelines recommended thromboprophylaxis with LMWH for up to 3 months after the resolution
of OHSS.[57]
Pregnancy-Associated VTE
VTE is one of the leading causes of maternal mortality in the Western world. According
to a 2014 WHO systematic analysis, PE accounted for 14% of the maternal deaths in
developed countries.[58] VTE complicates 1 to 2 per 1,000 pregnancies. Overall, the risk of VTE in pregnant
women is four- to fivefold higher than that in age-matched nonpregnant women. VTE
risk increases with gestational age and is highest around the time of delivery and
immediately postpartum. A Danish population-based study revealed an ≈20-fold risk
increase at term.[59] Approximately one-third of pregnancy-related DVT cases and half of pregnancy-related
PE cases occur after delivery. Risk factors and VTE prevention strategies have been
recently summarized in this journal by the working group “Women's Health” of the “Society
of Thrombosis and Haemostasis” (GTH) and are not further discussed here.[60]
Hormonal Replacement Therapy
Hormonal Replacement Therapy
Estrogen therapy is the most effective treatment option for relieving climacteric
symptoms such as hot flashes, sleep disturbances, depression, and urogenital complaints
due to the atrophy of mucous membranes. Because estrogen monotherapy is associated
with an elevated risk of endometrial cancer, estrogens are usually applied in combination
with a progestin component. In this regard, estrogens are applied in the lowest effective
dose and in the shortest time frame needed to provide the relief of symptoms.
The different estrogen types used for hormone replacement therapy (HRT)—mainly CEE
or E2—are less potent than EE used in CHCs. Oral HRT preparations, in monotherapy
as well as in combination with a progestin, increase the risk of VTE two- to threefold.[61]
[62] For women with prior VTE, an up to fourfold risk increase has been described.[63]
[64] Similar to CHCs, the risk of VTE increases with the estrogen dose and depends on
the progestin compound. The risk is highest within the first year of application and
remains elevated for the duration of therapy. A further risk increase is observed
with the presence of additional VTE risk factors, such as older age, a higher body
mass index, or the presence of thrombophilia. Because HRT patients are generally older,
HRT populations have a higher absolute baseline risk than CHC populations. The overall
annual incidence of DVT among Swedish women before the age of 40 was less than 20
per 100,000 per year but increased to more than 50 per 100,000 per year after the
age of 50.[65] In contrast to CHCs, transdermal applications for HRT exert no influence on the
risk of VTE.[66]
[67]
[68]
Sex-Specific Aspects of VTE Recurrence
Sex-Specific Aspects of VTE Recurrence
VTE Recurrence in Women
In general, the risk of VTE recurrence depends on the site and manifestation of the
initial VTE, the presence or absence of transient or persistent risk factors, and
comorbidities. Several risk prediction scores have been validated in recent years
with the objective of predicting the risk of VTE recurrence and individualizing the
duration and intensity of therapy (e.g., DASH, HERDOO2, Vienna prediction model).[69]
[70]
[71] The sex aspect was included in each model.
Several studies have identified the male sex as an independent risk predictor of VTE
recurrence.[72]
[73]
[74] A patient-level meta-analysis including 2,554 VTE patients who were followed for
2.5 years on average after stopping anticoagulant therapy revealed a 2.2-fold higher
risk of recurrence after a first unprovoked VTE in men than in women.[75] Even after adjustment for women with hormone-related VTE, the recurrence risk in
men remained significantly higher (HR: 1.8; 95% CI: 1.4–2.5).
In a multinational cohort management study (REVERSE-II), the HERDOO2 clinical decision
rule was prospectively validated to identify women at low risk of recurrent VTE.[70] Women with none or one of the four HERDOO2 criteria (i.e., hyperpigmentation, oedema,
or redness in either leg; a D-dimer level ≥250 µg/L while using anticoagulants; a
body mass index ≥30 kg/m2; or an age ≥65 years) had a low risk of recurrence (3.0%
per patient year; 95% CI: 1.8–4.8%) and therefore could safely discontinue anticoagulants
after 5 to 12 months of treatment. In contrast, high-risk women (scores ≥ 2) and men
who discontinued anticoagulant therapy had a substantially higher risk of VTE recurrence
(8.1% per patient-year; 95% CI: 5.2–11.9%). Whether these high-risk women and men
require indefinite anticoagulation remains controversial since the annual risk of
recurrent VTE seems to decrease after the first 1 to 2 years.
Risk of Recurrence after Hormone-Related VTE
According to the current risk estimations, CHC use is generally considered to be a
minor transient risk factor for VTE.[76] However, the continuation of CHC intake after the termination of anticoagulant therapy
is supposed to put women at high risk for VTE recurrence. Therefore, CHCs should definitely
be discontinued or switched to estrogen-free contraception at least 6 weeks before
the cessation of anticoagulant therapy.
In general, women have a lower risk of recurrence than men of the same age after a
first VTE. Among women with a first unprovoked VTE, the cumulative incidence for recurrent
VTE was 8.9% in the first year, 13.6% at 2 years, 21.5% at 5 years, and 28.8% at 10
years.[2] The corresponding incidence rates for VTE recurrence in men were 11.9% in the first
year, 18.3% at 2 years, 28.6% at 5 years, and 41.2% at 10 years. The risk of recurrence
is also lower after hormone-related VTE than after unprovoked VTE.[75]
[77]
[78] In cohort studies, the absolute annual risk of recurrence was 1.1 to 2.5%.[79]
[80]
[81] This must be balanced against the risk of major bleeding under continued anticoagulant
therapy. For example, phase 3 trials comparing DOACs with the traditional anticoagulation
regimen of heparin followed by VKAs revealed an average major bleeding rate of 1.0%
for patients taking DOACs and 1.7% for those taking VKA.[82] Some studies report an even higher bleeding risk for women than for men.[83] When balancing the risks and benefits of anticoagulation after a CHC-related VTE,
therapy—in the absence of additional or persistent risk factors—is often limited to
3 to 6 months.
Final Remarks
Because estrogens and estrogen–progestin combinations put women in jeopardy of thromboembolic
events, an individual risk assessment and weighing of risks and benefits, including
a woman's preferences, is mandatory before introducing any hormone therapy. The woman
must be informed about the risk increase and the symptoms of DVT and PE and should
be instructed to contact her general practitioner or an emergency physician as soon
as symptoms occur. The same applies for women for whom ART or HRT due to climacteric
symptoms is planned. As a matter of principle, a CHC composition with the lowest known
risk of VTE should be chosen. Only recently, new formulations (e.g., combinations
containing E4 and DRSP as well as DRSP-only pills) have been approved in Europe and
many other countries and can be considered as an important step forward in terms of
efficacy, better bleeding profile, and low risk of thromboembolic complications.
