IV Guideline
1 Introduction
Counselling and treating couples with RM is a diagnostic and therapeutic challenge
as several possible causes for RM are known, but no risk factor for RM is identified
in the majority of affected patients.
2 Incidence and Definition
Approximately 1 – 3% of all couples of reproductive age experience recurrent miscarriage
[4]. A miscarriage is defined as the loss of a fetus at any time between conception
and the 24th week of gestation (GW) or the loss of a fetus weighing < 500 g [5]. The World Health Organization (WHO) definition of recurrent spontaneous miscarriage
is: “three and more consecutive miscarriages before the 20th GW” [5]. The American Society for Reproductive Medicine (ASRM) already defines the occurrence
of two consecutive miscarriages as RM [3], [6]. This definition increases the incidence of RM to 5% of all couples of reproductive
age [7]. This guideline takes the WHO definition (≥ 3 consecutive recurrent miscarriages)
as the basis for its recommendations on diagnostic and therapeutic procedures.
If a woman has not previously given birth to a live infant, the loss of the fetus
is referred to as primary RM; if the woman has had a previous live birth, the pregnancy
loss is referred to as secondary RM [8]. Another classification, which refers to the course of the miscarriages, classifies
miscarriages into repeated loss of embryonic pregnancy (sporadic loss) or loss of
fetal pregnancy (detectable heart beat on sonography or histologically verifiable
embryo) [3].
The risk of recurrent miscarriage varies significantly, depending on a number of different
factors. In addition to maternal age, the number of previous miscarriages also affects
the risk of recurrence. [Table 4] presents the data from a retrospective registry study [9].
Table 4 Probability of recurrent miscarriage depending on maternal age and the number of
previous miscarriages, based on the study of Nybo-Andersen et al. [9].
|
Previous miscarriages
|
Risk of recurrence
|
|
25 – 29 years
|
30 – 34 years
|
35 – 39 years
|
40 – 44 years
|
|
1 miscarriage
|
~ 15%
|
~ 16 – 18%
|
~ 21 – 23%
|
~ 40%
|
|
2 miscarriages
|
~ 22 – 24%
|
~ 23 – 26%
|
~ 25 – 30%
|
~ 40 – 44%
|
|
≥ 3 miscarriages
|
~ 40 – 42%
|
~ 38 – 40%
|
~ 40 – 45%
|
~ 60 – 65%
|
3 Diagnosis and Treatment of Relevant Risk Factors
3.1 Lifestyle and behavior
3.1.1 Stress
Some studies have indicated that higher stress levels during pregnancy might be associated
with an increased risk of pregnancy loss. A case-control study of 45 patients with
RM concluded that stress levels were higher compared to 40 control patients [10]. A study of 301 patients with RM (defined as ≥ 3 miscarriages) compared to women
wanting to children reported similar findings [11]. Because of the small number of cases, it is not possible, based on the currently
available data, to conclude that stress increases the risk of miscarriage.
3.1.2 Coffee consumption
A few observational studies have reported a dose-dependent relationship between coffee
intake and late loss of pregnancy [12]. A larger case-control study was also able to show that coffee consumption had an
impact on early miscarriage [13]. Another retrospective case-control study demonstrated a significantly increased
risk of RM following coffee consumption in the periconceptional period and in early
pregnancy. It was not possible to show a linear association between the amount of
coffee consumed and the risk of RM [14].
3.1.3 Nicotine and alcohol consumption
There is a strong association between nicotine consumption and poor obstetric and
neonatal outcomes such as ectopic pregnancy, stillbirth, placenta previa, premature
birth, low birthweight and congenital malformation. Cessation of smoking should therefore
be recommended to all pregnant women [15]. The impact of smoking and of cessation of smoking on the risk of RM is not clear.
A retrospective study compared 326 patients with RM with 400 control patients who
had had at least one live birth. The study showed that even passive smoking significantly
increased the risk of RM [16]. Another study came to the conclusion that maternal nicotine, alcohol or coffee
consumption was not associated with a higher probability of RM [17].
A prospective study, which evaluated the impact of paternal smoking on the risk of
miscarriage, investigated 526 couples and was able to show that women who were heavy
smokers (> 20 cigarettes per day) had a higher risk of early miscarriage. Heavy smoking
(more than 20 cigarettes per day) had a significantly greater impact than moderate
smoking (< 20 cigarettes per day) [18]. There are currently no studies on the impact of smoking cessation on the chances
of giving birth to a live infant for couples with RM.
3.2 Genetic factors
3.2.1 Chromosomal anomalies
Embryonic/fetal chromosomal abnormalities are the most common cause of spontaneous
miscarriage [19], [20]. The earlier the miscarriage occurs, the more likely that an embryonic/fetal chromosomal
anomaly was present [21]. The risk of embryonic/fetal trisomy resulting from chromosomal aberrations increases
with higher maternal age. The most common cause of miscarriage is trisomy 16, followed
by trisomy 22. Triploidy was detected in approximately 15% of cytogenetically abnormal
fetuses. Monosomy X is responsible for approximately 20% of miscarriages which occur
in the first trimester of pregnancy. No association with maternal age has been found
for monosomy X, polyploidy or structural chromosomal disorders [22]. A balanced chromosomal abnormality in one of the partners was confirmed in around
4 – 5% of couples who had 2 or more miscarriages [23]. In around 1% of pregnancies, an unbalanced set of chromosomes was detected during
prenatal diagnostic procedures or after the birth [24], [25].
|
Consensus-based Recommendation 3-2.E1
|
|
Expert consensus
|
Level of consensus ++
|
|
Cytogenetic analysis must be done if a woman experiences recurrent miscarriages. This
can be done either by chromosome analysis of both partners prior to conception or
using tissue samples from the miscarried fetus.
|
It is not possible to carry out standard chromosomal analysis in around 18% of miscarried
fetuses, and array analysis cannot be done in around 5% of miscarried fetuses [26]. Overall, molecular cytogenetic analysis only detected additional chromosomal disorders
in around 5% of cases, and the routine use of array analysis to identify the cause
of miscarriage is therefore not useful at present [26].
|
Consensus-based Recommendation 3-2.E2
|
|
Expert consensus
|
Level of consensus +++
|
|
Both partners must undergo cytogenetic testing if a structural chromosomal disorder
is detected in the tissue of the miscarried fetus. The couple must be informed of
the findings during genetic counselling carried out by a specialist for human genetics
or a physician with the relevant qualifications in accordance with national legal
regulations.
|
|
Consensus-based Statement 3-2.S1
|
|
Expert consensus
|
Level of consensus ++
|
|
If a balanced chromosomal aberration is detected in one of the parents, the risk of
miscarriage or of giving birth to an infant with a chromosomal disorder increases,
depending on the chromosomes involved. This will affect antenatal diagnostic procedures
in any further pregnancies.
|
3.2.2 Monogenetic disease
X-linked dominant disorders that are lethal in males have an increased risk of miscarriage.
But autosomal dominant and recessive disorders with severe malformations can also
result in increased intrauterine mortality. In these cases, examination of the fetus
should include genetic and pathological testing, particularly if the disorder was
not identified prenatally.
|
Consensus-based Recommendation 3-2.E3
|
|
Expert consensus
|
Level of consensus ++
|
|
If there are indications that monogenetic disease may be the cause of miscarriage,
genetic counselling must include genetic testing.
|
3.2.3 Results of association studies
Numerous studies suggest possible maternal, paternal or fetal genetic effects, but
these appear to have very little impact on the risk of miscarriage [27].
|
Consensus-based Recommendation 3-2.E4
|
|
Expert consensus
|
Level of consensus +++
|
|
Molecular genetic analysis for gene variants detected in association studies is not
recommended for couples with RM.
|
3.2.5 Therapeutic options
It is not possible to treat the causes of chromosomal disorders. Previous studies
have not shown that PGD after IVF results in an increased rate of live births in women
with RM compared to spontaneous pregnancies, not even for couples who are genetically
at risk because one partner has a balanced chromosomal aberration. Neither the ESHRE
and RCOG guidelines nor the ASRM Statement recommend preimplantation genetic diagnosis
for couples with RM.
|
Consensus-based Recommendation 3-2.E5
|
|
Expert consensus
|
Level of consensus +++
|
|
Preimplantation genetic diagnosis to prevent miscarriage is not recommended for couples
with RM who have no familial chromosomal disorder and no monogenetic disease.
|
3.3 Anatomical factors
3.3.1 Diagnosis of anatomical factors
3.3.1.1 Congenital malformations
Hysteroscopic examinations of patients who had 2, 3 and ≥ 4 consecutive miscarriages
found no difference in the prevalence of congenital (uterine malformations) or acquired
(adhesions, polyps, submucosal fibroids) intrauterine pathologies [28]. The increased probability of miscarriage in women with subseptate uterus is well
known, but the cause of this association is unknown [29]. It is not clear whether there is an association between RM and other uterine malformations
such as arcuate uterus or bicornuate uterus. Ludwin et al. [30] reported significantly better diagnostic results when using sonohysterography (SHG)
to diagnose congenital uterine malformations compared with hysterosalpingography (HSG)
or hysteroscopy. But it is difficult to evaluate statements comparing diagnostic methods,
because even when hysteroscopy videos were presented to experienced international
observers, interobserver agreement was found to be poor [31]. When diagnosing uterine malformations, the decision whether to use hysteroscopy
– possibly in combination with laparoscopy or 3D sonography – must be made on an individual
basis [32]. 3D sonography is recommended for the diagnostic workup of uterine malformations
in high-risk populations and MRI and endoscopic examinations are recommended for diagnostic
problems or suspected complex malformations [33].
3.3.1.2 Acquired malformations
Although the study populations consisted only of women undergoing IVF, a meta-analysis
of 19 observational studies showed a higher but not statistically significant rate
of miscarriages in women with intramural fibroids and no submucosal involvement (relative
risk [RR] 1.24; 95% CI: 0.99 – 1.57) [34]. In an evaluation of retrospective and prospective data of patients with RM, the
incidence of submucosal fibroids was 2.6% (25/966) [35]. These study data suggest an association between submucosal fibroids and the occurrence
of miscarriage, but the quality of the data is poor. A Cochrane analysis which only
included a few studies showed no significant reduction in the risk of miscarriage
after uterine fibroids had been resected (intramural: OR 0.89, 95% CI: 0.14 – 5.48;
submucosal: OR 0.63, 95% CI: 0.09 – 4.40) [36].
It is not clear whether – as with submucosal fibroids – intracavitary polyps also
increase the risk of miscarriage.
|
Consensus-based Recommendation 3-3.E6
|
|
Expert consensus
|
Level of consensus +++
|
|
Vaginal sonography and/or hysteroscopy is recommended in women with RM to rule out
uterine malformations, submucosal uterine fibroids and polyps. Hysteroscopy is recommended
to rule out intrauterine adhesions.
|
3.3.2 Treatment for anatomical factors
Hysteroscopic septum dissection is recommended for women with RM and septate uterus
[37]. A meta-analysis carried out in 2017 showed that no randomized studies on the therapeutic
effect of septum dissection have been carried out to date [38]. But retrospective uncontrolled studies suggest that the surgical intervention is
beneficial. Postoperative healing takes about 2 months [39], and there are no reasons to avoid pregnancy thereafter. Surgical intervention is
not indicated for other uterine malformations such as bicornuate uterus, uterus didelphys
and arcuate uterus [40] – [42].
|
Consensus-based Recommendation 3-3.E7
|
|
Expert consensus
|
Level of consensus +++
|
|
Hysteroscopic septum resection is recommended to treat women with RM and septate uterus.
|
Hysteroscopic adhesiolysis is the therapy of choice to treat intrauterine adhesions
[43]. It is not clear whether or to what extent intrauterine adhesions affect the risk
of miscarriage and whether adhesiolysis will reduce the risk of RM.
|
Consensus-based Recommendation 3-3.E8
|
|
Expert consensus
|
Level of consensus ++
|
|
Hysteroscopic adhesiolysis is recommended to treat women with RM and intrauterine
adhesions.
|
There are no randomized studies on the therapeutic benefits of fibroid resection in
women with RM.
|
Consensus-based Recommendation 3-3.E9
|
|
Expert consensus
|
Level of consensus ++
|
|
Surgical resection should be performed in women with RM and submucosal fibroids.
|
A meta-analysis showed that hysteroscopic resection of intrauterine polyps visible
on ultrasound carried out before intrauterine insemination can increase clinical pregnancy
rates [44]. The resection of persistent polyps can be considered if there is no other explanation
for RM.
|
Consensus-based Recommendation 3-3.E10
|
|
Expert consensus
|
Level of consensus ++
|
|
Hysteroscopic resection should be carried out to prevent miscarriage in women with
RM and persistent polyps.
|
3.4 Microbiological factors
3.4.1 Diagnostic workup of microbiological factors
Because the association between infections and RM is unclear, general screening for
vaginal infections which goes beyond the routine screening carried out as part of
prenatal care is not recommended. However, chronic endometritis, as evidenced by the
finding of plasma cells in the endometrial biopsy, was detected in 7 – 67% of otherwise
asymptomatic women with RM and in 30 – 66% of women with recurrent implantation failure
[45], [46], [47], [48], [49]. Endometrial biopsy may be performed in women with RM to exclude chronic endometritis
(supported by immunohistochemical staining for the plasma cell surface antigen CD138).
|
Consensus-based Recommendation 3-4.E11
|
|
Expert consensus
|
Level of consensus +++
|
|
Infectious screening using vaginal swab specimens is not recommended in asymptomatic
women with RM.
|
|
Consensus-based Recommendation 3-4.E12
|
|
Expert consensus
|
Level of consensus ++
|
|
An endometrial biopsy may be performed in women with RM to rule out chronic endometritis
(supported by immunohistochemical staining for the plasma cell surface antigen CD138).
|
3.4.2 Treatment for microbiological factors
Pregnant women suspected of having a vaginal infection should receive proper testing
and treatment [50], [51]. Antibiotic therapy with doxycycline (e.g. 200 mg 1 – 0 – 0 for 14 days) is indicated
for chronic endometritis; in the event of persistent endometritis as evidenced by
the persistence of plasma cells, treatment can consist of ciprofloxacin with/without
metronidazole [45].
|
Consensus-based Recommendation 3-4.E13
|
|
Expert consensus
|
Level of consensus ++
|
|
Antibiotic therapy may be administered to women with RM and chronic endometritis to
prevent miscarriage.
|
3.5 Endocrine factors
3.5.1 Diagnostic workup of endocrine factors
According to a retrospective analysis, manifest hyperthyroidism is associated with
increased miscarriage rates [52]. This also applies to manifest hypothyroidism. It is still unclear, however, whether
latent hypothyroidism (i.e. increases in TSH concentrations despite thyroid hormone
concentrations within normal ranges) also increases the risk of miscarriage. A meta-analysis
of two studies reported that the LBR was not lower for women with RM and TSH concentrations
of > 2.5 mU/L [53]. Increased levels of thyroid hormone autoantibodies appear to be associated with
higher rates of spontaneous miscarriage [54]. PCOS, hyperandrogenemia (which is often PCOS-related [55]), insulin resistance [56], [57] and diabetes [58] are all associated with a higher tendency to miscarry. PCOS per se is not a predictive
factor for miscarriage or RM [59], whereas obesity per se appears to increase the rate of miscarriages.
|
Consensus-based Recommendation 3-5.E14
|
|
Expert consensus
|
Level of consensus ++
|
|
An endocrine workup determining TSH levels is recommended in women with RM. If TSH
levels are found to be abnormal, fT3, fT4 and thyroid hormone autoantibody concentrations
must also be determined.
|
|
Consensus-based Recommendation 3-5.E15
|
|
Expert consensus
|
Level of consensus +++
|
|
The BMI of women with RM should be determined. Women with a BMI ≥ 30 kg/m2 may be investigated further to determine whether they have a metabolic syndrome.
|
3.5.2 Treatment for endocrine factors
It is important to diagnose and treat manifest hyperthyroidism or hypothyroidism.
A meta-analysis of studies of IVF patients with increased levels of thyroid hormone
autoantibodies (RM was no inclusion criterion) and pregnant women with higher levels
of TPO antibodies showed that substitution of thyroid hormones decreased the rate
of miscarriages [54]. No statement was made about the rate of live births. However, other studies such
as the study by Negro et al. published in 2016 [60] were unable to demonstrate the effect. It is therefore possible that patients with
RM and TPO autoantibodies benefit from the substitution of thyroid hormones in terms
of a lower rate of miscarriages, but currently there is no data specifically on patients
with RM.
|
Consensus-based Recommendation 3-5.E16
|
|
Expert consensus
|
Level of consensus +++
|
|
Manifest hypothyroidism or hyperthyroidism must be treated before conception.
|
|
Consensus-based Recommendation 3-5.E17
|
|
Expert consensus
|
Level of consensus ++
|
|
Thyroid hormone substitution therapy can be administered to prevent miscarriage in
women with RM and latent hypothyroidism, i.e. pathologically increased TSH concentrations
despite fT3 and fT4 concentrations within normal ranges or the presence of TPO autoantibodies.
|
A meta-analysis found that the administration of metformin had no effect on the risk
of sporadic miscarriage [61], and the guideline can therefore not make any recommendation regarding the administration
of metformin.
There are many medical reasons why women with a high BMI should lose weight (cf. the
S3 guideline on “Gestational Diabetes”, AWMF guideline 057/008). A Danish cohort study
[62] showed that the risk of miscarriage increases for women with a BMI ≥ 30 kg/m2 (OR 1.23; 95% CI: 0.98 – 1.54).
|
Consensus-based Recommendation 3-5.E18
|
|
Expert consensus
|
Level of consensus +++
|
|
Women with RM and a high body mass index should be encouraged to lose weight.
|
3.6 Psychological factors
Evidence-based medicine has not been able to show that RM can be directly caused by
psychological factors such as stress alone [10], [63], [64].
|
Consensus-based Recommendation 3-6.E19
|
|
Expert consensus
|
Level of consensus +++
|
|
Women with a prior history of mental illness, women who are involuntarily childless,
and women who lack or have only limited social resources or are struggling with feelings
of guilt related to processing their experience of RM must be given information about
psychosocial assistance and support (including self-help groups and internet forums).
|
|
Consensus-based Recommendation 3-6.E20
|
|
Expert consensus
|
Level of consensus +++
|
|
A psychotherapist must be called in if there is a suspicion that the patient is suffering
from reactive depression following RM to determine whether the affected patient/couple
require(s) further treatment.
|
|
Consensus-based Statement 3-6.S2
|
|
Expert consensus
|
Level of consensus +++
|
|
The efficacy of “tender loving care” as a therapeutic intervention to prevent miscarriage
in women with RM is not proven. However, tender loving care can provide psychological
support.
|
3.7 Immunological factors
3.7.1 Diagnosis of immunological factors
3.7.1.1 Alloimmune factors
Activation of the immune system (particularly of the Th1 response) results in unfavorable
conditions for implantation and is associated with an increased probability of RM
[51], [65], [66], [67], [68], [69]. It has not yet been clearly proven that an increase in the Th1/Th2 ratio or T4/T8
index leads to an increased risk of miscarriage [51], [66], [70], [71], [72], [73]. Several studies have pointed to an increase in natural killer cells in peripheral
blood (pNK cells) in patients with RM compared to healthy controls [74], [75], [76], [77]. Recent studies have also pointed to a significant increase in uterine natural killer
cells (uNK cells) in patients with idiopathic RM [78], [79].
|
Consensus-based Recommendation 3-7.E21
|
|
Expert consensus
|
Level of consensus ++
|
|
Alloimmune investigations such as determining the Th1/Th2 ratio or the T4/T8 index,
analysis of pNK and/or uNK cells, NK toxicity tests, lymphocyte function tests, molecular
genetic testing for non-classical HLA groups (Ib) or KIR receptor families and determination
of HLA should not be done in women with RM outside clinical studies, unless there
is evidence of a pre-existing autoimmune disorder.
|
3.7.1.2 Autoimmune factors
Although the data are not consistent, the majority of studies report increased ANA
titer levels in women with RM [80], [81], [82], [83], [84], [85], [86]. Celiac disease is characterized by gluten sensitivity; its association with RM
is still controversially discussed. Testing for immunoglobulin A (IgA) antibodies
against tissue transglutaminase can be done in women with a history of food sensitivity
(food intolerances, irregular bowel motions) and RM, followed by biopsy of the small
intestine if the findings are positive [87].
Testing for antiphospholipid syndrome using clinical and laboratory parameters is
recommended in women with RM ([Fig. 1]). Non-specific antibodies against anionic phospholipids such as cardiolipins and
β2 glycoproteins, also known as antiphospholipid antibodies (aPLAb) have been detected
in some women with RM. However, according to the definition given in [Fig. 1], antiphospholipid (aPL) syndrome is only present if both clinical and laboratory
criteria are fulfilled. Between 2% and 15% of women with RM suffer from aPL syndrome
[88]. During the diagnostic workup, it is important to determine whether aPL antibody
titer is still moderate to high at the 12-week follow-up after first determining the
titer, i.e., whether it is in the > 99th percentile compared to unremarkable controls
[89].
Fig. 1 Diagnostic criteria for antiphospholipid syndrome [89]. Clinical and laboratory criteria can be present either in combination or singly.
By definition, however, at least one clinical and one laboratory criterion must be present to make a diagnosis of antiphospholipid syndrome. [rerif]
A few studies have considered the possibility of so-called “non-criteria” aPL syndrome,
particularly when manifestations (livedo reticularis, ulcerations, renal microangiopathies,
neurological disorders and cardiac manifestations) are present and the diagnostic
criteria for classic aPL syndrome are not fulfilled or only in part (e.g., low aPLAb
titer or s/p 2 miscarriages) [89].
|
Consensus-based Recommendation 3-7.E22
|
|
Expert consensus
|
Level of consensus +++
|
|
Testing for antiphospholipid syndrome based on clinical and laboratory parameters
([Fig. 1]) is recommended for women with RM.
|
|
Consensus-based Recommendation 3-7.E23
|
|
Expert consensus
|
Level of consensus +++
|
|
Interdisciplinary care must be offered to women with RM and an autoimmune disease
already present prior to conception.
|
|
Consensus-based Recommendation 3-7.E24
|
|
Expert consensus
|
Level of consensus ++
|
|
Testing for non-criteria antiphospholipid syndrome based on clinical and laboratory
parameters should be done in women with RM, particularly if clinical manifestations
are present (livedo reticularis, ulcerations, renal microangiopathies, neurological
disorders and cardiac manifestations).
|
3.7.2 Treatment for immunological factors
3.7.2.1 Glucocorticoids
The results of existing clinical studies which administered glucocorticoids to women
with RM are inconsistent [90], [91], [92], [93]. Treatment with glucocorticoids – particularly at higher doses – can induce side
effects such as gestational diabetes, arterial hypertension, preterm birth, low birthweight
(SGA) and disorders of neurological development in the infant [94] – [96]. This type of treatment should therefore be reserved for patients with pre-existing
autoimmune diseases which require therapy with glucocorticoids even during pregnancy.
|
Consensus-based Recommendation 3-7.E25
|
|
Expert consensus
|
Level of consensus ++
|
|
Glucocorticoids must not be administered outside clinical studies as prophylaxis to
prevent miscarriage in women with RM but without pre-existing autoimmune disease.
|
3.7.2.2 Intravenous immunoglobulins
A few studies have pointed out that intravenous administration of immunoglobulins
(IVIG) reduces the concentrations and activities of natural killer cells in peripheral
blood and affects Th1-mediated immune response [97]. The studies were very heterogeneous and the majority were done in patients with
idiopathic RM but without a specific immunological diagnostic workup prior to starting
therapy. The data is inconsistent [97], [98], [99], [100].
A recent meta-analysis which included 11 randomized studies of the type described
above found no significantly higher LBR for the group of patients who received IVIG
(RR 0.92; 95% CI: 0.75 – 1.12) compared to placebo or standard care [101]. A subgroup analysis showed a trend towards a benefit from IVIG in the cohort of
patients with secondary RM compared to placebo (RR for no live births 0.77; 95% CI:
0.58 – 1.02; p = 0.06). There are currently no clearly defined indications for the
administration of immunoglobulins, and they should therefore not be administered outside
clinical studies. Side effects which can even include anaphylactic shock and the transmission
of infectious pathogens are rare, but the incidence of occurrence is significantly
higher in the verum group compared to controls.
|
Consensus-based Recommendation 3-7.E26
|
|
Expert consensus
|
Level of consensus ++
|
|
Therapy with intravenous immunoglobulins to prevent miscarriage should not be given
to women with RM outside clinical studies.
|
3.7.2.3 Lipid infusion
Current studies showed that soybean-oil-based lipid infusions reduced both NK cell
activity and the formation of pro-inflammatory cytokines [102], [103], [104], [105], [106]. Small observational studies have shown that lipid infusions administered to women
with RM or implantation failure and increased NK cell activity achieved the same live
birth rates as treatment with IVIG [107], [108], [109]. A randomized placebo-controlled double-blind study carried out in Egypt investigated
the impact of a single lipid infusion in a cohort von 296 women (with no tubal pathology
and aged less than 40 years) undergoing IVF. The investigated women all had ≥ 3 idiopathic
RM (consecutive clinical miscarriages after spontaneous conception or IVF/ICSI) and
had elevated levels of peripheral blood NK cells (pNK cells > 12%) [110]. No significant difference in the rate of biochemical pregnancies was found between
groups, but the rate of intact pregnancies > 12th GW and the rate of live births (37.5
vs. 22.4%, respectively; p = 0.005) was significantly higher in the group which had
received a lipid infusion.
|
Consensus-based Recommendation 3-7.E27
|
|
Expert consensus
|
Level of consensus +++
|
|
Lipid infusion to prevent miscarriage should not be administered to women with RM
outside clinical studies.
|
3.7.2.4 Allogeneic lymphocyte transfer (LIT)
The transfer of allogeneic lymphocytes (usually paternal lymphocytes, rarely donor
lymphocytes, also known as lymphocyte immunization therapy) is a means of readying
the maternal immune system to cope with the embryoʼs foreign antigens (HLA). The data
on the uses of this therapy in women with RM is inconsistent. Two recent meta-analyses
pointed to higher LBR in patients with idiopathic RM who received LIT. However, these
meta-analyses were strongly influenced by the weighting of an Asian study, published
in 2013, which showed a positive effect [111] – [113]. Older studies found no benefit [114], [115], [116], meaning that, here too, further studies will be necessary. It should be noted that
the transfusion of blood products can lead to complications (e.g., transmission of
infections, formation of irregular autoantibodies, induction of autoimmune disorders).
|
Consensus-based Recommendation 3-7.E28
|
|
Expert consensus
|
Level of consensus +
|
|
Allogeneic lymphocyte transfer to prevent miscarriage should not be carried out in
women with RM outside clinical studies.
|
3.7.2.5 TNF-α receptor blockers
Subgroups of patients with RM have been reported to have increased TNF-α concentrations,
abnormal TNF-α/IL-10 ratios or numbers of TNF-α-producing CD3+CD4+ lymphocytes, and
these subgroups could benefit from the administration of TNF-α receptor blockers (e.g.,
adalimumab or infliximab) [100], [117]. However, only one retrospective study has looked at this issue to date. In addition
to TNF-α receptor blockers, the study also used low-dose acetylsalicylic acid (ASA),
low-molecular-weight heparin (LMWH) and immunoglobulins [100]. Well-known side effects ranged from skin reactions to infections and even rare
adverse events such as drug-induced lupus [118]. There are also concerns regarding the possible induction of malignant disease by
TNF-α blockers [119]. At present, the administration of TNF-α receptor blockers should be reserved for
controlled clinical studies and for specific conditions (e.g., autoimmune diseases
such as Crohnʼs disease or chronic polyarthritis).
|
Consensus-based Recommendation 3-7.E29
|
|
Expert consensus
|
Level of consensus ++
|
|
Therapy with TNF-α receptor blockers should not be given to women with RM outside
clinical studies.
|
3.7.2.6 Treatment for autoimmune factors
Because of the inconsistent data on the prevalence of antinuclear antibodies in women
with RM, the current therapy strategies (ASA, glucocorticoids, low-molecular-weight
heparin) are inconsistent and the guideline cannot offer any recommendations. There
is currently only one retrospective study on the therapy of women with celiac disease
and RM (n = 13) [120]. The women in the study benefitted from a gluten-free diet.
Therapy with low-dose acetylsalicylic acid and low-molecular-weight heparin is recommended
for women with RM and antiphospholipid syndrome. Treatment with acetylsalicylic acid
and heparin must be initiated as soon as the pregnancy test is positive. Aspirin administration
must continue until GW 34 + 0, with heparin administration continuing for at least
6 weeks post partum. Numerous studies have shown that patients with RM and APS benefited
from the administration of aspirin (50 – 100 mg/d) and low-molecular-weight heparin
in prophylactic doses [121], [122], [123], [124], [125]. In contrast to the administration of LMWH and aspirin, other therapeutic approaches
such as the administration of corticoids, immunoglobulins or aspirin alone did not
result in any significant improvement in the LBR of patients with RM and APS [121].
Based on current studies, non-criteria aPL syndrome should be treated in exactly the
same way, as a few studies have indicated a potential benefit from the administration
of LMWH in combination with ASA [126], [127], [128], [129], [130].
|
Consensus-based Recommendation 3-7.E30
|
|
Expert consensus
|
Level of consensus +++
|
|
Therapy with low-dose acetylsalicylic acid and low-molecular-weight heparin is recommended
for women with RM and antiphospholipid syndrome. Treatment with acetylsalicylic acid
and heparin must be initiated as soon as the pregnancy test is positive. Aspirin administration
must continue until GW 34 + 0, with heparin administration continuing for at least
6 weeks post partum.
|
|
Consensus-based Recommendation 3-7.E31
|
|
Expert consensus
|
Level of consensus +++
|
|
Therapy with low-dose acetylsalicylic acid and low-molecular-weight heparin is recommended
for women with RM and non-criteria antiphospholipid syndrome. Treatment with acetylsalicylic
acid and heparin must be initiated as soon as the pregnancy test is positive. Aspirin
administration must continue until GW 34 + 0, with heparin administration continuing
for at least 6 weeks post partum.
|
3.8 Coagulation
3.8.1 Diagnosis of congenital thrombophilic factors
Hereditary thrombophilic parameters are present in up to 15% of the Caucasian population
[131]. In recent years, the assessment of maternal thrombophilia as an important risk
factor for RM has significantly changed. Thrombophilia testing done only to prevent
miscarriage is not recommended. International guidelines (ASRM, ACCP, RCOG) do not
recommend routine testing for hereditary thrombophilia in women with RM [1], [3], [132], [133], [134]. The RCOG guideline considers testing for maternal hereditary thrombophilia to be
only indicated in the context of scientific studies [133]. The ASRM recommendations propose thrombophilia testing for women with RM only if
they have a medical or familial history of thromboembolic events [1], [3], [132], [133], [134].
Abnormalities in thrombophilic parameters may be an indication to treat pregnant women
for medical reasons (prevention of thrombembolic events). Anticoagulation therapy
to prevent maternal thromboembolism may be justified in pregnant women who have an
increased risk of thromboembolic events (VTE) due to specific conditions (e.g., antithrombin
deficiency, homozygous FVL mutation, combined heterozygous FVL and PT mutation, etc.)
and in women with additional risk factors for VTE in pregnancy (immobilization, surgery,
excessive weight gain, etc.) (ACOG 2013, AWMF 2015).
|
Consensus-based Recommendation 3-8.E32
|
|
Expert consensus
|
Level of consensus ++
|
|
Testing for thrombophilia to prevent miscarriage is not recommended.
|
|
Consensus-based Recommendation 3-8.E33
|
|
Expert consensus
|
Level of consensus +++
|
|
Women with RM who are at risk of thromboembolic events must be tested for thrombophilia.
This includes determination of antithrombin activity and plasma protein C and protein
S levels and molecular genetic analysis for factor V Leiden mutation and prothrombin
G20210A mutation.
|
3.8.2 Treatment for women at risk of thrombophilic events
3.8.2.1 Heparin
Unfractionated and low-molecular-weight heparins differ with regard to their molecular
weight, plasma protein binding, biological half-life and rate of side effects. In
addition to their anticoagulation effect, they also have numerous effects at the molecular
level of the embryo-maternal interface which are still not completely understood [135]. No heparins cross the placenta. The administration of low-molecular-weight heparin(s)
during pregnancy is considered comparatively safe [136]. The administration of heparins in pregnancy represents an off-label use. If the
administration of heparin in pregnancy is indicated, low-molecular-weight heparins
should be used because of their superior side-effects profile and ease of administration
[132]. The enthusiasm at the turn of the century about the impact of prophylactic heparin
administration in women with RM (in whom APS had been excluded) on the prevention
of miscarriage could not be confirmed in either large prospective randomized studies
[137], [138], [139] or in more recent meta-analyses [140].
The general maternal administration of heparin in subsequent pregnancies to women
with RM without confirmed thrombophilia is not indicated because of the lack of proof
of efficacy [141] – [143]. There is also no evidence for a beneficial effect of administering heparin prior
to or during the conception period on the prevention of further miscarriages.
To what extent subgroups of patients (e.g., patients with confirmed hereditary thrombophilia)
actually benefit from the administration of heparin in subsequent pregnancies requires
further studies, such as the currently recruiting, multinational ALIFE2 trial [144]. At present, the general administration of heparin outside clinical studies for
the indication “prevention of miscarriage” alone is not indicated, even in thrombophilic
women with RM (in whom APS has not been confirmed) [132], [145].
|
Consensus-based Recommendation 3-8.E34
|
|
Expert consensus
|
Level of consensus ++
|
|
Treatment with heparin with the sole purpose of preventing miscarriage is not recommended
for women with RM. This also applies to women with hereditary thrombophilia.
|
|
Consensus-based Recommendation 3-8.E35
|
|
Expert consensus
|
Level of consensus ++
|
|
Thromboprophylaxis for maternal indication should be given during pregnancy to women
with RM and an increased risk of thromboembolic events.
|
3.8.2.2 Acetylsalicylic acid (ASA)
The use of ASA in pregnancy to prevent miscarriage represents an off-label use. The
administration of ASA in low doses starting in the 1st trimester of pregnancy reduces
the risk of placenta-associated complications in late pregnancy [146], although it has not been possible to confirm any protective effect on the rate
of miscarriages. The prospective randomized ALIFE trial in women with idiopathic RM
reported that administration of aspirin prior to conception (80 mg/day) did not reduce
the rate of miscarriages compared to placebo [138]. A systematic Cochrane meta-analysis found no benefit from the prophylactic administration
of ASA in women with idiopathic RM (RR 0.94; 95% CI: 0.80 – 1.11) [147]. This also applies to the administration of aspirin prior to conception [148].
|
Consensus-based Recommendation 3-8.E36
|
|
Expert consensus
|
Level of consensus +++
|
|
Acetylsalicylic acid therapy to prevent miscarriage is not recommended for women with
RM.
|
3.8.3 Monitoring during pregnancy – D dimers
|
Consensus-based Recommendation 3-8.E37
|
|
Expert consensus
|
Level of consensus +++
|
|
Monitoring of plasma coagulation markers (D dimers, prothrombin fragments, etc.) during
pregnancy is not recommended in women with RM. Determination of these markers must
not be used as an indication to initiate therapy to prevent miscarriage.
|
3.9 Idiopathic RM
3.9.1 Diagnosis of idiopathic RM
Idiopathic RM is present if the criteria for a diagnosis of RM are met, and genetic,
anatomical, endocrine, established immunological and hemostatic factors have been
ruled out. The percentage of idiopathic RM in the total population of women with RM
is high and amounts to 50 – 75% [2]. The percentage of live births for women with idiopathic RM who did not receive
treatment is 35 – 85% [138], [149].
|
Consensus-based Recommendation 3-9.E38
|
|
Expert consensus
|
Level of consensus +++
|
|
The term “idiopathic RM” is only used if the diagnostic workup described in this guideline
is carried out and no cause of RM has been found.
|
3.9.2 Therapy for idiopathic RM
A Cochrane meta-analysis of nine randomized studies which included 1228 women with
idiopathic RM who had had at least two spontaneous miscarriages found no statistically
significant effect of ASA with/without heparin on the LBR compared to placebo [147]. A randomized study of 364 women with idiopathic RM found that ASA administration
had no impact on LBR compared to ASA and nadroparin or placebo [138]. Another meta-analysis of six randomized studies of 907 women with idiopathic RM
also found no improvement in live birth rates following the administration of ASA
and heparin [147].
|
Consensus-based Recommendation 3-9.E39
|
|
Expert consensus
|
Level of consensus ++
|
|
Treatment with acetylsalicylic acid with or without additional heparin to prevent
miscarriage is not recommended in women with idiopathic RM.
|
A meta-analysis, published in 2017, of 10 randomized studies of 1586 women with idiopathic
RM reported a positive effect following therapy with progestogens in the first trimester
of pregnancy, both in terms of the rate of miscarriages (RR 0.72; 95% CI: 0.53 – 0.97)
and the rate of live births (RR 1.07; 95% CI: 1.02 – 1.15). Synthetic progestogens,
but not natural progesterone, were associated with a lower risk of recurrent miscarriage
[150]. Synthetic progestogens can therefore be administered to women with idiopathic RM
in the first trimester of pregnancy to prevent miscarriage. However, the optimal time
for administration and the optimal dosage of the progestin are not yet clear.
In the PROMISE trial, a total of 836 women with idiopathic RM were randomized to receive
either placebo or 400 mg micronized progesterone applied by vaginal suppository [151]. Treatment was initiated soon after positive urinary pregnancy test and continued
up to and including the 12th week of gestation. The LBR was the same in both study
arms (63 and 66%, respectively). However, a randomized study of 700 women with RM
carried out in Egypt reported significantly higher live birth rates compared to placebo
(91 vs. 77%) for 2 × 400 mg progesterone administered intravaginally, starting in
the luteal phase [152].
|
Consensus-based Recommendation 3-9.E40
|
|
Expert consensus
|
Level of consensus ++
|
|
Treatment with natural micronized progesterone in the first trimester of pregnancy
to prevent miscarriage is not recommended for women with idiopathic RM.
|
|
Consensus-based Recommendation 3-9.E40
|
|
Expert consensus
|
Level of consensus ++
|
|
Synthetic progestogens can be administered to women with idiopathic RM in the first
trimester of pregnancy to prevent miscarriage.
|
The effect of human chorionic gonadotropin (hCG) in doses of 5000 – 10 000 IE in the
first and second trimester of pregnancy was evaluated in five randomized studies of
596 women with RM, including women with idiopathic RM. A Cochrane meta-analysis of
these five studies found that the administration of hCG led to a significant reduction
in the frequency of miscarriage. However, this positive effect was no longer statistically
significant when the analysis was done without the two methodologically weaker studies
(OR 0.74; 95% CI: 0.44 – 1.23) [153]. The studies did not include data on LBR. There was no separate subgroup analysis
for women with idiopathic RM. It is therefore currently not possible to recommend
the administration of hCG to treat women with RM.
Scarpellini et al. carried out a randomized study in 68 women with RM who had previously
had at least 4 consecutive spontaneous miscarriages. The women were randomized to
receive either placebo or rh-G-CSF (1 µg/kg/day) starting on the 6th day after ovulation
[154]. LBR for the active study arm was 83% (29/35) compared to 48% in the control group
(16/33). In a retrospective cohort study Santjohanser et al. reported on 127 women
with RM (for the purposes of that study, RM was defined as at least 2 spontaneous
early miscarriages) who had IVF/ICSI [155]. Forty-nine of the women received either rh-G-CSF at a dose of 34 million units/week
or 2 × 13 million units/week until the 12th week of gestation. The LBR was 32% higher
following G-CSF administration compared to 13 – 14% for other patient groups. As a
number of issues (e.g., the optimal dose) relating to G-CSF therapy are still unresolved,
G-CSF should only be administered in the context of a clinical study.
|
Consensus-based Recommendation 3-9.E41
|
|
Expert consensus
|
Level of consensus ++
|
|
With the exception of clinical trials, administration of G-CSF to prevent miscarriage
is not recommended for women with idiopathic RM.
|
|
Consensus-based Recommendation 3-9.E42
|
|
Expert consensus
|
Level of consensus +++
|
|
Treatment with acetylsalicylic acid with or without additional heparin to prevent
miscarriage is not recommended in women with idiopathic RM.
|
