Are imprinting disorders more

prevalent after human in vitro

fertilization or intracytoplasmic
sperm injection?
Jan P. W. Vermeiden, Ph.D. and Rob E. Bernardus, M.D., Ph.D.
^ s Fertility Center, Wolvega, the Netherlands
Nij Barrahu

Objective: To review the literature and present original data to answer the question of whether in vitro fertilization (IVF) or intracy-
toplasmic sperm injection (ICSI) is associated with an increase in imprinted diseases in offspring. If the answer is positive, to investigate
whether there is a causal relationship between IVF or ICSI and the imprinted diseases.
Design: Review study.
Result(s): Eight epidemiologic studies were suitable to calculate the weighted relative risk for the birth of a child with Beckwith-
Wiedemann syndrome following IVF or ICSI compared with the risk in the normal population. This relative risk was 5.2 (95% CI
1.67.4). In one study the relative risk was corrected for parents fertility problems and no signicant association was found. Data
on the Silver-Russell syndrome are too sparse to draw conclusions, but a positive association with IVF or ICSI treatment is probable.
No signicant associations were found between the incidences of the Angelman and Prader-Willi syndromes and IVF or ICSI
treatments. Children with Prader-Willi syndrome or Angelman syndrome are more likely to be born to parents with fertility
problems. All retinoblastomas in children born after IVF or ICSI could be explained by de novo mutations in the RB1 gene and were
not associated with imprinted genes. Imprinted diseases result from methylation errors already present in sperms or oocytes. There
is no proof of a causal relationship between imprinted diseases and IVF or ICSI treatments.
Conclusion(s): Imprinting disorders are more prevalent after human IVF or ICSI. Future studies should correct for fertility problems in
the affected and comparison groups. It is highly improbable that assisted reproduction technol-
ogies cause imprinted diseases in humans. (Fertil Steril 2013;99:64251. 2013 by American
Society for Reproductive Medicine.) Use your smartphone
Key Words: Beckwith-Wiedemann syndrome, Silver-Russell syndrome, Angelman syndrome, to scan this QR code
Prader-Willi syndrome, retinoblastoma, IVF, ICSI and connect to the
discussion forum for
this article now.*
Discuss: You can discuss this article with its authors and with other ASRM members at http://
fertstertforum.com/vermeidenjp-ivf-icsi-imprinted-diseases/ * Download a free QR code scanner by searching for QR
scanner in your smartphones app store or app marketplace.

T
here are reports that in vitro fertil- the paternally derived copy. This mono- curs in the zygote as well as the de
ization (IVF) or intracytoplasmic allelic expression of imprinted genes is novo methylation during differentia-
sperm injection (ICSI) is associ- associated with epigenetic modica- tion of somatic cells through as yet
ated with imprinting disorders, and it is tions and DNA methylation and is mostly unknown mechanisms. In early
even presumed that IVF or ICSI can parent-of-origin specic. This parent- embryonic germ cell development, the
cause imprinted diseases. In the present specic modication pattern present parental patterns are erased, to be set
paper we aimed to evaluate the pub- in the germlines leads to differentially anew in adult life. The paternal imprint-
lished evidence on these problems. methylated regions (DMRs) after fertil- ing pattern is set during spermatogene-
Imprinted genes are transcribed ization. These stay more or less intact sis. The maternal pattern is set before,
from only one specic parental allele, after fertilization and escape the during, or after fertilization. This timing
either the maternally derived copy or genome-wide demethylation that oc- is gene specic. The DMRs originating
in the germ line are called primary
Received November 21, 2012; revised January 15, 2013; accepted January 22, 2013. DMRs. If paternal genes are silenced
J.P.W.V. has nothing to disclose. R.E.B. has nothing to disclose. due to a primary DMR, maternal genes
Reprint requests: Jan P. W. Vermeiden, Ph.D., Nij Barrahu ^ s Fertility Center, Heerenveenseweg 99b,
8471 ZA Wolvega, the Netherlands (E-mail: j.vermeiden@xs4all.nl). are predominantly or exclusively ex-
pressed, or vice versa. This distin-
Fertility and Sterility Vol. 99, No. 3, March 1, 2013 0015-0282/$36.00 guishes primary DMRs from secondary
Copyright 2013 American Society for Reproductive Medicine, Published by Elsevier Inc.
http://dx.doi.org/10.1016/j.fertnstert.2013.01.125 DMRs, which arise only after de novo

642 VOL. 99 NO. 3 / MARCH 1, 2013


methylation in the postimplantation embryo. Secondary
DMRs would be developmental stage, cell, and tissue
specic, and would in general not show allelic-specic
methylation (1).
In imprinted diseases, the monoallelic expression is dis-
turbed by aberrant methylation of the imprinting control re-
gion of a specic gene causing the disease. This can be either
hypomethylation or hypermethylation. The situation can be
complex. For example, imprinting of H19 (maternally ex-
pressed) is reciprocal with the imprinting of IGF2 (paternally
expressed). Hypermethylation of H19 is associated with the
Beckwith-Wiedemann syndrome (BWS) and hypomethyla-
tion with the Silver-Russell syndrome (SRS).
IVF and ICSI are well accepted assisted reproduction
techniques (ART) for subfertile couples. ART does not heal
the causes of subfertility but rather makes it possible for sub-
fertile couples to conceive. The rst IVF baby was born in
1978, and now, 35 years later, optimal culture conditions
have not been determined and we do not know the effect
of the treatment on the children; we do not know whether
and how the treatments effect the epigenome. Studies follow-
ing the health of children born after IVF or ICSI are reassur-
ing, but the mean values of several variables differ from the
general population: Slightly higher fasting glucose and
higher blood pressure have been observed compared with
those found in spontaneously conceived children. The indi-
vidual values of these variables are mostly within normal
range, indicating slight difference among IVF/ICSI children
(2). Children conceived and born in periods of famine and
stress also showed higher fasting glucose and increased blood
pressure. Before mechanisms behind these phenomena were
delineated the term programming was used (3, 4). Today, Several papers have been published with original data on
programming can be explained by adaptations of the
epigenome, as demonstrated by differential methylation of
various DMRs of IGF2/H19 (5). These epigenomic
adaptations are normal physiologic variations and are not
imprinted diseases.
Imprinted diseases are real diseases, as opposed to phys-
iologic adaptations to a changing environment. The rst en-
counter of ART with imprinted diseases was the large
offspring syndrome. Because of certain culture conditions
(high concentration of serum and coculture with somatic
cells), in vitro maturation of bovine oocytes and in vitro fer-
tilization of these oocytes resulted in large calves (compared
with 45 kg normal weight, up to 90 kg for the large calves).
It appeared that the imprinting control center of Igf2r was hy-
pomethylated, resulting in the overgrowth of the calves (6).
From 2002 on, alarming papers were published on an increase
of imprinted diseases because of IVF and ICSI (7), which raised
the question of whether IVF and ICSI caused the imprinted
diseases BWS, SRS, Angelman syndrome(AS), Prader-Willi
syndrome (PWS), and retinoblastoma. However, because of
the rarity of these diseases in the general population as well
as in children conceived by IVF or ICSI, and confusion in un-
derstanding the implications of programming and imprinted
diseases, and nally because studies were performed with
differing methodologies that produced conicting results,
the associations between imprinted diseases and ART are
not clear.
VOL. 99 NO. 3 / MARCH 1, 2013
Fertility and Sterility
BECKWITH-WIEDEMANN SYNDROME AND ART
BWS is a fetal overgrowth syndrome. The diagnosis is based
on patient physical characteristics; in 80% of all cases
conrmation by genetic molecular tests is possible. The ge-
netic errors are found on chromosome 11. The major causes
are: uniparental disomy (UPD) of H19 (20% of cases; UPD in-
volves both imprinting centers 1 and 2); hypomethylation of
LIT1 (maternally methylated, 50% of cases); hypermethyla-
tion of H19 (paternally methylated, 5% of cases; and micro-
deletions, translocation, inversion, duplications (each <1% of
cases). In 20% of cases the cause is unknown (8). In ART-
associated cases, almost all BWS was due to hypomethylation
of LIT1. BWS is a rare disease, with population prevalence es-
timated to be 1:14,50013,700 in Western Europe and North
America (8, 9). A prevalence of 1:287,000 has been reported in
Japan (10).
In the literature, the standard expression for hypomethy-
lation of imprinted genes such as LIT1 is loss of methylation.
Loss refers to a formerly methylated gene that subsequently
lost this methylation. However, it is very possible that the
gene was never methylated. Therefore neutral terms, such as
absence of methylation or hypomethylation, are preferred.
This absence can be caused either by loss or by failure of
methylation. The term hypomethylation is applied after as-
sessment of the methylation index and indicates that this in-
dex is low. Hypomethylation is a neutral term that describes
the concrete methylation status of a gene in a tissue or in
a cell suspension.
Epidemiologic Studies of BWS and ART
BWS and IVF/ICSI. Most of the papers used data from volun-
tary registries. Only three were based on complete registries
(1113). The differences in methodology did not allow for
proper meta-analysis, but we calculated a weighted relative
risk (RR) based on all available data.
DeBaun et al. (14) used the data of two voluntary regis-
tries of BWS. They estimated that BWS is at least 6
(95%
condence interval [CI] 2.0
18.2
) more prevalent in chil-
dren born after IVF/ICSI than in the general population. They
described the molecular basis of seven BWS children born af-
ter ART: Five had hypomethylation of LIT1, one had normal
LIT1 and normal H19, and one had methylation errors in
both LIT1 and H19. They did not report the incidence of
BWS in other forms of ART, such as intrauterine insemination
(IUI), donor insemination, and the use of fertility drugs.
In the United Kingdom several papers with new data were
published on BWS related to IVF/ICSI (9, 1517). These papers
were all based on voluntary registries. These papers repeated
some data from the same patients. For that reason, only one
paper was selected for the calculation of the weighted RR (16).
A questionnaire was sent to 213 families and a response
was received from 83. Of these, 79 had sporadic BWS, six
(7.6%) of these individuals were conceived with IVF or ICSI.
After correction for nonresponders at least 2.9% (95% CI
1.4%6.3%) of the BWS children were born after IVF or ICSI
(expected rate 0.8%; RR 3.6 (95% CI 1.67.8). Thus, even after
correction for nonresponders, there was a signicant
643
VIEWS AND REVIEWS
association between BWS and IVF/ICSI. Using the original
data, we calculated a RR of 9.5 (95% CI 4.320.1). We used
this gure to calculate the weighted RR of all epidemiologic
studies, because in all other studies no correction for nonre-
sponders was made. The cited paper (16) reported the molecu-
lar data of 8/11 patients. All eight had hypomethylation of
LIT1, but the type of treatment (IVF/ICSI or just fertility drugs)
was not indicated. Of the same group, the molecular data of 13
post-ICSI and 12 post-IVF BWS patients were reported (17):
24/25 had hypomethylation of LIT1. It is presumed that the
data of the six BWS IVF/ICSI patients of Sutcliffe et al. were
incorporated into these data (16).
Bowdin et al. (9) sent a questionnaire to 2,492 families
with IVF/ICSI children living in Ireland and central England;
1,542 responded, and one BWS child was reported. If the ex-
pected number of patients was calculated, using both the RR
of 9.5 and the corrected RR of 3.6 (16), the incidence of BWS
in this group would have been between 1:4,0001,500 and
0.41.0 BWS child could be expected. Thus the results are
within expected values, but the study lacked sufcient power
for a solid conclusion.
Gicquel et al. (11) based their work on a complete nation-
wide registry (France) of all BWS patients. The study period
was not dened, but six BWS patients were born after IVF
or ICSI; nationwide, 149 BWS children were born in the
same period, meaning that 4% of the BWS children were
born after IVF/ICSI, giving an RR of 3.2 (95% CI 1.46.8) com-
pared with the normal population.
Rossignol et al. (from the same group as Gicquel) pub-
lished the molecular results of new BWS patients after ART
(18). Together the result of 12 BWS patients after ART were
published. All 12 patients had hypomethylation of LIT1,
and two of them also had hypomethylation of H19.
Halliday et al. (19) published the results of BWS and IVF/
ICSI in the state of Victoria, Australia. Among 1,316,500
live births in Victoria from 1983 to 2003, 37 people with
BWS were detected, giving an overall BWS prevalence of
1:35,580 live births for this period. In the same period,
14,894 children were born by IVF/ICSI in the State of Victoria.
The incidence of BWS in these children was 4 out of 14,894
(1:3,723; RR 9.6, 95% CI 3.824.1). No gures of other fertility
treatments were given.
Kallen et al. (12) reviewed BWS registries in Sweden. In
the cohort of children born after IVF/ICSI, one child had
PWS and one had SRS. In this period (19822001) 16,280
children were born after IVF/ICSI. It was concluded that in
Sweden, IVF/ICSI was not associated with an increased inci-
dence of imprinted diseases. No epidemiologic data on BWS
births in the Swedish population were available, but 1 (2)
out of 16,280 could be expected (assuming that the incidence
in Sweden equals that of Western Europe [1:13,700]).
Lidegard et al. (13) reviewed the registries in Denmark. All
ART children and all congenital birth defects are centrally
registered in Denmark. In the 6,052 children born after IVF/
ICSI no BWS child was reported. It was concluded that ART
was not associated with an increased incidence of imprinted
diseases in Denmark. In fact, no BWS children were reported
in the database of all children born in Denmark. Because at
least 21 BWS children could have been expected (number of
644
children born [442,349] divided by the incidence of BWS
[13,700], minus 2
SD 32  11 21), these ndings are
difcult to understand.
Doornbos et al. (20) published the gures of imprinted
diseases and ART in The Netherlands during the period
19832003. The study was based on a voluntary registry. Tak-
ing the calculated incidence of BWS in the Netherlands, 24%
of the families with BWS children participated in the study.
Seventy-one families with a BWS child participated in the
study, of whom four children were born after IVF or ICSI.
The RR was 6.1 (95% CI 2.5.9) compared with the normal
population. Fifteen of these 71 families reported fertility prob-
lems (21.1%, compared with 5.6% in the Dutch population;
RR 4.6, 95% CI 2.35.6). After correction for fertility problems
in the group of parents with BWS children and for children
born in families with fertility problems of any kind in the nor-
mal population, there was no signicant increase of BWS
births in the group of IVF/ICSI children.
Hiura et al. (10) published a paper on the association be-
tween ART and imprinted diseases in Japan. That paper re-
ported an incidence of BWS of 1:287,000. The gure was
based on a questionnaire sent to 3,158 pediatric departments
in Japan. The published incidence of BWS in Western Europe
varies from 1:14,500 to 1:13,700. This could mean that the in-
cidence of BWS is approximately 20
lower in Japan than in
Western Europe. However, the authors indicated in the dis-
cussion of their manuscript that the response rate to the ques-
tionnaire was not optimal (10). They reported on six BWS
patients out of 70 conceived by IVF/ICSI, or 8.6%. Because
0.64%0.98% of the total number of babies born in Japan
were a result of IVF and ICSI, the RR of BWS after ART was
estimated to be 12.5 (95% CI 4.931.8).
To calculate the RR of the association between IVF/ICSI
and the incidence of BWS, we assumed that in Denmark
and in Sweden the incidence of BWS would be equal to other
Western European countries (1:13,700). We included the as-
sumed data of these countries, because no ART-related BWS
were reported and we did not want to overestimate the RR.
The weighted RR of the eight epidemiologic studies was 5.2
(95% CI 1.67.4), indicating a signicant positive association
between IVF/ICSI treatment and the incidence of BWS.
Other Studies on ART and BWS
Kobayshi et al. (21) analyzed 78 aborted conceptuses con-
ceived by IVF or ICSI and 38 aborted spontaneously or IUI-
conceived conceptuses. The age of the conceptuses was
from 6 to 9 weeks. They assessed the methylation index of
LIT1, and in all groups low methylation indices were found.
The IVF/ICSI group did not differ from the other groups.
The notion that IVF/ICSI causes imprinted diseases in hu-
mans is supported by differences in the proportion of the var-
ious molecular causes of BWS observed in the normal
population and in the population of IVF/ICSI children. The
molecular data of 46 BWS IVF/ICSI children are reported in
the papers discussed above: 44/46 had hypomethylation of
LIT1, and in 3/44 abnormal methylation of H19 was also
found, i.e., in 95.7% (44/46) of the children with BWS born af-
ter IVF/ICSI the disease was associated with hypomethylation
VOL. 99 NO. 3 / MARCH 1, 2013

of LIT1. In the general population this percentage is 50%.
Reductio ad absurdum led to the conclusion that IVF/ICSI
treatment is the cause. However, a similarly high percentage
of hypomethylation of LIT1 (89.5%; 17/19) was found in nat-
urally conceived discordant BWS monozygotic twins (2224).
Therefore, the high proportion of hypomethylation of LIT1 of
BWS children born after IVF/ICSI does not support a causal
relationship.
Monozygotic twinning is increased in the IVF/ICSI popu-
lation (25). One can argue that the BWS children with hypo-
methylation of LIT1 are the twins that survived while the
others would have been categorized as vanished twins. How-
ever, most of the monozygotic twins discordant for BWS are
female. In reported BWS ART children, the male-female ratio
appeared to be balanced. In spontaneous conception, mono-
zygotic twinning with BWS has to be explained by a different
mechanism than in the BWS children born after ART. It was
presumed that in the discordant monozygotic twins hypome-
thylation of LIT1 occurs after cleaving in the BWS twin (24).
Geuns et al. (26) assessed the methylation of LIT1 in 16
human oocytes and reported one human oocyte with absent
methylation of LIT1. The other 15 were normally methylated.
These observations suggest that BWS can be explained by ab-
sence of methylation of LIT1 in the oocyte. Experimental
work with nuclear transfer from immature murine oocyte nu-
clei into mature enucleated murine oocytes resulted in pups
with hypomethylation of imprinting genes in Igf2r, Snrp,
and Mest, demonstrating that hypomethylation can originate
in oocytes (27). Aberrant methylation can also be explained
by dysfunction of methylation enzymes. Bestor (28) suggested
that aberrant function of DNMT1 (DNA methyltransferase 1)
is associated with BWS. Kobayashi et al. (21) reported DNA
sequence variations in the gene encoding DNMT3L (DNA
methyltransferase 3L), which was associated with the abnor-
mal paternal DNA methylation.
There are several arguments to support the idea that IVF/
ICSI can induce imprinted diseases in humans. One of the
strongest arguments is based on results from bovine and
ovine IVF. IVF of oocytes of these species can give rise to
the large offspring syndrome. A rich culture system was
used (surplus of serum and coculture with broblasts). After
changing the culture system, large offspring syndrome disap-
peared. Large offspring syndrome can be explained by hypo-
methylation of the ovine imprinted Igf2r gene (6). We agree
that analogue reasoning is a methodologically sound way
to design experiments, but the nding that large offspring
syndrome is caused by in vitro culture and IVF and in vitro
growth of embryos is not proof that BWS has to be caused
by IVF. It must also be noted that timing of methylation dif-
fers in these species (29).
Another question is whether ART other than IVF/ICSI,
such as IUI in a stimulated cycle or use of fertility drugs
and natural conception, are associated with an increase of
the incidence of BWS. Doornbos et al. (20) reported a nonsig-
nicant RR of 2.4 (95% CI 0.69.4). Sutcliffe et al. (16) re-
ported ve BWS children born after fertility drug treatment
without IVF/ICSI. The number of children born in the relevant
period in the United Kingdom after ART treatment other than
IVF/ICSI was not reported. To estimate risk, we assumed that
VOL. 99 NO. 3 / MARCH 1, 2013
Fertility and Sterility
in the United Kingdom the number of these treatments is pro
rata similar to that of The Netherlands. In our estimate the RR
was 5.6 (95% CI 2.413.3). The combined data of Doornbos
et al. (20) and Sutcliffe et al. (16) gave an RR of 4.0 (95% CI
2.08.3). These results indicate a positive association between
fertility treatments other than IVF/ICSI and BWS. Apparently
the in vitro phase is not necessary to increase the incidence of
BWS. Further research is needed to investigate this question.
Another concern raised by the above studies is the use of
a suitable comparison group. In almost all studies, the normal
population is used as control group. However, in the normal
population only a small proportion of the children are born
to parents with a fertility problem (5.6% in the Dutch popula-
tion [20]), whereas 100% of the children conceived after IVF/
ICSI are born to parents with fertility problems. Controlling
for fertility problems would result in a truer picture of the as-
sociations between fertility problems, fertility treatments, and
imprinted diseases.
Another difculty presented by rare diseases is the sample
size needed to reach signicant and meaningful conclusions.
Rancourt et al. (30) compared the difference in methylation of
six DMRs in placentas or cord blood from 61 spontaneously
conceived children, 57 children conceived by IVF, and 21
children conceived by ovulation induction. No meaningful
differences were found. Puumala et al. (31) compared imprint-
ing in lymphocytes and buccal smears in 67 ART and 31 spon-
taneously conceived children and found no meaningful
differences. If we assume that in the group of children born
after IVF/ICSI the prevalence of BWS is 1:2,700 and that the
prevalence in the normal population is 1:13,700, study groups
of >45,000 would be necessary to demonstrate a statistically
signicant difference with an a of 5% and b of 20% (see:
www.biomath.info/power/chsq.htm).
Association of BWS and ART: Conclusions
There is a signicant positive association between IVF/ICSI
treatment and BWS, with an RR of 5.2 (95% CI 1.67.4).
With the population prevalence of 1:13,700, this suggests
that among every 2,700 IVF/ICSI births one BWS child will
be born. One study found that after correcting for fertility
problems among the parents with BWS children and in the
control group, the association was no longer signicant.
This indicates that the IVF/ICSI treatment in itself was likely
not the cause of the increased incidence of BWS in these chil-
dren, but rather the (epi)genetic background of the parents
was. Many causal relationships are possible, but none have
been proven in humans. Possible causes are the culture sys-
tem, ovarian hyperstimulation, monozygotic twinning, ge-
netic background of the parents, such as genetic variation
in DNMT1 and DNMT3L, and aberrant methylation in both
oocytes and sperm.
SILVER-RUSSELL SYNDROME AND ART
SRS is a disease associated with imprinted genes. The main
characteristics of SRS are severe intrauterine and postnatal
growth restriction, relative macrocephaly, and a small trian-
gular face (32). Approximately one-half of children with
SRS have signicant cognitive impairment (33). In 50% of
645
VIEWS AND REVIEWS
the patients with clinical diagnosis of SRS, molecular conr-
mation or explanation of the disease is not found. Prevalence
rates from 1:392,000 to 1:3,000 have been reported, support-
ing the statement that the prevalence of SRS is not known
(32). This makes discussions of an increased incidence of
SRS following ART problematic.
SRS is the only known imprinted disease that can be
caused by two different genetic mechanisms on two different
chromosomes: hypomethylation of H19 (chromosome 11),
and UPD of maternal chromosome 7 (mUDP7). Hypomethyla-
tion of H19 is often associated with other epimutations. Hy-
pomethylation of H19 seems to represent the more severe
forms of SRS than the mUPD7 form, however this is ques-
tioned by some investigators (32).
We will discuss the results of four case studies (3437) and
four epidemiologic studies. A study from Denmark reported no
SRS birth in the ART cohort (13), the study from Sweden
reported one child (12), from Japan ve children (10), and
from The Netherlands three SRS children born after IVF/ICSI
(38). Together these papers report 13 SRS children born after
IVF/ICSI. The molecular data of 11 patients were reported:
ten hypomethylation of H19 and one normal methylation of
H19 but hypermethylation of PEG1/MEST DMR (37).
Case Studies
Three of the case studies reported on the birth of a child with
SRS; all three children were born after ICSI. Two had hypome-
thylation of H19 (34, 36). No molecular data were included for
the third child (35).
The fourth case study was of special interest (37). It re-
ported on a girl with SRS of dizygotic twins conceived by
IVF. Diagnosis of SRS was based on clinical features. Genetic
and molecular diagnosis revealed no mUPD7, normal methyl-
ation of H19 and partial hypermethylation at the DMR of
PEG1/MEST. Her father also showed aberrant hypermethyla-
tion at the DMR of PEG1/MEST, though to a lesser degree than
the girl, suggesting that the aberrant methylation of the father
was heritable. We do not know whether PEG1/MEST is of rel-
evance in SRS, but the observation that aberrant imprinting
patterns can be heritable is of importance for our discussion
on the association of imprinted diseases and ART.
Epidemiologic Studies
One child with SRS was reported in Sweden (12). The child
was born after an ICSI treatment. The molecular data were
not reported.
Lidegard et al.(13) concluded from the Danish data that
there was no increase of imprinted diseases in association
with ART. In a study in which all Danish children born Janu-
ary 1, 1995December 31, 2001, were included (total 442,349
singleton non-IVF children and 6,052 IVF/ICSI children). In
the non-IVF group, two SRS and three PWS cases were regis-
tered. In the IVF/ICSI group, no children with imprinted dis-
eases were found. If we presume that in Western Europe
and in North America each of the four imprinted diseases
has an average incidence of R1:16,000, the combined inci-
dence would be 1:4,000. This means that a minimum of 90
646
children with BWS, SRS, AS, or PWS should have been regis-
tered in the non-IVF group. The results of the Danish registry
study are therefore puzzling.
Hiura et al. (10) reported ve patients with SRS, all born
after IVF. In a nationwide questionnaire, the birth of 42 SRS
children was reported, of which four had been born after
IVF or ICSI. If the average ART birthrate is 0.75% in Japan,
the RR of SRS associated with ART would be 12.6 (95% CI
5.032.2). While interpreting these data we have to keep in
mind that the authors had indicated that the response rate
to their questionnaires was not optimal.
Lammers et al. (38) had access to a complete Dutch SRS
registry containing 38 SRS children. The parents of these chil-
dren were asked to participate in a study and 23 agreed. Three
had mUPD7 and 20 had hypomethylation of H19. Two of these
23 children were born after IVF and one after ICSI. In the
relevant period, 2% of all children in The Netherlands were
born after IVF/ICSI. This gives an RR of 6.7 (95% CI 2.319.1).
These two papers (10, 38) indicate a positive signicant
association between ART and SRS.
Methylation Studies
A methylation index of H19 can inform us of the time period
in which the (de)methylation occurred. A methylation index
of 0% indicates no gain of demethylation at any time, or
that demethylation occurred before fertilization (34, 39).
Hiura et al. (10) did not present the gures of the
methylation indices, but reported that 4/5 of the ART SRS
patients were mosaic for methylation of H19, suggesting
that the methylation errors were established after
fertilization and cleavage divisions.
Kobayashi et al. (21) analyzed the methylation of seven
autosomal imprinted loci and the XIST locus of 78 aborted fe-
tuses conceived by ART and 38 aborted fetuses conceived
spontaneously. Although not reported, serious complications
or diseases of the conceptuses had to be involved to have
abortion performed by women conceived by ART, so it is
questionable how representative these conceptuses were of
live-born ART children. Molecular analyses were performed
on fetal tissue, placental tissue was carefully removed. Hypo-
methylation of H19 was found in six ART fetuses (methyla-
tion index 4%23.5%). In none of the spontaneously
conceived conceptuses were similarly low values of the meth-
ylation index of H19 found. This could suggest that hypome-
thylation of H19 occurred after fertilization. However,
Kobayashi et al. (21) also found, with the use of DNA poly-
morphisms at the H19 and GTL2 loci, seven cases where ex-
actly the same DNA methylation error was present in the
sperm of the fathers as in their conceptuses, suggesting that
the methylation errors were preexisting and were not a conse-
quence of ART. In other words, the ART procedure had not af-
fected the methylation status of these genes. We conclude that
imprinted diseases can be caused by aberrant methylation of
the genes in the sperm.
Obata et al. (27) reported that in murine oocytes aberrant
methylation patterns may be present and passed on to the off-
spring, as judged by their presence in the pups. Although this
does not prove that this also can happen in humans, it is proof
VOL. 99 NO. 3 / MARCH 1, 2013

of the principle that aberrant methylation patterns are present
in the gametes and can be transferred into the conceptuses.
Association of SRS and ART: Conclusions
Of 13 SRS children born after ART, ve were born after ICSI
and eight after IVF. There seemed to be no prevalence for
IVF or ICSI, but the numbers are too small for other
conclusions.
Two epidemiologic studies, with low power, did not show
an association between ART and the incidence of SRS. Two
other epidemiologic studies reported a positive signicant as-
sociation between the incidence of SRS and IVF/ICSI (RRs
12.6 [95% CI 5.032.2] [10] and 6.7 [95% CI 2.319.1] [38]).
We would like to conclude that there is probably a signicant
positive association between the incidence of SRS and IVF/
ICSI treatment, but the number of published cases is small.
Many of the SRS children, both spontaneous or IVF/ICSI
conceived were mosaic, suggesting that in these children the
aberrant methylation patterns were established in the early
differentiation phase of the embryos (10, 39). Because of the
low number of observations we do not know whether the
mosaic methylation pattern is associated with IVF/ICSI.
Aberrant methylation patterns detected in sperm were
also detectable in offspring. This indicated that aberrant
methylation can occur during spermatogenesis, suggesting
that in vitro procedures such as ICSI and IVF facilitate the
transfer of the aberrant sperm but are not the intrinsic cause
of the methylation errors and imprinted diseases.
ANGELMAN SYNDROME AND ART
AS is a neurogenetic disorder characterized by severe mental
retardation, delayed motor development, poor balance ac-
companied by jerky movements, absence of speech, and
happy disposition. It is caused by the absence of function of
the maternal allele of the UBE3A gene on chromosome 15, re-
sulting from a deletion, a point mutation, UPD, or an imprint-
ing defect. Only in 5% of the cases is there absence of
imprinting of the maternal allele. In 5%10% of the cases
that appear to have the major clinical phenotypic features,
there are no identiable genetic abnormalities. AS is a rare
disease. The reported incidence varies from 1:20,000 to
1:12,000 in the West (41) and from 1:134,000 in Japan (10).
A recent review was published by Dagli et al. (40).
Case Studies
In the years 20022012 two case studies (7, 41) and four
epidemiologic papers with original data (10, 16, 20, 42) were
published on this disease and its association with subfertility
and fertility treatments. Cox et al. (7) described two patients,
both born after ICSI. Both patients had absence of
methylation of the maternal allele. According to Cox et al.
(7), absence of methylation occurs only in 1:300,000 AS
patients. These two cases (probability of 1:300,0002),
evidence that maternal methylation at chromosome 15 is
established at or during fertilization, and that ICSI and
embryo culture is a process outside the body, as well as the
observations of Young et al. (4) that IVF of ovine oocytes
VOL. 99 NO. 3 / MARCH 1, 2013
Fertility and Sterility
can result in hypomethylation of the imprinting control
element of the maternal Igf2r allele, led Cox et al. (7) to
conclude that there are arguments that ICSI might interfere
with the establishment of the maternal imprint (impaired
gain of methylation) in the oocyte or preembryo and
increases the risk of imprinting defects. rstavik et al. (41)
reported a third AS patient with hypomethylation after ICSI
treatment, reinforcing the conclusions of Cox et al. (7).
Epidemiologic Studies
Ludwig et al. (42) performed a large nationwide study in Ger-
many. The authors contacted all the members of the German
AS Support Group. Seventy-nine valid questionnaires were
received from 270 members. Twenty of the 79 children were
born to couples with fertility problems (25.3%). In Germany
the proportion of children born in families with fertility prob-
lems of any kind is 5.3% (44). This resulted in an RR of 4.8
(95% CI 3.37.0). We conclude that the incidence of AS is sig-
nicantly associated with fertility problems. Three of the 79
AS children were born after ICSI, none after IVF. If we com-
pared this with the German population, assuming that 0.9%
of the children were born after IVF/ICSI in the relevant period,
we calculated an RR of 4.1 (95% CI 1.412.4). We conclude
that in Germany, in the period before the year 2005, AS is sig-
nicantly associated with ART. We also concluded that the
incidence of AS is signicantly associated with fertility prob-
lems of the parents.
Ludwig et al. (43) published the genetic data of 16 pa-
tients. They found three patients with imprinting defects,
ten with deletions, and three with unknown defects. The three
patients with imprinting defects were born after time to preg-
nancy >2 years, time to pregnancy >2 years and hormonal
treatments, and after time to pregnancy >2 years and hor-
monal treatments and ICSI. One AS patient had absence of
methylation of the maternal allele after ICSI, and two AS pa-
tients without ICSI treatment also displayed absence of meth-
ylation of the maternal allele. The rare form of AS (absence of
methylation), incidence of 1:300,000, observed by Cox et al.
(7) in each of two ICSI patients, was here also observed in
the non-ICSI AS children. This seems to indicate that absence
of methylation of the maternal allele is associated with fertil-
ity problems and not with the fertility treatment. These num-
bers are far too low for nal conclusions, but the simplest and
most likely explanation is that parents with fertility problems
have a higher chance to have a child with AS than parents
without fertility problems. It was also concluded that ovarian
hyperstimulation is a variable positively and independently
associated with AS (43). But again, that conclusion is based
on small numbers.
Sutcliffe et al. (16) sent questionnaires to 384 families
with AS children and received 81 usable responses: Six had
familial AS, 75 were sporadic. Three of these 75 (4%) were
born to families with fertility problems, but no IVF or ICSI
was involved. Assuming that the percentage of families
with fertility problems and children was similar to those of
Germany and The Netherlands (5.3% and 5.6%, respectively),
then there was no association between fertility problems and
AS in the United Kingdom.
647
VIEWS AND REVIEWS
In 2007 Doornbos et al. (20) also performed a study based
on results from questionnaires sent to members of AS support
groups. Sixty-three valid questionnaires were received. Re-
sults similar to the ndings of Ludwig et al. (43) were achieved.
Although none of the 63 children were born after IVF or ICSI,
there was a signicant association between fertility problems
of the parent and the incidence of AS (RR 3.4, 95% CI 1.95.4).
In the AS group, four (6.3%) children were born after any form
of ART. This percentage was 2.1% in the Dutch population (RR
3.1, 95% CI 1.27.9), giving a signicant association between
any form of ART and the incidence of AS.
Hiura et al. (10) reported the results of an epidemiologic
study on imprinted diseases and IVF/ICSI. They found no as-
sociation between AS and IVF/ICSI.
The four discussed epidemiologic studies reported ve AS
cases born after IVF or ICSI, in a total of 340 cases (10, 16, 20,
42). Considering the number of IVF/ICSI births and the total
number of births in the four countries in the relevant
periods, the weighted RR was 1.7 (95% CI 0.74.2). We
conclude that there was no signicant association between
AS and IVF or ICSI treatments.
AS and ART: Conclusions
Taking all the published evidence together, we conclude that
there is probably a positive association between fertility prob-
lems and the incidence of AS. As a consequence of this asso-
ciation we expected a positive association between IVF/ICSI
and AS, but that was not demonstrated.
The imprinting of the AS imprinting center (AS-IC) is es-
tablished during or after fertilization (7). It was presumed that
because of this, AS-IC would be vulnerable to methylation er-
rors associated with ICSI or IVF. In the literature discussed
here, there were seven cases of AS reported with methylation
errors: Four were born after ICSI treatment and three after
hormonal treatment or in families with a history of fertility
problems. Because of these observations and because of the
absence of association between AS and IVF/ICSI, we conclude
that it is highly improbable that ICSI or IVF prevents proper
methylation of AS-IC. Studies investigating (micro)deletions
in the gene methylation enzymes, such as the DNMT3L
gene, will be much more fruitful in explaining these methyl-
ation errors (21). From a clinical standpoint, AS has no impact
on ART owing to its low incidence.
PRADER-WILLI SYNDROME AND ART
PWS is a complex neurodevelopmental disorder resulting
from errors in genomic imprinting with loss of methylation centage was 1.0%). The same group published a new study
of imprinted genes that are paternally expressed in the chro-
mosome 15q11q13 region, the same region that is responsi-
ble for AS. PWS is characterized by severe hypotonia and
feeding difculties in early infancy, followed in later infancy
or early childhood by excessive eating and gradual develop-
ment of morbid obesity. All individuals have some degree of
cognitive impairment. The worldwide prevalence is
1:30,00010,000. Approximately 70% of PWS cases stem
from de novo deletion in the paternally derived chromosome
15q11q13 region. About 25% of the cases can be explained
by maternal disomy of chromosome 15. In about 3% of cases
648
PWS can be explained either by genomic imprinting defects
due to microdeletions or epimutations of the imprinting cen-
ter located in the 15q11q13 region or by chromosome 15
translocations or rearrangements (44, 45).
There are only a few studies in which original data are re-
ported on the association between IVF/ICSI and PWS (10, 12,
16, 20). All of the studies concluded that there was no
association between ART and PWS.
In the period 19822001 the birth of 16,280 IVF/ICSI chil-
dren was reported in Sweden, and one of these had PWS (12).
Sutcliffe et al. (16) contacted 522 families with PWS chil-
dren, 169 replied, and six families had a history of PWS. Two
children (1.2%) were born after ICSI, none after IVF. These
numbers did not differ from that of the normal population.
Seven children (4.3%) were born after the use of fertility
drugs. The authors indicated that 2% of the children in the
United Kingdom are born after any type of fertility treatment.
In the PWS population 5.5% were born after fertility treat-
ment (RR 2.8, 95% CI 1.025.21). This gure indicates that
there is an association between fertility problems of the par-
ents and an increased incidence of PWS.
Doornbos et al. (20) performed a similar study. They re-
ceived replies from 86 families with PWS children. Twelve
of those families had fertility problems of any kind (RR 2.5,
95% CI 1.54.2, compared with the normal population). There
was no signicant association between IVF/ICSI and the inci-
dence of PWS (RR 2.5, 95% CI 0.910.0).
Hiura et al. (10) reported the birth of four IVF/ICSI chil-
dren in a group of 261 PWS children (1.5%). This proportion
did not differ from that of IVF/ICSI children born in the nor-
mal population.
PWS and ART: Conclusions
Children with PWS are more likely to be born to parents with
a fertility problem (RR 2.52.8), however, there is no signi-
cant association between PWS and IVF/ICSI. The same para-
dox was observed in the group of children with AS.
RETINOBLASTOMA
Retinoblastoma is a cancer originating from the retina of one
or both eyes in young children. There are several reports of as-
sociations between IVF/ICSI and retinoblastoma. Moll et al.
(46) studied the incidence in the period 20002002 and calcu-
lated an RR of 4.9 (95% CI 1.611.3) or 7.2 (95% CI 2.417.0)
(4.9 if it was assumed that 1.5% of the children in The Nether-
lands were born after IVF/ICSI and 7.2 if this assumed per-
in 2009 in which the incidence of retinoblastoma was studied
over the period 20022007 (47). The RR in this period was 1.29
(95% CI 0.10- 4.66), and it was concluded that there was no
signicant association between IVF/ICSI and retinoblastoma.
In an abstract of another study, an RR of 4.5, without con-
dence interval, was reported; no follow-up of that result
was published (48).
Retinoblastoma can be heritable and can be caused by
two mutations in the RB1 gene or by hypermethylation of
the promoter of the RB1 gene. If the retinoblastomas observed
in IVF/ICSI children is caused by hypermethylation of this
VOL. 99 NO. 3 / MARCH 1, 2013

promoter, a relationship between the fertility treatment and
the occurrence of retinoblastoma can not be excluded. In
2012 a paper was published in which the molecular origin
of the tumors was delineated. For all tumors, two causative
RB1 mutations were found. None of the tumors showed hy-
permethylation of the RB1 promoter. This demonstrated
that an association between IVF or ICSI and retinoblastoma
through this epigenomic mechanism is highly unlikely (49).
DISCUSSION
Alarming studies on the association between imprinted dis-
eases and IVF/ICSI were published in recent years. The inci-
dences of BWS was reported to be up to 12 times higher in
children born after IVF/ICSI than in the normal population.
Also, the birth of calves and lambs with the large offspring
syndrome after IVF with both ovine and bovine gametes,
and the observation that this could be explained by demethy-
lation of the Igf2r gene, raised the question of whether in vitro
culture of oocytes and embryos could cause methylation er-
rors and imprinted diseases.
Imprinted diseases are rare, with reported incidences of
1:30,00010,000. An incidence as low as 1:392,000 for SRS
has been reported in Japan (10), although that study was based
on questionnaires sent to departments of pediatrics with subop-
timal response rate. It can be expected that the reported preva-
lence in Japan will be adjusted. However, it has not been
established whether prevalence of imprinted diseases differs
among different human ethnic backgrounds. A consequence
of the rarity of these diseases is the difculty to perform studies
with sufcient power. If we assume that in the group of children
born after IVF/ICSI the rate of BWS may reach 1:2,700 com-
pared with 1:14,000 in the normal population, study groups
of >45,000 are needed to demonstrate a statistically signicant
difference with an a of 5% and a b of 20%.
Studies of the association between ART and retinoblas-
toma illustrate the problem of imprinted disease and studies
lacking sufcient statistical power. The highly signicant effect of the treatments they received, and we must search for
association rst reported vanished over time as greater num-
bers were collected. Subsequently it was reported that for all
tumors two causative RB1 mutations were found and none
of the tumors showed hypermethylation of the RB1 promoter.
It is concluded, in light of these additional data, that an asso-
ciation between IVF or ICSI and retinoblastoma through this
epigenetic mechanism is unlikely (49).
Keeping these pitfalls in mind, we examined the existence
of a statistical relationship between the incidence of an im-
printed disease and fertility treatments. If such a relationship
exists, the possible etiology must be investigated before we
can conclude that there is a causal relationship between im-
printed diseases and ART. However, because it would be
highly unethical to try to induce imprinted diseases by exper-
imental studies in human embryos, the question of causality
will be very difcult to answer.
Based on the extant literature, it was possible to calculate
a weighted RR for BWS. We included eight studies: two stud-
ies in which no association was found (12, 13) and six (10, 11,
14, 16, 19, 20) with a positive association. The weighted RR
was 5.2 (95% CI 1.67.8). We concluded that there is
VOL. 99 NO. 3 / MARCH 1, 2013
Fertility and Sterility
a signicant positive association between IVF/ICSI
treatment and risk of BWS.
Four epidemiologic studies examined the relationship be-
tween SRS and IVF/ICSI. Two studies with low power found
no association, whereas two other studies that included eight
children with SRS after IVF/ICSI treatment reported a signi-
cant positive association (RRs 12.6 [95% CI 5.032.2] [10] and
6.7 [95% CI 2.319.1] [38]). The incidence of SRS is not
known. Many children born small for gestational age may
have some form of SRS. Improvement in the diagnosis will
be essential for studying the putative association between
SRS and IVF/ICSI.
We did not nd a positive association between the risk of
AS and IVF/ICSI. Two studies found a positive association be-
tween AS and other fertility treatments (20, 42).
We also did not nd a positive association between the
incidence of PWS and IVF/ICSI. However, we concluded
that there is a signicantly higher risk for a child with AS
or PWS to be born to parents with fertility problems
(16, 20). The apparent paradox of increased fertility
problems but no increased numbers of IVF/ICSI treatments
can be explained by the rarity of the diseases.
Another problem of investigation into the statistical rela-
tionships between ART and imprinted diseases is the compar-
ison group. In all of the studies but one (20), the normal
population was used as the comparison group. Couples under-
going IVF or ICSI have fertility problems, whereas in the gen-
eral population the vast majority of children are conceived
spontaneously. Valid comparison between IVF/ICSIcon-
ceived children and control subjects must account for fertility
problems in both groups. Applying this principle, the statisti-
cal relationship between imprinted diseases and IVF/ICSI dis-
appeared in the aforementioned study. Thus the increased
incidence of imprinted diseases after IVF/ICSI can be ex-
plained by the fertility problem of the parents.
Apparently the increased incidence of imprinted diseases
in children born after IVF or ICSI treatment is not an intrinsic
other causes. Several publications reported methylation errors
in sperm and oocytes that could be traced in the offspring, ex-
plaining BWS or SRS (21, 26, 36). It may be that absence of
methylation and mosaic patterns of methylation present in
SRS patients are caused by improper enzyme functions of
the methylation enzyme DNMT3L caused by microdeletions
(21, 36). This would indicate that the imprinted error is not
caused by the culture system, but that the in vitro system is
only permissive for the transfer and propagation of
aberrantly methylated genes.
In conclusion, retinoblastoma, AS, and PWS are not asso-
ciated with IVF/ICSI, BWS is signicantly associated with
IVF/ICSI, and the number of cases of SRS is small but a posi-
tive association is probable. The incidences of AS and PWS
are positively associated with families with fertility problems,
but not with IVF/ICSI. This paradox can be explained by the
low power of the studies due to the rarity of the diseases. There
are no indications for a causal relationship between ART and
imprinted diseases in humans. That does not imply that the
IVF/ICSI procedures do not affect the epigenome (31). Chil-
dren conceived during stressful situations can display
649
VIEWS AND REVIEWS
differences in glucose tolerance and blood pressure. This is as-
sociated with small but signicant differences in methylation
of DMRs (5), and it is presumed that the differences in glucose
tolerance and birth weight in children born after IVF or ICSI
can be explained by these mechanisms. These adaptations
of the epigenome are normal physiologic reactions and are
not diseases (1, 5, 50).
REFERENCES
1. Enklaar T, Zabel BU, Prawitt D. Beckwith-Wiedemann syndrome: multiple
molecular mechanisms. Expert Rev Mol Med 2006;8:119.
2. Ceelen M, van Weissenbruch MM, Vermeiden JP, van Leeuwen FE, Delemarre-
van de Waal HA. Cardiometabolic differences in children born after in vitro fer-
tilization: follow-up study. J Clin Endocrinol Metab 2008;93:16828.
3. Barker DJ, Bagby SP, Hanson MA. Mechanisms of disease: in utero program-
ming in the pathogenesis of hypertension. Nat Clin Pract Nephrol 2006;2:
7007.
4. Young LE, Sinclair KD, Wilmut I. Large offspring syndrome in cattle and
sheep. Rev Reprod 1998;3:15563.
5. Tobi EW, Slagboom PE, van Dongen J, Kremer D, Stein AD, Putter H, et al.
Prenatal famine and genetic variation are independently and additively asso-
ciated with DNA methylation at regulatory loci within IGF2/H19. PLoS One
2012;7:e37933.
6. Young LE, Fernandes K, McEvoy TG, Butterwith SC, Gutierrez CG,
Carolan C, et al. Epigenetic change in IGF2R is associated with fetal over-
growth after sheep embryo culture. Nat Genet 2001;27:1534.
7. Cox GF, B urger J, Lip V, Mau UA, Sperling K, Wu BL, et al. Intracytoplasmic
sperm injection may increase the risk of imprinting defects. Am J Hum Genet
2002;71:1624.
8. Shuman C, Beckwith JB, Smith AC, Weksberg R. Beckwith-Wiedemann syn-
drome. Updated December 14, 2010. In: Pagon RA, Bird TD, Dolan CR, Ste-
phens K, Adam MP, eds. Gene reviews [internet]. Seattle: University of
Washington, 19932000.
9. Bowdin S, Allen C, Kirby G, Brueton L, Afnan M, Barratt C, Kirkman-Brown J,
Harrison R, et al. A survey of assisted reproductive technology births and im-
printing disorders. Hum Reprod 2007;22:323740.
10. Hiura H, Okae H, Miyauchi N, Sato F, Sato A, van de Pette M, et al. Charac-
terization of DNA methylation errors in patients with imprinting disorders
conceived by assisted reproduction technologies. Hum Reprod 2012;27:
25418.
11. Gicquel C, Gaston V, Mandelbaum J, Siffroi JP, Flahault A, le Bouc Y. In vitro
fertilization may increase the risk of Beckwith-Wiedemann syndrome related
to the abnormal imprinting of the KCNQ1OT gene. Am J Hum Genet 2003;
72:133841.
12. Kall
en B, Finnstrom O, Nygren KG, Olausson PO. In vitro fertilization (IVF) in
Sweden: risk for congenital malformations after different IVF methods. Birth
Defects Res A Clin Mol Teratol 2005;73:1629.
13. Lidegaard O, Pinborg A, Andersen AN. Imprinting diseases and IVF: Danish
national IVF cohort study. Hum Reprod 2005;20:9504.
14. DeBaun MR, Niemitz EL, Feinberg AP. Association of in vitro fertilization with
Beckwith-Wiedemann syndrome and epigenetic alterations of LIT1 and
H19. Am J Hum Genet 2003;72:15660.
15. Maher E, Brueton L, Bowdin S, Luharia A, Cooper W, Cole T, et al. Beckwith-
Wiedemann syndrome and assisted reproduction technology (ART). J Med
Genet 2003;40:624.
16. Sutcliffe AG, Peters CJ, Bowdin S, Temple K, Reardon W, Wilson L, et al. As-
sisted reproductive therapies and imprinting disordersa preliminary British
survey. Hum Reprod 2006;21:100911.
17. Lim D, Bowdin SC, Tee L, Kirby GA, Blair E, Fryer A, et al. Clinical and molec-
ular genetic features of Beckwith-Wiedemann syndrome associated with as-
sisted reproductive technologies. Hum Reprod 2009;24:7417.
18. Rossignol S, Steunou V, Chalas C, Kerjean A, Rigolet M, Viegas-Pequignot E,
et al. The epigenetic imprinting defect of patients with Beckwith-
Wiedemann syndrome born after assisted reproductive technology is not re-
stricted to the 11p15 region. J Med Genet 2006;43:9027.
650
19. Halliday J, Oke K, Breheny S, Algar E, J Amor D. Beckwith-Wiedemann syn-
drome and IVF: a case-control study. Am J Hum Genet 2004;75:5268.
20. Doornbos ME, Maas SM, McDonnell J, Vermeiden JP, Hennekam RC. Infer-
tility, assisted reproduction technologies and imprinting disturbances:
a Dutch study. Hum Reprod 2007;22:247680.
21. Kobayashi H, Hiura H, John RM, Sato A, Otsu E, Kobayashi N, et al. DNA
methylation errors at imprinted loci after assisted conception originate in
the parental sperm. Eur J Hum Genet 2009;17:158291.
22. Weksberg R, Shuman C, Beckwith JB. Beckwith-Wiedemann syndrome. Eur
J Hum Genet 2010;18:814.
23. Gaston V, Le Bouc Y, Soupre V, Burglen L, Donadieu J, Oro H, et al. Analysis
of the methylation status of the KCNQ1OT and H19 genes in leukocyte DNA
for the diagnosis and prognosis of Beckwith-Wiedemann syndrome. Eur J
Hum Genet 2001;9:40918.
24. Bliek J, Alders M, Maas SM, Oostra RJ, Mackay DM, van der Lip K, et al. Les-
sons from BWS twins: complex maternal and paternal hypomethylation and
a common source of haematopoietic stem cells. Eur J Hum Genet 2009;17:
162534.
25. Aston KI, Peterson CM, Carrell DT. Monozygotic twinning associated with as-
sisted reproductive technologies: a review. Reproduction 2008;136:37786.
26. Geuns E, Hilven P, van Steirteghem A, Liebaers I, de Rycke M. Methylation
analysis of KvDMR1 in human oocytes. J Med Genet 2007;44:1447.
27. Obata Y, Hiura H, Fukuda A, Komiyama J, Hatada I, Kono T. Epigenetically
immature oocytes lead to loss of imprinting during embryogenesis. J Reprod
Dev 2011;57:32734.
28. Bestor TH. Imprinting errors and developmental asymmetry. Philos Trans R
Soc Lond B Biol Sci 2003;358:14115.
29. Thurston A, Taylor J, Gardner J, Sinclair KD, Young LE. Monoallelic expres-
sion of nine imprinted genes in the sheep embryo occurs after the blastocyst
stage. Reproduction 2008;135:2940.
30. Rancourt RC, Harris HR, Michels KB. Methylation levels at imprinting control
regions are not altered with ovulation induction or in vitro fertilization in
a birth cohort. Hum Reprod 2012;27:220816.
31. Puumala SE, Nelson HH, Ross JA, Nguyen RH, Damario MA, Spector LG. Sim-
ilar DNA methylation levels in specic imprinting control regions in children
conceived with and without assisted reproductive technology: a cross-
sectional study. BMC Pediatr 2012;20:1233.
32. Eggermann T, Gonzalez D, Spengler S, Arslan-Kirchner M, Binder G,
Scho nherr N. Broad clinical spectrum in Silver-Russell syndrome and conse-
quences for genetic testing in growth retardation. Pediatrics 2009;123:
e92931.
33. Lai KY, Skuse D, Stanhope R, Hindmarsh P. Cognitive abilities associated
with the Silver-Russell syndrome. Arch Dis Child 1994;71:4906.
34. Bliek J, Terhal P, van den Bogaard MJ, Maas S, Hamel B, Salieb-Beugelaar G,
et al. Hypomethylation of the H19 gene causes not only Silver-Russell syn-
drome (SRS) but also isolated asymmetry or an SRS-like phenotype. Am J
Hum Genet 2006;78:60414.
35. Galli-Tsinopoulou A, Emmanouilidou E, Karagianni P, Grigoriadou M,
Kirkos J, Varlamis GS. A female infant with Silver Russell syndrome, mesocar-
dia and enlargement of the clitoris. Hormones (Athens) 2008;7:7781.
36. Chopra M, Amor DJ, Sutton L, Algar E, Mowat D. Russell-Silver syndrome
due to paternal H19/IGF2 hypomethylation in a patient conceived using in-
tracytoplasmic sperm injection. Reprod Biomed Online 2010;20:8437.
37. Kagami M, Nagai T, Fukami M, Yamazawa K, Ogata T. Silver-Russell syn-
drome in a girl born after in vitro fertilization: partial hypermethylation at
the differentially methylated region of PEG1/MEST. J Assist Reprod Genet
2007;24:1316.
38. Lammers TH, van Haelst MM, Alders M, Cobben JM. Het Silver-Russell-
syndroom in Nederland. Tijdschr Kindergeneeskd 2012;80:8691.
39. Higashimoto K, Nakabayashi K, Yatsuki H, Yoshinaga H, Jozaki K, Okada J,
et al. Aberrant methylation of H19-DMR acquired after implantation was
dissimilar in soma versus placenta of patients with Beckwith-Wiedemann
syndrome. Am J Med Genet A 2012;158A:16705.
40. Dagli A, Buiting K, Williams CA. Molecular and clinical aspects of Angelman
syndrome. Mol Syndromol 2012;2:10012.
41. rstavik KH, Eiklid K, van der Hagen CB, Spetalen S, Kierulf K, Skjeldal O,
et al. Another case of imprinting defect in a girl with Angelman syndrome
VOL. 99 NO. 3 / MARCH 1, 2013

who was conceived by intracytoplasmic semen injection. Am J Hum Genet
2003;72:2189.
42. Ludwig M, Katalinic A, Gross S, Sutcliffe A, Varon R, Horsthemke B.
Increased prevalence of imprinting defects in patients with Angel-
man syndrome born to subfertile couples. J Med Genet 2005;42:
28991.
43. Gnoth C, Godehardt D, Godehardt E, Frank-Herrmann P, Freundl G. Time to
pregnancy: results of the German Prospective Study and impact on the man-
agement of infertility. Hum Reprod 2003;18:195966.
44. Butler MG. Prader-Willi syndrome: obesity due to genomic imprinting. Curr
Genomics 2011;12:20415.
45. Driscoll DJ, Miller JL, Schwartz S, Cassidy SB. Prader-Willi Syndrome. Prader-
Willi syndrome. Updated October 11, 2012. In: Pagon RA, Bird TD, Dolan
CR, Stephens K, Adam MP, eds. Gene reviews [internet]. Seattle: University
of Washington, 19931998.
VOL. 99 NO. 3 / MARCH 1, 2013
Fertility and Sterility
46. Moll AC, Imhof SM, Cruysberg JR, Schouten-van Meeteren AY, Boers M,
van Leeuwen FE. Incidence of retinoblastoma in children born after in-
vitro fertilisation. Lancet 2003;361:30910.
47. Marees T, Dommering CJ, Imhof SM, Kors WA, Ringens PJ, van Leeuwen FE,
et al. Incidence of retinoblastoma in Dutch children conceived by IVF: an ex-
panded study. Hum Reprod 2009;24:32204.
48. Viot GB, Epelboin S, Olivennes F. Epigenetic risks in assisted reproductive
technologies (ART): results from a French cohort composed of 15,162 chil-
dren conceived by in vitro fertilization or intracytoplasmic sperm injection.
Fertil Steril 2010;94:S22.
49. Dommering CJ, van der Hout AH, Meijers-Heijboer H, Marees T, Moll AC. IVF
and retinoblastoma revisited. Fertil Steril 2012;97:7981.
50. Dumoulin JC, Land JA, Van Montfoort AP, Nelissen EC, Coonen E,
Derhaag JG, et al. Effect of in vitro culture of human embryos on birth
weight of newborns. Hum Reprod 2010;25:60512.
651