Journal of Reproductive Immunology 93 (2012) 5863

Contents lists available at SciVerse ScienceDirect

Journal of Reproductive Immunology

journal homepage: www.elsevier.com/locate/jreprimm

The distribution of immune cells and macrophages in the

endometrium of women with recurrent reproductive failure. II:
adenomyosis and macrophages
Kelton P. Tremellen a,b, , Peter Russell c,d
a
b
c
d

Repromed, Adelaide, Australia

School of Pharmacy and Medical Science, University of South Australia, Australia
GynaePath, Douglass Hanly Moir Pathology, Australia
Department of Obstetrics and Gynaecology, The University of Sydney, Australia

a r t i c l e

i n f o

Article history:
Received 22 September 2011
Received in revised form
23 November 2011
Accepted 5 December 2011

Keywords:
Endometrium
IVF
CD163
Macrophages
Adenomyosis

a b s t r a c t
Adenomyosis, a condition usually associated with multiparity, is not generally seen as a
cause of infertility. However, recent studies have reported a reduction in IVF implantation
rates and a link with miscarriage, suggesting that adenomyosis may interfere with successful implantation. To investigate this hypothesis, the clinical records and laboratory results,
which routinely include immunohistochemical examination of a late luteal phase endometrial biopsy for leukocytes, were retrospectively reviewed for 64 women with implantation
failure and who previously had been screened for the presence of adenomyosis by pelvic
MRI.
The presence of either diffuse or adenomyoma type of adenomyosis was associated
with a marked increase (p = 0.004) in the density of macrophages and natural killer cells
in the endometrial stroma, compared to those women with mild focal adenomyosis or no
disease. These ndings point to an immunological mechanism by which adenomyosis might
interfere with successful embryo implantation.
Crown Copyright 2012 Published by Elsevier Ireland Ltd. All rights reserved.

1. Introduction
Adenomyosis, a disorder related to endometriosis (Kunz
et al., 2005; Leyendecker et al., 2006), is characterised
by heterotopic endometrial glands and stroma in the
myometrium, and is an established cause of menorrhagia, dysmenorrhoea and pelvic pressure symptoms (Peric
and Fraser, 2006). Most commonly it is diagnosed in multiparous women in their 5th decade of life and therefore is
not traditionally associated with an inability to conceive
(Vercellini et al., 2006). Adenomyosis is primarily diagnosed at pathological examination of the uterus following

Corresponding author at: Repromed, 180 Fullarton Road, Dulwich,

South Australia 5065, Australia. Fax: +61 8 8333 8188.
E-mail address: ktremellen@repromed.com.au (K.P. Tremellen).

hysterectomy, a sterilising procedure unlikely to be undertaken in an infertile patient. Since as many as one third
of women with pathologically conrmed adenomyosis are
symptom-free (Peric and Fraser, 2006), it is likely that adenomyosis in infertile women is frequently missed.
Recent studies have suggested that adenomyosis may
be associated with impaired reproduction. A study of
46 women undergoing resection of bowel endometriosis
reported that the subsequent natural and IVF assisted pregnancy rate in those with co-existing adenomyosis was very
signicantly reduced (Ferrero et al., 2009). Furthermore,
two small case series have linked the presence of adenomyosis with early miscarriage (Kano et al., 1997; Olive et al.,
1982). All of these studies suggest that adenomyosis may
interfere with successful implantation.
Pelvic Magnetic Resonance Imaging (MRI) is the radiological investigation of choice for the diagnosis of

0165-0378/$ see front matter Crown Copyright 2012 Published by Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.jri.2011.12.001

K.P. Tremellen, P. Russell / Journal of Reproductive Immunology 93 (2012) 5863

adenomyosis, with a sensitivity and specicity approaching 90% (Reinhold et al., 1998; Tamai et al., 2006), now
enabling the non-invasive examination of the relationship
between adenomyosis and infertility. Two prospective MRI
studies (Kissler et al., 2006; Maubon et al., 2010) and a more
recent case-series reported from our group (Tremellen and
Russell, 2011) have conrmed a link between failure of successful implantation of good quality embryos during IVF
treatment and adenomyosis. However, other investigators
have not conrmed a signicant correlation between MRI
evidence of adenomyosis and implantation failure (Kunz
and Beil, 2010; Turnbull et al., 1994). Therefore we sought
to examine in a large study cohort if adenomyosis may produce implantation failure, by investigating differences in
endometrial function between women with and without
adenomyosis detected by MRI scan.
Whilst the exact mechanisms by which adenomyosis
may limit fertility are still under debate, alterations in
the endometrial immune environment were suggested as
a possible cause in a small case series of patients with
adenomyosis, who exhibited excessively high endometrial
macrophage density compared to published normal ranges
(Russell et al., 2011; Tremellen and Russell, 2011). Since it
is well recognised that macrophages have the capability
of releasing embryo-toxic cytokines and reactive oxygen
species (Agarwal et al., 2005; Haddad et al., 1995; Lee et al.,
2004), we specically examined the relationship between
MRI diagnosed adenomyosis and endometrial immune cell
populations in a larger cohort of women experiencing IVF
implantation failure.
2. Materials and methods
2.1. Study population
Investigation of recurrent implantation failure in
Repromed (Adelaide, South Australia) generally consists
of an assessment for thrombophilias (lupus anticoagulant, anticardiolipin antibody, factor V Leiden mutation,
prothrombin mutation, MTHFR genotype, homocysteine),
evidence for autoimmunity (thyroid antibodies, ANA),
coeliac antibody screen and an endometrial biopsy in the
late luteal phase (endometrial morphology and assessment of selected immunocompetent cells). A diagnostic
hysteroscopy and laparoscopy are performed if any abnormalities are found on ultrasound or symptoms suggest
pelvic pathology such as endometriosis. Patients with submucous or cavity-distorting intramural broids or pelvic
pathology such as hydrosalpinx were excluded from the
study as these conditions are already known to alter
endometrial immune cell populations (Copperman et al.,
2006; Miura et al., 2006).
Since late 2009, we have routinely performed a pelvic
MRI in patients with recurrent IVF failure, and therefore
we were able to retrospectively correlate the MRI diagnosis of adenomyosis with changes in the numbers of
selected endometrial immunocompetent cells. This study
was approved by the local scientic advisory committee
and did not require formal ethics committee review due
to its purely retrospective nature, in line with Australian
National Health and Medical Research Council guidelines.

59

Fig. 1. Immunostain of CD163+ macrophages in the supercial endometrial stroma (dark brown) in a background of non-reactive stromal cells
(pale blue nuclei). (For interpretation of the references to colour in this
gure legend, the reader is referred to the web version of the article.)

2.2. Endometrial tissue collection and processing

Endometrial biopsies were conducted between 8 and
12 days post ovulation, corresponding to days 22 through
to 26 of a standardised 28-day menstrual cycle. In
patients with irregular cycles, serum hormone proles
were used to time biopsies. Anovulatory patients were
given clomiphene citrate to produce ovulation or an articial hormone replacement cycle (oral oestradiol valerate
for a minimum 2 weeks before commencing vaginal progesterone). All patients had the endometrial sample dated
according to the time-honoured criteria of Noyes et al.
(1975). An additional six cases where histological dating
of the endometrium, using these traditional morphological criteria, was outside the day 2226 period of interest,
were omitted from the study.
All endometrial samples were xed in 10% neutral
buffered formalin, processed within 48 h into parafn, had
sections cut at 4 m and stained with haematoxylin and
eosin according to a standard protocol. Serial sections were
also cut at 4 m and immunostained in a Ventana Benchmark XT using ULTRA view DAB detection kit (Ventana
Medical Systems, Oro Valley, USA), using commercially
available monoclonal antibodies, for natural killer (NK)
cells (CD56) and macrophages (CD163), as previously published by our group (Russell et al., 2011). All histological
studies were performed by one specialist pathologist (PR).
The only information provided to the pathologist before
reporting of the biopsy was the date of the last menstrual
period and a clinical history of recurrent implantation
failure. The pathologist was blinded to the MRI result
so as to remove any chance of bias in the report ndings. CD56+ NK cell and CD163+ macrophage numbers
were quantied both as immunopositive cells per mm2 of
endometrial stroma in the supercial functional layer of the
endometrium, and as a percentage of stromal cells in the
same area (Fig. 1). Stromal macrophage counts were performed on 5 high power elds (40 objective), devoid of
endometrial glands. Percentage counts were performed on

60

K.P. Tremellen, P. Russell / Journal of Reproductive Immunology 93 (2012) 5863

Table 1
Subject baseline characteristics.
Adenomyosis
positive
Number subjects
31
Maternal age (years)
37.3 4.4
Number prior
1 (02)
miscarriages
Number of prior live
0 (01)
births
7 (412)
Total number of
embryos transferred
during IVF
(fresh + cryopreserved)
Aetiology infertility
22.6%
1. Male
2. Female
29%
29%
3. Combined
19.4%
4. Unknown

Fig. 2. Immunostain of CD163+ macrophages within the gland lumens

of the supercial endometrium (dark brown) this case corresponding
to +++ in severity. (For interpretation of the references to colour in this
gure legend, the reader is referred to the web version of the article.)

a single oil immersion eld (100 objective) deemed to be

representative of the cell density, as has been previously
reported by our group (Russell et al., 2011). In addition,
the presence of CD163+ macrophages within the glandular lumens were semi-quantied on a +, ++, +++ and ++++
grading scale, when present (Fig. 2).
2.3. Pelvic MRI protocol
All pelvic MRIs were performed on a Philips Eclipse 1.5
T (Philips, Netherlands) MRI scanner, with a 1.5 T magnet
and 27-mT/m gradients. Sequences included T2-weighted
sagittal, frontal and axial sequences. Patients were advised
to preferably have their scan in the second half of the menstrual cycle when adenomyosis is most prominent.
Patients found to have a junctional zone less than 8 mm
in thickness were considered to be normal with no evidence of adenomyosis (Reinhold et al., 1998; Tamai et al.,
2006). Patients identied as having a thickened junctional
zone 12 mm were considered to have sufcient MRI evidence of adenomyosis, irrespective of whether or not they
had additional features of adenomyosis such as the presence of sub-mucosal cysts (Reinhold et al., 1998). Those
patients with features of adenomyosis localised to a small
area of the uterus less than 2 cm in width were classied
as having focal mild adenomyosis, whilst those patients
exhibiting widespread junctional zone thickening of the
majority of the anterior and/or posterior uterine wall, or
a nodular adenomyoma type lesions producing distorsion of the endometrium, were considered to have severe
adenomyosis.
2.4. Statistical analysis
Statistical analysis was performed using the Graphpad
Prism 5 software (San Diego, USA). Of all the continuous
variables recorded, only maternal age and CD56 density
were normally distributed and therefore expressed as a
mean (SEM) and analysed using the non-paired t-test. The

Adenomyosis
negative

p-Value

33
36.0 3.4
1 (02)

0.061a
0.25b

0 (00.25)

0.297b

8 (512)

0.53b

21.2%
33.3%
30.3%
15.2%

0.96c

Maternal age is expressed as mean SEM. Other variables are expressed

as median (inter-quartile range).
a
Unpaired t-test.
b
Mann Whitney test.
c
Chi square analysis.

remaining continuous variables were expressed as medians with inter-quartile (25th75th percentile) ranges and
analysed by the non-parametric Mann Whitney test. Categorical variables such as infertility diagnosis were analysed
by a Chi-square test. The differences between comparison groups were considered to be statistically signicant
at p < 0.05.
3. Results
3.1. Clinical data
Sixty-four women with recurrent implantation failure were identied as having had a pelvic MRI and late
luteal phase endometrial biopsy between January 2010
and March 2011. Their baseline clinical data are given in
Table 1. In summary, the mean age of the adenomyosis
group (n = 31) was slightly older than women found not
to have adenomyosis (n = 33), but this difference did not
reach statistical signicance (p = 0.061). All women had
undergone a signicant number of unsuccessful embryo
transfers, and the majority were experiencing primary
rather than secondary infertility (Table 1). Laparoscopically
conrmed endometriosis was present in ve of 31 (16%)
adenomyosis cases and in three of 33 (9.1%) women with
no evidence of adenomyosis on MRI (p = 0.46). Other aetiological factors for infertility were shared equally between
the two groups.
3.2. Immune cell populations
Fig. 1 shows the typical representative section displaying the immunohistochemistry stain patterns of CD163+
macrophages in the supercial endometrial stroma, with
Fig. 2 being an example of a marked inltration of
CD163+ macrophages into the supercial endometrial
gland lumens. The differences in median (inter-quartile
range) CD163+ density in the endometrial stroma and

K.P. Tremellen, P. Russell / Journal of Reproductive Immunology 93 (2012) 5863

Table 2
Immunohistochemistry analysis.

0.013

1 (02)

1 (02.5)

0.418

1160 124

913 93

0.112

Macrophages quantied using CD163 and uterine Natural Killer cells

with CD56 immunostains. CD163 data were not normally distributed
and therefore are presented as median (inter-quartile range) and analysed with Mann Whitney test. CD56 data were normally distributed and
expressed as a mean (SEM) and analysed by an unpaired t-test.

the supercial glands observed between patients with

mild focal adenomyosis, severe diffuse adenomyosis and
those with no MRI evidence of adenomyosis is recorded
in Table 2, and is represented in graphical form in Fig. 3.
The median day of sampling was the same in the adenomyosis and non-adenomyosis groups, enabling legitimate
comparisons between these two groups.
The median density of CD163+ stromal macrophages
(Table 2) was greatly increased in the adenomyosis
group compared with the non-adenomyosis group (median
280/mm2 vs 210/mm2 , p = 0.013). However, if the adenomyosis population is divided into those with mild focal
adenomyosis and those with severe diffuse or adenomyoma type of adenomyosis, the above difference is almost
entirely due to increased macrophage populations in the
severely affected patients (Fig. 3). The density of stromal
macrophages does not signicantly differ between subjects
with no MRI evidence of adenomyosis (median 210/mm2 ,
IQR 170245/mm2 ) compared with those women with

CD163 positive cells (mm

2)

Endometrial macrophage density

a

600
400
200

ad
en
o
se
ve

re

ad
ild
m

ad
en
o

en
o

Fig. 3. Immunohistochemical analysis of macrophage density (CD163)

in the endometrial stromal tissue. Horizontal lines represent the group
medians as data were not normally distributed. Differences between the
groups were analysed by the MannWhitney U test. Groups labelled with
different letters identify statistically signicant differences between those
groups.

2000
1000
0
ad
en
o

210 (170245)

re

280 (190360)

Se
ve

Stromal CD 163
density (mm2 )
Supercial
glandular CD163
density (0 to 4+)
Stromal CD 56
density (mm2 )

a
3000

en
o

0.321

ad

24 (2324)

M
ild

24 (2324)

ad
en
o

Day of biopsy

4000

p-Value

Adenomyosis
negative

CD56 positive cells (mm2)

Endometrial Natural Killer cell density

Adenomyosis
positive

800

61

Fig. 4. Immunohistochemical analysis of Natural Killer cell density (CD56)

in endometrial stromal tissue. Horizontal lines represent the mean SEM.
Differences between group means were analysed using an unpaired t-test.
Groups labelled with different letters identify statistically signicantly
differences between those groups.

mild adenomyosis (median 220/mm2 , IQR 172310/mm2 ).

However, women with severe adenomyosis have a statistically signicant greater stromal macrophage density
(median 335/mm2 , IQR 263560/mm2 ) than both nonadenomyosis controls and mild focal adenomyosis cases
(p = 0.004). Analysis of macrophage numbers in the supercial glandular lumens (Table 2) using the semi-quantitative
0 to 4+ scale revealed no signicant differences between
the adenomyosis and non-adenomyosis groups.
Whilst less pronounced than changes seen in
macrophage density, signicant differences were also
observed in CD56+ uNK cell densities (Fig. 4) between the
severe adenomyosis and the non-adenomyosis group.
4. Discussion
This study is the rst to report an increased stromal macrophage population in the functional layer of
the endometrium in patients experiencing implantation
failure with co-existing severe diffuse or the adenomyoma variant of adenomyosis. A previous histological study
has described such an increase within the myometrium
of women with adenomyosis, but did not comment on
endometrial macrophage numbers (Ota et al., 1991). A
more recent histological study also identied signicant
numbers of macrophages in the eutopic endometrium of
patients with endometriosis, myomas and adenomyosis,
but the authors did not comment on any signicant
difference in endometrial macrophage density between
these pathological groups (Khan et al., 2010). Interestingly,
several investigators have reported a link between the
presence of endometriosis on laparoscopy and an increase
in macrophage density in the eutopic endometrium (Berbic
et al., 2009; Khan et al., 2004; Leiva et al., 1994). Only
one study to date has reported a signicant decrease
in macrophage density in the eutopic endometrium of
women with endometriosis (Braun et al., 2002).

62

K.P. Tremellen, P. Russell / Journal of Reproductive Immunology 93 (2012) 5863

As both endometriosis and adenomyosis have been

linked to impaired implantation potential (Barnhart et al.,
2002; Ferrero et al., 2009; Maubon et al., 2010; Tremellen
and Russell, 2011), it is possible that the observed
increase in macrophage density may represent a common underlying pathological mechanism by which these
cells contribute to a hostile immune environment for
the implanting embryos. Whilst not all subjects in this
study underwent laparoscopy to identify the presence of
endometriosis, those patients with symptoms suggestive of
endometriosis usually did. Furthermore, pelvic MRI is a relatively good imaging modality to non-invasively diagnose
severe endometriosis (endometriomas, thick peritoneal
deposits). Taken together, it is unlikely that we have
signicantly under-estimated the existence of signicant endometriosis disease co-existing with adenomyosis.
Finally, the observation that only severe adenomyosis, not
mild focal disease, is linked with an increased endometrial macrophage density makes it less likely that mild
undiagnosed endometriosis is primarily responsible for the
observed macrophage inltration.
It is presently unknown how adenomyosis stimulates an increased endometrial macrophage density or
whether, indeed, these phenomena are related or merely
co-incidental. A recent study has reported that long term
GnRH agonist down-regulation therapy can produce a
very signicant fall in endometrial macrophage numbers,
with a co-existent decline in endometrial tissue production of MCP-1, a cytokine known to be chemotactic for
macrophages (Khan et al., 2010). Furthermore, this study
reported that down-regulation therapy produced a marked
reduction in endometrial capillary density, a parameter
that is known to be signicantly increased in the adenomyotic endometrium (Ota and Tanaka, 2003). Taken together,
these studies suggest that under local control of oestrogen there is an increase in endometrial capillary density
and production of cytokines chemotactic for macrophages,
both likely to result in a net inux of macrophages into the
endometrium. Adenomyosis is known to be characterised
by a local state of oestrogen dominance (Kitawaki, 2006).
Whilst serum levels of oestrogen do not differ between
women with or without adenomyosis, menstrual blood
oestradiol levels have been reported to be raised in the adenomyosis group, reecting an increased level of aromatase
expression in the adenomyotic endometrium (Kitawaki,
2006; Takahashi et al., 1989). An excess of oestrogen action
in the endometrium may result in an increase in capillary
density and production of pro-inammatory cytokines,
resulting in an elevated endometrial macrophage density.
When appropriately activated, macrophages are known
to have the capacity to secrete cytokines such as TNF and
IL-1, plus reactive oxygen and nitrogen species, all potentially toxic to embryos (Agarwal et al., 2005; Haddad et al.,
1995). The endometrial environment of patients with adenomyosis has been reported to contain elevated levels of
nitric oxide (Ota et al., 1998), a free radical chemical
linked with impaired embryo development and poor pregnancy rates (Lee et al., 2004). Furthermore, endometrial
biopsies taken from patients with adenomyosis contain
elevated amounts of anti-oxidant enzymes such as catalase, superoxide dismutase and glutathione peroxidase, a

clear sign of local oxidative stress caused by excessive reactive oxygen species production (Agarwal et al., 2005; Ota
et al., 2002). These reactive oxygen and nitrogen species
will directly damage embryos and have been reported to
increase local production of prostaglandin F2, leading to
an increase in myometrial contractility that is likely to
further impede the implantation process (Agarwal et al.,
2005).
Whilst the cause and effect relationship, if any, between
severe adenomyosis and increased macrophage populations in the endometrium requires elucidation, our novel
observation now provides a pathological mechanism by
which adenomyosis may impede successful implantation of an embryo. However, as these observations are
only preliminary ndings of a small retrospective case
series, they must be veried in a larger prospective study
before rm conclusions can be made. Furthermore, in the
future it would be useful to correlate the immunohistochemistry dened macrophage inltrations seen with
adenomyosis with changes in endometrial inammatory
mediators (cytokines and prostaglandins) and the production of reactive oxygen species, as these observations
would help strengthen the causative link between adenomyosis and failure of implantation of good quality embryos.
References
Agarwal, A., Gupta, S., Sharma, R.K., 2005. Role of oxidative stress in female
reproduction. Reprod. Biol. Endocrinol. 3, 28.
Barnhart, K., Dunsmoor-Su, R., Coutifaris, C., 2002. Effect of endometriosis
on in vitro fertilization. Fertil. Steril. 77, 11481155.
Berbic, M., Schulke, L., Markham, R., Tokushige, N., Russell, P., Fraser, I.S.,
2009. Macrophage expression in endometrium of women with and
without endometriosis. Hum. Reprod. 24, 325332.
Braun, D.P., Ding, J., Shen, J., Rana, N., Fernandez, B.B., Dmowski, W.P., 2002.
Relationship between apoptosis and the number of macrophages in
eutopic endometrium from women with and without endometriosis.
Fertil. Steril. 78, 830835.
Copperman, A.B., Wells, V., Luna, M., Kalir, T., Sandler, B., Mukherjee, T.,
2006. Presence of hydrosalpinx correlated to endometrial inammatory response in vivo. Fertil. Steril. 86, 972976.
Ferrero, S., Anserini, P., Abbamonte, L.H., Ragni, N., Camerini, G.,
Remorgida, V., 2009. Fertility after bowel resection for endometriosis.
Fertil. Steril. 92, 4146.
Haddad, E.K., Duclos, A.J., Baines, M.G., 1995. Early embryo loss is associated with local production of nitric oxide by decidual mononuclear
cells. J. Exp. Med. 182, 11431151.
Kano, T., Furudono, M., Nabetani, M., 1997. The incidence of endometriosis and adenomyosis in patients with habitual abortion in relation to
immunological abnormalities. Jpn. J. Fertil. Steril. 42, 113118.
Khan, K.N., Kitajima, M., Hiraki, K., Fujishita, A., Sekine, I., Ishimaru, T.,
Masuzaki, H., 2010. Changes in tissue inammation, angiogenesis and
apoptosis in endometriosis, adenomyosis and uterine myoma after
GnRH agonist therapy. Hum. Reprod. 25, 642653.
Khan, K.N., Masuzaki, H., Fujishita, A., Kitajima, M., Sekine, I., Ishimaru, T., 2004. Differential macrophage inltration in early and
advanced endometriosis and adjacent peritoneum. Fertil. Steril. 81,
652661.
Kissler, S., Hamscho, N., Zangos, S., Wiegratz, I., Schlichter, S., Menzel, C.,
Doebert, N., Gruenwald, F., Vogl, T.J., Gaetje, R., Rody, A., Siebzehnruebl, E., Kunz, G., Leyendecker, G., Kaufmann, M., 2006. Uterotubal
transport disorder in adenomyosis and endometriosis a cause for
infertility. BJOG 113, 902908.
Kitawaki, J., 2006. Adenomyosis: the pathophysiology of an oestrogendependent disease. Best Pract. Res. Clin. Obstet. Gynaecol. 20,
493502.
Kunz, G., Beil, D., 2010. Characterization of the uterine junctional zone
prior to IVF/ICSI: an observational study. Arch. Gynecol. Obstet. 281,
945953.
Kunz, G., Beil, D., Huppert, P., Noe, M., Kissler, S., Leyendecker, G., 2005.
Adenomyosis in endometriosis prevalence and impact on fertility.

K.P. Tremellen, P. Russell / Journal of Reproductive Immunology 93 (2012) 5863

Evidence from magnetic resonance imaging. Hum. Reprod. 20,
23092316.
Lee, T.H., Wu, M.Y., Chen, M.J., Chao, K.H., Ho, H.N., Yang, Y.S., 2004. Nitric
oxide is associated with poor embryo quality and pregnancy outcome
in in vitro fertilization cycles. Fertil. Steril. 82, 126131.
Leiva, M.C., Hasty, L.A., Lyttle, C.R., 1994. Inammatory changes of the
endometrium in patients with minimal-to-moderate endometriosis.
Fertil. Steril. 62, 967972.
Leyendecker, G., Kunz, G., Kissler, S., Wildt, L., 2006. Adenomyosis and
reproduction. Best Pract. Res. Clin. Obstet. Gynaecol. 20, 523546.
Maubon, A., Faury, A., Kapella, M., Pouquet, M., Piver, P., 2010. Uterine
junctional zone at magnetic resonance imaging: a predictor of in vitro
fertilization implantation failure. J. Obstet. Gynaecol. Res. 36, 611618.
Miura, S., Khan, K.N., Kitajima, M., Hiraki, K., Moriyama, S., Masuzaki, H.,
Samejima, T., Fujishita, A., Ishimaru, T., 2006. Differential inltration
of macrophages and prostaglandin production by different uterine
leiomyomas. Hum. Reprod. 21, 25452554.
Noyes, R.W., Hertig, A.T., Rock, J., 1975. Dating the endometrial biopsy.
Am. J. Obstet. Gynecol. 122, 262263.
Olive, D.L., Franklin, R.R., Gratkins, L.V., 1982. The association between
endometriosis and spontaneous abortion. A retrospective clinical
study. J. Reprod. Med. 27, 333338.
Ota, H., Igarashi, S., Hatazawa, J., Tanaka, T., 1998. Endothelial nitric oxide
synthase in the endometrium during the menstrual cycle in patients
with endometriosis and adenomyosis. Fertil. Steril. 69, 303308.
Ota, H., Igarashi, S., Maki, M., 1991. Distribution of tissue macrophages in
uterine muscle layers in patients with adenomyosis. Med. Sci. Res. 19,
473474.

63

Ota, H., Igarashi, S., Sato, N., Tanaka, H., Tanaka, T., 2002. Involvement
of catalase in the endometrium of patients with endometriosis and
adenomyosis. Fertil. Steril. 78, 804809.
Ota, H., Tanaka, T., 2003. Stromal vascularization in the endometrium
during adenomyosis. Microsc. Res. Tech. 60, 445449.
Peric, H., Fraser, I.S., 2006. The symptomatology of adenomyosis. Best
Pract. Res. Clin. Obstet. Gynaecol. 20, 547555.
Reinhold, C., Tafazoli, F., Wang, L., 1998. Imaging features of adenomyosis.
Hum. Reprod. Update 4 (4), 337349.
Russell, P., Anderson, L., Lieberman, D., Tremellen, K., Yilmaz, H., Cheerala,
B., Sacks, G., 2011. The distribution of immune cells and macrophages
in the endometrium of women with recurrent reproductive failure. I:
techniques. J. Reprod. Immunol. 91 (12), 90102.
Takahashi, K., Nagata, H., Kitao, M., 1989. Clinical usefulness of determination of estradiol level in the menstrual blood for patients with
endometriosis. Nippon Sanka Fujinka Gakkai Zasshi 41, 18491850.
Tamai, K., Koyama, T., Umeoka, S., Saga, T., Fujii, S., Togashi, K., 2006. Spectrum of MR features in adenomyosis. Best Pract. Res. Clin. Obstet.
Gynaecol. 20, 583602.
Tremellen, K., Russell, P., 2011. Adenomyosis is a potential cause of recurrent implantation failure during IVF treatment. Aust. N. Z. J. Obstet.
Gynaecol. 51, 280283.
Turnbull, L.W., Rice, C.F., Horsman, A., Robinson, J., Killick, S.R., 1994. Magnetic resonance imaging and transvaginal ultrasound of the uterus
prior to embryo transfer. Hum. Reprod. 9, 24382443.
Vercellini, P., Vigano, P., Somigliana, E., Daguati, R., Abbiati, A., Fedele,
L., 2006. Adenomyosis: epidemiological factors. Best Pract. Res. Clin.
Obstet. Gynaecol. 20, 465477.