Best Practice & Research Clinical Obstetrics and Gynaecology 28 (2014) 621636

Contents lists available at ScienceDirect

Best Practice & Research Clinical

Obstetrics and Gynaecology
journal homepage: www.elsevier.com/locate/bpobgyn

Ultrasound imaging in the management of

bleeding and pain in early pregnancy
Jure Knez, MD, Clinical Fellow,
Andrea Day, MBBS, Specialist Registrar in Obstetrics and
Gynaecology,
Davor Jurkovic, MD, FRCOG, Consultant Gynaecologist, Director
of Gynaecology Diagnostic and Outpatient Treatment Unit *
Gynaecology Diagnostic and Outpatient Treatment Unit, Elizabeth Garrett Anderson Wing, University
College Hospital, 235 Euston Road, London NW1 2BU, UK

Keywords:
Bleeding and pain are experienced by 20% of women during the
early pregnancy rst trimester of pregnancy. Although most pregnancies compli-
ultrasound cated by pain and bleeding tend to progress normally, these
miscarriage symptoms are distressing for woman, and they are also associated
ectopic pregnancy with an increased risk of miscarriage and ectopic pregnancy. Ul-
safety trasound is the rst and often the only diagnostic modality that is
used to determine location of early pregnancy and to assess its
health. Ultrasound is an accurate, safe, painless and relatively
inexpensive diagnostic tool, which all contributed to its widespread
use in early pregnancy. Pain and bleeding in early pregnancy are
sometimes caused by concomitant gynaecological, gastrointestinal,
and urological problems, which could also be detected on ultra-
sound scan. In women with suspected intra-abdominal bleeding,
ultrasound scan can be used to detect the presence of blood and
provide information about the extent of bleeding.
In this chapter, we comprehensively review the use of ultrasound
in the diagnosis and management of early pregnancy complica-
tions. We include information about the diagnosis of gynaeco-
logical and other pelvic abnormalities, which could cause pain or
bleeding in pregnancy. We also provide a summary of the current
views on the safety of ultrasound in early pregnancy.
2014 Elsevier Ltd. All rights reserved.

* Corresponding author. Tel.: 207 725 0521; Fax: 20 7486 7019.

E-mail addresses: davor.jurkovic@nhs.net, Jurkovic.davor@gmail.com (D. Jurkovic).

http://dx.doi.org/10.1016/j.bpobgyn.2014.04.003
1521-6934/ 2014 Elsevier Ltd. All rights reserved.
622 J. Knez et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 28 (2014) 621636

Introduction

Pain and vaginal bleeding are experienced by about one in ve women during the rst trimester of
pregnancy [1]. The symptoms may vary from light painless bleeding to severe pain, accompanied by
haemorrhagic shock. Clinical presentation is helpful in determining the likely cause of womens
symptoms, but, in most cases, ultrasound imaging is required to reach a conclusive diagnosis. About
one-half of pregnancies complicated by bleeding will progress normally beyond the rst trimester,
with the remaining women suffering an early pregnancy failure [2]. A total of 1015% women pre-
senting with vaginal bleeding have an ectopic pregnancy [3]. A timely and accurate diagnosis of ectopic
pregnancy is important to minimise the risk of serious adverse outcomes. It is important to remember
that bleeding may also be caused by conditions that are coincidental to pregnancy, such and cervical
polyps, vulval and vaginal lesions, and injury during sexual intercourse. Abdominal pain is often a late
symptom in women with miscarriages and ectopic pregnancy. It may also be caused by intra-
abdominal haemorrhage caused by a ruptured corpus luteum cyst or by ovarian torsion. Non-
gynaecological causes of pain include acute cystitis, ureteric stones, appendicitis, and constipation.
Differential diagnosis between these conditions is difcult, based on clinical symptoms alone, and
some form of imaging is almost always required to reach the correct diagnosis.
The development of high-resolution transvaginal ultrasound technology in the 1980s has enabled
signicant progress in early pregnancy imaging. Since then, transvaginal ultrasonography has become
the accepted standard for examining women with suspected complications of early pregnancy [4,5].
With advances in ultrasound diagnosis, pregnancies could be identied earlier in gestation compared
with transabdominal ultrasound, resulting in better diagnostic accuracy. Diagnosis of early pregnancy
complications has been further enhanced by combining ultrasound ndings with the results of
biochemical tests such as human chorionic gonadotrophin (hCG) and progesterone. In this chapter, we
aim to provide a summary of recent development in ultrasound imaging in pregnancy, with plenty of
practical advice to those providing clinical care to women with suspected early pregnancy complications.

Normal early intrauterine pregnancy

When performing ultrasound scan in early pregnancy, it is important to conrm the location of
pregnancy, its viability, and gestational age [6]. Although the ultrasound diagnosis of intrauterine
pregnancy is generally considered to be straightforward, criteria to diagnose an intrauterine pregnancy
are conspicuously absent from the literature. This makes it hard to differentiate between a pregnancy
that is normally implanted within the uterine cavity and several types of ectopic pregnancies conned
to the uterus, such as interstitial, cervical, or intramural pregnancies. A normal intrauterine pregnancy
should be located within the uterine cavity, which represents a virtual space lined with endometrium
extending from the internal cervical orice to tubal ostia. In addition, the trophoblast should not be
invading beyond the endometrial myometrial junction [7,8].
A normal intrauterine pregnancy can be consistently visualised on ultrasound scan 3 weeks after
conception (i.e. 5 weeks after last menstrual period in a woman with regular 28-day cycle) [9]. When
assessing the uterus, the probe is moved in the transverse plane from the internal orice to the fundus
in search of an intrauterine pregnancy. Longitudinal view is then used to show the location of the
gestational sac beneath the endometrial surface and to conrm its intrauterine location by demon-
strating a communication between the cervical canal and the uterine cavity. If a gestational sac is not
immediately visible, the endometrial morphology should be assessed to search for the signs of retained
products. Endometrial thickness measurements are not useful to discriminate between an early in-
trauterine pregnancy, incomplete miscarriage, and ectopic pregnancy [10]. Sweeping the probe both
side to side and up and down will facilitate the assessment of the cervix, caesarean section scar (if
present), myometrium, and interstitial portion of the Fallopian tubes, to rule out unusual sites of
ectopic pregnancy implantation.
In a normal pregnancy, the gestational sac rst becomes visible on transvaginal scan at 4 weeks and
3 days of gestation (Fig. 1a) [9]. The sac usually appears as a circular structure with a thick echogenic
rim surrounding a clear anechoic centre, which represents early chorionic cavity. The sac is located just
 J. Knez et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 28 (2014) 621636 623

Fig. 1. Examples of normal rst-trimester intrauterine pregnancies. (a) A 4-week gestational sac is seen implanted just below the
surface of the endometrium (mid-line echo), surrounded by echogenic trophoblast; (b) a 5-week pregnancy implanted in the fundal
area of the uterine cavity with a visible yolk sac; (c) a small embryo is visible in a pregnancy corresponding to 6 weeks gestation.
Embryonic heart rate is measured using M-mode ultrasound; (d) a three-dimensional scan showing a longitudinal view of a 7-weeks
old embryo. The head is clearly visible with a prominent fourth ventricle (arrow). FHR, fetal heart rate. a) GS gestational sac; b) YS
yolk sac.

below the midline echo, buried into the decidualised endometrium. Myometrial cysts in women with
adenomyosis may sometimes appear similar to an early gestational sac. They are typically located
within the myometrium beyond the endometrialmyometrial junction, which facilitates the differ-
ential diagnosis [11]. Accumulation of blood within the uterine cavity may also resemble an early
gestational sac. This is described as a pseudosac, which is particularly common in ectopic pregnancy,
and it should not be confused with an intrauterine gestation. A pseudosac appears like an elongated
structure inside the cavity, surrounded by a single decidual layer. On colour Doppler examination, the
pseudosac is avascular and its shape may change during the scan, whereas a normal pregnancy tends to
have high peripheral blood ow and its shape is stable. The gestational sac is measured using the mean
of three perpendicular diameters. In early gestation, the sac grows about 1 mm in mean diameter per
day. A large, empty, slowly growing and irregular gestational sac is suggestive of pregnancy failure (this
will be discussed in more detail in the miscarriage section).
The yolk sac becomes visible within the gestational sac (i.e. chorionic cavity) from 5 weeks
gestation (Fig. 1b). Measurement technique is similar to that of the gestational sac, always including
three perpendicular diameters from the centre of the yolk sac wall. If the yolk sac is large (>5.6 mm)
[12] or not visible when the mean gestational sac diameter reaches over 13 mm [13], repeat ultrasound
in a week is advised, as these ndings are strongly associated with early pregnancy failure.
Early embryo is typically located adjacent to the yolk sac, and it appears as a linear bright echo
(Fig. 1c). The two structures are connected by the vitelline duct. From 7 weeks gestation onwards,
crown (head) can be distinguished from rump (trunk) (Fig. 1d). Crownrump length should be
measured from a sagittal section of the embryo, with care to avoid including the yolk sac into the
measurement [14].
624 J. Knez et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 28 (2014) 621636

Cardiac activity can be documented at around 5 weeks and 5 days gestation, when the embryo
measures 25 mm in length. Background movement and maternal pulsation can sometimes be mis-
interpreted as embryonic cardiac activity. Heart rate should be measured using M-mode; pulsed
Doppler examination produces high-energy acoustic outputs and should not be used in early rst-
trimester pregnancy (Fig. 1c) [15]. It has been reported that heart rates below 100 beats per minute
are suggestive of an abnormal pregnancy [16,17]. It is important to recognise, however, that, in early
normal gestations, heart rate could be below 100 beats per minute, yet increase rapidly between 67
weeks gestation. If a heartbeat is not visible in an embryo, colour Doppler may be useful to conrm a
negative nding provided it is used for a short period of time with both mechanical and thermal index
set to less than 1.
From 7 weeks gestation, the amniotic cavity becomes visible and should be measured in three
dimensions. In addition, rhombencephalon and spine may be distinguished, and the umbilical cord
seen. At 8 weeks gestation, forebrain, midbrain, hindbrain and skull are apparent, and limb buds start
to grow. The amniotic cavity expands, whereas the umbilical cord and vitelline duct lengthen and a
mid-gut hernia appears. Between 79 weeks gestation, amniotic and chorionic cavities are not fused,
and chorionicity and amnionicity of multiple pregnancies can be condently established.

Multiple pregnancy

The diagnosis of multiple pregnancy in the rst trimester is important, as determination of cho-
rionicity and amnionicity is relatively simple, and women could be advised about the risk associated
with different types of multiple pregnancies [9].
Dizygotic twins develop from two different oocytes fertilised by two different sperm, and they
implant separately. As a result, each embryo has its own gestational sac (chorion), amniotic sac
(amnion), and placenta. Consequently, dizygotic twins are dichorionic and diamniotic (Fig. 2a).
Monozygotic twins arise from a single oocyte fertilised by one sperm, which then divides. The stage of
development at which it divides is what determines its chorionicity and amnionicity. If it divides before
implantation, it results in a dichorionic and diamniotic pregnancy. If the division occurs after im-
plantation, a single gestational sac is formed, hence pregnancy is monochorionic. If division occurs on
days 48 of development, the pregnancy is monochorionic and diamniotic, and if it occurs on days 913
of development, the pregnancy is monochorionic and monoamniotic (Fig. 2b).
On ultrasound, the rst signs of a multiple pregnancy are the presence of more than one gestational
sac at around 5 weeks gestation. The number of gestational sacs and yolk sacs may be variable, and it
does not always correlate with the number of embryos in multiple pregnancies [9]. It is imperative to
examine the entire chorionic cavity to ensure that all embryos are seen when pregnancy progresses
beyond 6 weeks gestation. Amnionicity should be assessed at around 7 weeks gestation when the
amnion is seen separate to the embryo. At this stage, chorion and amnion are not fused, so that

Fig. 2. Multiple pregnancy. (a) Dichorionic twins at 7 weeks gestation. Two gestational sacs are present, which are separated by a
thick chorionic membrane; (b) monochorionic, monoamniotic twins at 7 weeks gestation. Both embryos lie in a single amniotic
cavity. Amniotic membrane is clearly seen. CM chorionic membrane; AM amniotic membrane.
 J. Knez et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 28 (2014) 621636 625

chorionicity, and amnionicity of multiple pregnancies can be easily and condently established. After
amnion and chorion are fused, the assessment of chorionicity becomes more difcult, and it relies on
the assessment of thickness of dividing membrane and shape of its junction with the uterine wall
(lambda or T sign) [9]. If the dividing membrane between two embryos or fetuses within the same
gestational sac is absent, the pregnancy is classied as monochorionic and monoamniotic. In higher
multiples, the same principles are applied.

Miscarriage

Miscarriage is dened as a pregnancy failure occurring before the completion of 24 weeks gesta-
tion, and it is the most common complication of pregnancy. The reported rate of pregnancy loss in
women with a missed menstrual period and positive pregnancy test is 1224% [18]. The true rate of
miscarriage is probably even higher considering the preclinical pregnancy losses, before a menstrual
period is missed. Miscarriage can be suspected in women presenting with vaginal bleeding or
abdominal pain, but ultrasound examination is usually required to diagnose early pregnancy failure.
Presence of an embryonic heartbeat at the time of the initial ultrasound does not always indicate that
the pregnancy will progress normally. It has been reported that as many as 12.2% may experience
miscarriage after positive detection of embryonic cardiac activity in singleton pregnancies [19].

Early embryonic (fetal) demise

Early embryonic demise (also described as missed miscarriage or blighted ovum) refers to the early
stage in the natural history of a miscarriage. On ultrasound scan, an intact gestational sac is seen within
the uterine cavity, which does not contain an embryo or there is a visible embryo with no cardiac
activity (Fig. 3a). The main difculty when diagnosing early embryonic demise is to avoid confusing a
healthy, early, normal intrauterine pregnancy with a miscarriage. As the ultrasound diagnosis of early
fetal demise is based on negative ndings, the risk of diagnostic errors is considerable. The risk is
particularly high in women who are unsure of their dates, have irregular cycles, have conceived while
taking hormonal contraception, or have had less than three menstrual periods since their last preg-
nancy. Uterine abnormalities, such as congenital uterine anomaly, uterine broids, intra-abdominal
adhesions after previous caesarean sections, or other pelvic surgery affecting uterine position, in-
crease the risk of misdiagnosis. A recent systematic review [20] showed a lack of high-quality, pro-
spective data on which to base guidelines for the accurate diagnosis of early embryonic demise. A wide
range of cut-off points for the size of the gestational sac or embryo above which embryonic cardiac
activity should be visible in a normal early intrauterine pregnancy have been proposed. Although they
all include a certain margin of safety, the main cause of misdiagnosis is operator error, and this can
occur irrespective of a chosen cut-off value. Recent National Institute for Health and Care Excellence
guidelines suggest that fetal demise should be suspected in the absence of heartbeat when crown
rump length greater than 7 mm or gestational sac measuring greater than 25 mm without a visible
embryo [1]. National Institute for Health and Care Excellence recommend that, in all cases, the diag-
nosis should be conrmed at a follow-up visit 714 days later or by a second observer to minimise the
risk of diagnostic errors.
Other morphological features have been described, which are not diagnostic of miscarriage, but
they indicate that the risk of miscarriage occurring is increased. These include visualisation of an
amniotic and yolk sac without a viable embryo, irregular gestational sac with a thin trophoblastic layer,
early growth retardation [21], discrepancies between size of gestational sac and embryo [22,23], and
bradycardia (heart rate below the 5th percentile or less than 85 beats per minute) [24].
In multiple pregnancies, a signicant discrepancy in crownrump length between twins may be
associated with loss of one twin [25]. In some cases, an empty gestational sac may be seen alongside a
second one containing an embryo. This is sometimes referred to as vanishing twin syndrome [26,27].
An intrauterine subchorionic haematoma may sometimes resemble an empty gestational sac, and care
has to be taken to avoid misinterpreting this nding as a vanishing twin. The presence of a clearly
dened layer of trophoblast surrounding the sac and a nding of two or more corpora lutea support the
diagnosis of a multiple pregnancy.
626 J. Knez et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 28 (2014) 621636

Fig. 3. Ultrasound imaging of miscarriage. (a) Ultrasound image of an embryo corresponding to 78 weeks gestation. On colour
Doppler, there was no evidence of cardiac activity, which is conclusive of an early embryonic demise; (b) incomplete miscarriage. A
longitudinal section of the uterus showing a small amount of hyperechoic tissue within the uterine cavity; (c) on Doppler exami-
nation, the tissue is highly vascular, which conrms the diagnosis of an incomplete miscarriage. Colour Doppler can be used in non-
viable pregnancies, and it facilitates differentiation between blood clots and retained products of conception. E embryo.

Incomplete miscarriage

Incomplete miscarriage is dened by the presence of retained products of conception without a

well-dened gestation sac. Blood clots within the uterine cavity are often present in women who are
bleeding, and they are sometimes difcult to differentiate from trophoblastic tissue. Retained products
are usually seen as a well-dened area of hyperechoic tissue within the endometrial cavity (Fig. 3b)
[28] and frequently demonstrate increased vascularity on colour Doppler examination (Fig. 3c). Blood
clots on the other hand, are ill dened and avascular. The ultrasound diagnosis of incomplete
miscarriage is difcult, and no consensus exists about the best diagnostic criteria. Endometrial thick-
ness (measured as the anteriorposterior diameter of the uterine cavity) is often used in clinical
practice to aid with this diagnosis. A number of different cut-off levels have been proposed ranging
from 525 mm [29,30]. A recent prospective observational study, however, showed that none of these
criteria are accurate enough to diagnose the presence of chorionic villi within the uterine cavity [31].
Subjective assessment of the morphological characteristics of the tissue within the uterine cavity,
combined with colour Doppler assessment of its vascularity, has been proposed to overcome the
limitations of using cut-off measurements [32]. No good-quality prospective studies, however, have
determined the accuracy of these diagnostic criteria.

Complete miscarriage

The diagnosis of complete miscarriage is made in women in with no signs of any pregnancy tissue
within the uterine cavity on ultrasound scan. This diagnosis can be made with condence only in
women who had clear evidence of intrauterine pregnancy on previous ultrasound examinations. If no
scan has previously been carried out, the pregnancy should be described as a pregnancy of unknown
location and followed up with serial serum human chorionic gonadotrophin (b-hCG) measurements
 J. Knez et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 28 (2014) 621636 627

[1]. A decline in serum b-hCG by 13% or more over 48 h suggests a failing pregnancy [33]. Serum
progesterone reading less than 20 nmol/l is equally accurate in predicting a failing pregnancy without
the need for another blood test 2 days after the initial visit [34]. In women taking exogenous pro-
gestogens, however, measurements of serum progesterone are not reliable. Low hCG ratio and low
progesterone are used to estimate the probability of a pregnancy of unknown location resolving
spontaneously. They are not test that could rule out a possibility of an ectopic pregnancy. In view of
that, biochemical markers must be assessed in conjunction with clinical and ultrasound information
and should not be used alone as a diagnostic tool.

Molar pregnancies

In modern clinical practice, most molar pregnancies present during the rst trimester of pregnancy
with clinical and ultrasound signs of early pregnancy failure. Molar pregnancies have a potential to
develop into persistent trophoblastic disease (invasive hydatidiform mole, choriocarcinoma, placental
site trophoblastic tumour) and their early detection on ultrasound helps to optimise the management
and follow up. Ultrasound feature strongly suggestive of a complete molar pregnancy is thick, cystic
tissue within the uterine cavity without a visible gestational sac. Partial hydatidiform mole is often
suspected in women with intact gestational sac with cystic placental changes. The accuracy of ultra-
sound to diagnose molar pregnancy is difcult to assess in modern practice, as most miscarriages are
managed conservatively, and histological conrmation of diagnosis is available only in a few women.
The available data, however, show that ultrasound diagnosis of complete molar pregnancy is sensitive,
with the reported detection rates between 80 and 95% [35,36].
The diagnosis of partial molar pregnancy is less accurate, with the detection rates between 20 and
30%. This indicates that a signicant number of partial molar pregnancies may be managed non-
surgically in modern practice without the diagnosis ever being made. As the risk of persistent
trophoblastic disease is much lower in partial compared with complete moles (0.5% v 15%), the inability
to detect partial moles on ultrasound is unlikely to have severe adverse effects on maternal health [37].

Ectopic pregnancy

An ectopic pregnancy is any gestation that is implanted outside the uterine cavity. The prevalence
varies from 1120 per 1000 live births in developed countries, although it can be as high as 4% in
assisted conception populations [38]. Fallopian tubes are the most common site for the ectopic
pregnancy to implant, with 9398% of ectopic pregnancies being located there. Thus, the term tubal
pregnancy and ectopic pregnancy are often used interchangeably [3942]. A total of 57% of ectopic
pregnancies implant within the uterine wall, but outside the uterine cavity. Although rare, these non-
tubal ectopics are associated with signicantly higher mortality and morbidity than tubal ectopics, as
they are often difcult to diagnose and tend to present late with sudden rupture.
The triad of pain, vaginal bleeding, and amenorrhea was historically used to suspect a diagnosis of
ectopic pregnancy [43]. The amount of bleeding associated with ectopic pregnancy varies, although
classically the patient would present with brown discharge. Heavy bleeding, in the absence of further
ultrasound examination may sometimes lead to a misdiagnosis of miscarriage. Moreover, 1020% of
women report no bleeding [44]. Abdominal pain is usually a late feature in the clinical presentation of
ectopic pregnancy, and typically follows tubal rupture or tubal miscarriage with bleeding through the
mbrial end of the tube into the peritoneal cavity. The severity of pain does not necessarily correlate
with the amount of blood present in the abdominal cavity. Nevertheless, about 10% of women diag-
nosed with ectopic pregnancy do not report any abdominal pain [44]. Less common features of ectopic
pregnancy include nausea, vomiting, and diarrhoea. Hence, the symptoms of ectopic pregnancy are
often non-specic and difcult to differentiate from those of other gynaecological, intestinal, or uro-
logical disorders. A recent meta-analysis showed that physical examination, including the assessment
of cervical motion tenderness, detects less than 50% of tubal ectopics [45].
The potential value of ultrasound for the diagnosis of ectopic pregnancy was rst described in 1969
[46]. Initially, an ectopic pregnancy was suspected in all women with no conclusive evidence of in-
trauterine pregnancy on the scan. The use of high-resolution transvaginal ultrasound has changed the
628 J. Knez et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 28 (2014) 621636

diagnostic approach to one based on visualising the ectopic mass [47]. Transvaginal scanning has been
demonstrated to be superior to transabdominal ultrasound, with sensitivities for the diagnosis of
ectopic pregnancy in early studies of 7780% for transabdominal and 8890% for transvaginal ultra-
sound [5,48]. As expertise and equipment has improved since then, the sensitivity of transvaginal
ultrasound for the diagnosis of ectopic pregnancy has increased further [47,49].

Tubal ectopic pregnancy

Several observational studies have shown that presence of an adnexal mass on transvaginal ul-
trasound is highly specic for tubal pregnancy (Fig. 4a) [49]. A meta-analysis of ndings in 2216 women
showed that the presence of an adnexal mass other than a simple cyst separate from the ovary was a
highly sensitive (84.4%) and specic (98.9%) test for the diagnosis of ectopic pregnancy [50]. A more
recent systematic review reported similar results (sensitivity 88% and specicity 99%) [45].
Despite high sensitivity of the method, it is important to note that not every woman with ectopic
pregnancy will be diagnosed by transvaginal ultrasound, and many not at the rst visit. A study
comparing the features of ectopic pregnancy diagnosed at the rst visit with those rst classied as a
pregnancy of unknown location concluded that the latter pregnancies were small and probably too
early in their natural history to be visualised, rather than being missed [51]. Diagnostic difculties may
arise if there is complex adnexal pathology, large tender stimulated ovaries or a non-specic adnexal
mass made up of blood clots surrounding the ectopic gestation. The adnexal swelling suspicious of
ectopic pregnancy needs to move separately from the ovary during palpation with the ultrasound
probe. This sliding organs sign helps to distinguish ectopic pregnancy from a cystic corpus luteum
[52]. In addition, colour Doppler can be useful both in identication of a mass and conrmation of
trophoblastic ow within it. Corpora lutea demonstrate a vascular rim with the appearance of a ring of
re, and should not be mistaken for an ectopic pregnancy. In spontaneous conceptions, the location of
the corpus luteum is important, as 78% of ectopics are ipsilateral to it [53].
A group of experts from Europe and USA proposed that the term denite ectopic pregnancy should
only be used if a yolk sac, embryo, or both, are seen on ultrasound [54]. In the case of inhomogeneous
adnexal mass or extrauterine gestational sac-like structure, probable ectopic pregnancy should be
used [54]. This, more conservative approach for the diagnosis of denite ectopic pregnancy is based on
the fact that, although falsepositive diagnoses related to ultrasound imaging are rare, the conse-
quences of misdiagnosis could be serious. If methotrexate is used in such cases, this may lead to
inadvertent termination of an undetected intrauterine pregnancy or severe abnormalities in surviving
pregnancies [55].
A number of other ndings may suggest the presence of ectopic pregnancy, but they are not
diagnostic. The presence of uid collection (pseudosac) in the absence of adnexal mass has only 0.02%

Fig. 4. Tubal ectopic pregnancy. (a) A transverse section through the pelvis shows a gestational sac in the left Fallopian tube. Tube is
distended with blood, indicating haematosalpinx; (b) longitudinal section through the pelvis shows an empty uterus with blood
present in the utero-vesical pouch and in the pouch of Douglas after rupture of tubal ectopic pregnancy. GS gestational sac; B
blood.
 J. Knez et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 28 (2014) 621636 629

probability of ectopic pregnancy [56]. The presence of echogenic uid in the pelvis has been reported in
2856% of women with ectopic pregnancy [57,58]. This nding correlates with the surgical ndings of
haemoperitoneum, but not necessarily as a result of tubal rupture, as bleeding can be the result of tubal
miscarriage or rupture of a haemorrhagic ovarian cyst.
Various amounts of clear anechoic uid within the lesser pelvis are seen in most pregnant women,
but only the presence of blood constitutes a signicant nding (Fig. 4b) [57,58]. Blood clots appear
hyperechoic and irregular on scan, and may be mistaken for bowel loops. Checking for the presence of
peristalsis helps make the differential diagnosis. Blood clots are also usually covered with hypoechoic
serum and appear jelly-like when gently pressed by the probe. In cases of severe intra-abdominal
bleeding, blood can be present in the upper abdomen below the liver (Morrisons pouch). A trans-
abdominal scan should be therefore performed in all cases of suspected ruptured ectopics to provide
more information about the extent of blood loss.
In women with non-diagnostic scans, biochemical markers have been used to try to facilitate
detection of ectopic pregnancy. A recent cross-sectional study, however, showed that a single reading
of serum hCG above a certain level (discriminatory zone) is not helpful in differentiating intrauterine
from ectopic pregnancies in symptomatic women [59]. A systematic review by van Mello et al. [60]
showed that neither single nor serial measurements enable an accurate detection of ectopic preg-
nancy [60]. The results of biochemical tests should be interpreted with particular caution in preg-
nancies after assisted reproduction because of a high risk of multiple pregnancy. In normal multiple
pregnancies, gestational sacs may not be visible at the usual hCG threshold applicable to singleton
pregnancies. In addition the risk of heterotopic pregnancy is increased, which may lead to erroneous
clinical decisions based on hCG readings.

Non-tubal ectopic pregnancy

Non-tubal ectopics (Fig. 5a) have signicantly higher maternal morbidity and mortality usually due
to delayed diagnosis [61]. Interstitial ectopics occur when the gestational sac implants on the inter-
stitial portion of the Fallopian tube. These women remain asymptomatic longer because a thick
myometrial layer surrounds the ectopic pregnancy, containing any bleeding and preventing early
rupture. On ultrasound, this is seen as myometrium from the uterine fundus enclosing the gestation
(Fig. 5b). The most characteristic feature on ultrasound is visualisation of the proximal interstitial tube
communicating with the medial aspect of the ectopic gestation and the lateral aspect of the uterine
cavity. Three-dimensional ultrasound could be used to facilitate a correct diagnosis [62].
Cornual ectopic pregnancy occurs only in women with a unicornuate uterus, where implantation
takes place in the rudimentary horn. On ultrasound scan, the uterine shape is abnormal, with a single
interstitial tube and an ectopic gestation adjacent to the uterus. The rudimentary horn is usually mobile
and the gestational sac appears enveloped in a continuous myometrial mantle [63]. Cervical and
caesarean scar pregnancies (Fig. 5c) are caused by previous surgical trauma, with implantation
occurring in the myometrial defect. Women typically present with painless vaginal bleeding. Caesarean
scar pregnancies are located close to the internal orice, whereas cervical pregnancies tend to be
located lower in the cervix [64]. Management of cervical and caesarean ectopics is similar, so differ-
entiation between the two is not essential. It is imperative, however, to distinguish a true extrauterine
implantation from the cervical phase of a spontaneous intra-uterine miscarriage or a low (isthmic)
implantation. Visualization of the insertion of the uterine artery may help determine the level of the
internal os. In a miscarriage the internal os might be open and gentle pressure with the probe may
cause movement of the sac. Lack of peri-trophoblastic blood ow also supports the diagnosis of a
detached pregnancy retained in the cervical canal. Ovarian pregnancies are rare, characterized by the
presence of a gestational sac surrounded by healthy ovarian tissue where gentle palpation does not
separate it from the ovary (Fig. 5d) [65]. The echogenicity and vascularity can be used to differentiate
products of conception from a corpus luteum. Doppler ultrasound is particularly helpful in women
with suspected ovarian pregnancies. The main purpose of Doppler is to identify corpus luteum with
certainty. The presence of two areas of blood ow within the ovary in a woman with an empty uterus
raises the possibility of ovarian pregnancy. Ovarian pregnancy tends to present as a focus of vascular,
cystic, hyperechoic tissue, which usually looks different from a corpus luteum. When ovarian
630 J. Knez et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 28 (2014) 621636

Fig. 5. Non-tubal sites of ectopic pregnancy. (a) Location of different types of non-tubal ectopic pregnancies; (b) the gestational sac
(I) is located high right and lateral to the uterine cavity (UC). Three-dimensional ultrasound facilitates visualisation of the interstitial
portion of the tube (arrow) adjoining the gestational sac and the uterine cavity. This allows differentiation between intrauterine
pregnancy and interstitial ectopic; (c) longitudinal section through the uterus showing a gestational sac (S), which is implanted into
a decient lower-segment caesarean section scar outside the uterine cavity (UC). The anterior wall of the uterus is herniating
anteriorly towards the bladder due to the loss of myometrial mantle; (d) ovarian ectopic pregnancy (O) can be differentiated from
the corpus luteum (CL) by using colour Doppler ultrasound; (e) abdominal ectopic pregnancy (A) secondary to tubal miscarriage. An
inhomogeneous mass is seen attached to the peritoneum of the pouch of Douglas (D). A, abdominal pregnancy; I, interstitial
pregnancy; O, ovarian pregnancy; S, caesarean section scar pregnancy.

pregnancy presents as an empty sac the differential diagnosis from a cystic corpus luteum could be
difcult. In ovarian ectopic, however, the layer of trophoblast surrounding the sac tends to be thicker
and more echogenic compared with corpus luteum.
Abdominal pregnancies implant in the peritoneal cavity, and may result from primary implantation
or be secondary to a tubal abortion with trophoblastic invasion and re-implantation (Fig. 5e). Diagnosis
is made when pregnancy is seen outside of the uterus, tubes and ovaries.

Other causes of pain in early pregnancy

Aside from pregnancy complications, abdominal pain can be the consequence of a wide variety of
conditions coincidental to pregnancy. These could be of gynaecological, gastrointestinal or genito-
urinary tract origin. Ultrasound imaging also plays a pivotal role in the diagnosis and management of
these conditions.

Uterine broids

Uterine broids are often detected during pregnancy. They are usually asymptomatic but, in a small
minority of women, they can cause signicant problems. Fibroids tend to increase rapidly in size during
pregnancy, which may result in pressure symptoms. Rapid growth sometimes outstrips blood supply to
 J. Knez et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 28 (2014) 621636 631

the broid, which can result in ischaemia and necrosis (broid degeneration). Ultrasound appear-
ances of broids vary, but they typically present as well-dened hypoechoic lesions representing
muscle bres and connective tissue, which commonly generate acoustic shadowing. Ultrasound ap-
pearances of broids undergoing degeneration are often non-specic, although later in the course of
the disease they may appear partially cystic and contain hypoechoic uid and septations. It is also
important to determine the location of the broids in relation to the uterine cavity, as the risk of rst-
trimester miscarriage is increased in women with submucous broids [66]. Very large broids, which
are located posteriorly and in the lower part of the uterus, may obstruct urethra and cause urinary
retention. They are also associated with an increased risk of late miscarriage and pre-term labour [67].

Ovarian cysts

Although true ovarian tumours are relatively rare in pregnancy, ovaries are often enlarged with
functional cysts. They are usually clinically silent but, in rare cases, they could be complicated by
haemorrhage or torsion (Fig. 6). On ultrasound, they usually appear rounded, unilocular, anechoic, with
smooth thin internal borders. They are surrounded by healthy ovarian tissue, and are avascular on
Doppler examination. The size may vary, and most resolve by 16 weeks gestation [68]. The corpus
luteum may be cystic, solid, or may contain mixed internal echoes (spider web) or ground-glass
appearances, which both represent haemorrhage within the cyst. Haemorrhagic ovarian cysts are
best managed expectantly, as they tend to resolve spontaneously within a short period of time. When
bleeding is suspected, examination of the pouch of Douglas and Morrisons pouch for the presence of
blood should always be performed to evaluate the extent of bleeding. In most women with ruptured
haemorrhagic ovarian cysts, however, blood loss tends to be limited and it is rarely severe enough to
cause cardio-vascular instability.

Fig. 6. Causes of pain and bleeding not directly related to pregnancy. (a) Ovary, enlarged with an anechoic cyst. Note the oedema (O)
of the surrounding ovarian tissue. These ndings are suggestive of ovarian torsion; (b) haemorrhagic cyst with mixed internal echoes
(spider web) appearance; (c) ureteric stone (arrow) is seen as a hyperechoic swelling in the pelvic segment of the left ureter (U) in a
women presenting with severe colicky left iliac fossa pain.
632 J. Knez et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 28 (2014) 621636

Ovarian torsion

Heavy ovaries are prone to twisting with various degrees of vascular and lymphatic obstruction,
causing ovarian congestion, oedema, and subsequent infarction. Pregnancy and ovarian stimulation are
both risk factors for torsion, and this represents the main complication of adnexal tumours, particularly
in the rst trimester. The prevalence of ovarian torsion in pregnancy varies between 0.05 and 0.6% [69].
Size and contents of an ovarian lesion contribute to the risk, and should always be considered.
McWilliams et al. [70] found that the risk of torsion is highest when the lesion measures 812 cm.
When considering the nature of the mass, endometriomas or invading tumours are less likely to un-
dergo torsion owing to reduced mobility secondary to adhesions or invasion. Hyperstimulated ovaries,
on the other hand, are more likely to tort, with a rate reported to be as high as 7.5% [71].
The clinical picture is the main diagnostic measure, and any woman with sudden onset of severe
abdominal pain should be assessed promptly. On ultrasound, the ovary appears enlarged and
oedematous (Fig. 6a); comparing with the contralateral ovary may be useful. On the other hand,
assessment of blood supply with Doppler ultrasound may not be helpful, as the arterial perfusion can
be maintained even in severe cases of torsion. In most cases of torsion, twisted pedicle can be visualised
using colour Doppler. This has been referred to as a whirlpool sign, which is another helpful nding in
suspected cases [72].

Gastrointestinal causes

Pregnant women are not at increased risk of suffering pain owing to gastrointestinal causes.
Because of physiologic changes, it is more difcult to arrive at a correct diagnosis in gravid women. The
gastrointestinal causes include appendicitis and other inammatory, infectious or obstructive pro-
cesses of the bowel. The most common reason for non-obstetric related surgery during pregnancy is
acute appendicitis, complicating 1 in 1500 pregnancies [73]. Early diagnosis is important, as 36% of
women with perforated appendix suffer fetal loss compared with 1.59% in those with unruptured
inamed appendix [74]. Symptoms include abdominal pain, anorexia, nausea, and vomiting. The
clinical diagnosis is often difcult, mainly due to enlarged gravid uterus, which displaces the appendix
upwards and the non-specicity of symptoms in pregnancy. Appendix can be visualised by ultrasound
in pregnancy, and the sonographic criteria for diagnosing appendicitis do not differ from the criteria
used in non-pregnant patients. The ultrasound examination of the appendix, however, is highly
operator-dependent, and is further limited by pregnant body habitus. The studies analysing the reli-
ability of ultrasound in visualisation of acute appendicitis during pregnancy are scarce and based on a
low number of patients; sensitivities range from 50100% and specicities from 96100% [75,76].

Urinary tract

Abdominal pain in pregnancy can also originate from the urinary tract. Infectious causes, such as
cystitis or pyelonephritis, do not require imaging techniques for the diagnosis. On the other hand,
ultrasound is usually the rst imaging technique used to diagnose and resolve the possible causes of
hydronephrosis. Although urolithiasis is not more common in pregnant than in non-pregnant women,
physiologic hydronephrosis in pregnancy should be differentiated from the hydronephrosis owing to
renal or ureteral calculi (Fig. 6c). The sensitivity of ultrasound for detecting urinary tract calculi ranges
from 3495% [77,78].

Safety of ultrasound

The safety of ultrasound examination applied to pregnant women has been subject to much
research and debate. Ultrasound can produce biologic effect by multiple mechanisms, including
thermal (heating), cavitation, and microstreaming. It is well known that high levels of ultrasound are
capable of producing biological damage. The safety of ultrasound use has been indirectly correlated
to the increase in temperature it produces. Studies have shown that temperature rises up to 1  C are
considered to be safe for the developing embryo [79], and to give a rough guide of the possible
 J. Knez et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 28 (2014) 621636 633

temperature increase induced by ultrasound, thermal index (TI) has been introduced. A TI of 1.0
means that, under worst-case conditions, the temperature might increase by 1  C. The TI depends on
the organ being scanned. This has led to the introduction of three different TI indices: TIB (thermal
index for bone), TIS (thermal index for soft tissue) and TIC (thermal index for cranial bone). Corre-
spondingly, the MI (mechanical index) is the counterpart for the mechanical effects of ultrasound. As
in the case of TI, MI value of less than 1.0 is considered to be safe for the developing embryo. These
numbers are displayed in real-time, usually on the screen of modern ultrasound machines. The
examiner, however, should always adhere to as-low-as-reasonable achievable (ALARA) rule. This
means that the examiner should always use the lowest output settings in order to obtain the desired
information, and should not unnecessarily prolong the examination time. The British Medical Ul-
trasound Society has also issued recommendations about the duration of ultrasound examination at
different TI and MI [80].
The basic technique used in ultrasound examination, the B-mode ultrasound, is generally consid-
ered to be safe because of its low output intensity with TI and MI well below 1.0, unless associated with
Doppler technology. Three-dimensional ultrasound imaging does not entail higher risks than two-
dimensional B-mode ultrasound, as three-dimensional ultrasound only digitally combines multiple
two-dimensional images to produce a three-dimensional representation of the captured eld. When
using transvaginal or transabdominal B-mode ultrasound, a safety limit of 30 minutes should still be
followed, owing to concerns of disturbed neuron migration of animal fetuses with prolonged exposure
to continuous ultrasound [81]. The highest levels of TI are usually recorded when the machine is used
in pulsed Doppler mode. According to World Federation of Ultrasound in Medicine and Biology,
Doppler ultrasound should not be used routinely in early pregnancy assessment [82]. In a case when
Doppler mode is indicated, TI should be less or equal to 1.0, and the exposure time should be kept as
short as possible. Because of these concerns, only M-mode ultrasound should be used to document and
measure the fetal heart rate in routine early pregnancy assessment.

Conclusion

In most women with suspected early pregnancy failure, a single ultrasound examination will
identify the location of pregnancy and provide information about its potential to progress nor-
mally. Most ectopic pregnancies can be identied on ultrasound scan, which facilitates the use of
conservative management strategies, and surgery is mainly used to treat rather than to diagnose
ectopic pregnancies. Pain and bleeding in early pregnancy may also be caused by the concomitant
pelvic abnormalities, and all possible causes of womens symptoms during ultrasound examination
should be considered. There are no concerns about possible harmful effects of ultrasound on the
developing embryo as long as the basic safety principles are followed and this technique is likely to
remain a pivotal test for the diagnosis of early pregnancy complications for a foreseeable future.

Practice points

 Transvaginal ultrasound is the method of choice for the diagnosis of early pregnancy
complications.
 There is no evidence that ultrasound examination in early pregnancy is harmful provided that
low-intensity imaging is used, and the examiners adhere to as-low-as-reasonable achievable
(ALARA) rule.
 Ultrasound diagnosis of early embryonic demise is based on negative findings, and the
diagnosis should be checked by another independent operator or on a follow up visit in order
to minimise the risk of diagnostic errors.
 Most tubal ectopic pregnancies should be detected on a single ultrasound examination.
 Non-tubal ectopic pregnancies are rare, are more difficult to diagnose, and are associated
with higher maternal morbidity and mortality.
634 J. Knez et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 28 (2014) 621636

Research agenda

 To devise more accurate criteria for the diagnosis of incomplete miscarriage.

 To develop better algorithms for non-interventional management of miscarriages and
ectopic pregnancies.
 To improve diagnostic criteria for detecting partial hydatidiform mole.
 To refine diagnostic criteria and improve management of acute ovarian torsion.

Conict of interest statement

None declared.

References

*[1] National Institute for Health and Care Excellence. Ectopic pregnancy and miscarriage: diagnosis and initial management
in early pregnancy of ectopic pregnancy and miscarriage. London: National Institute for Health and Care Excellence; 2012.
[2] Everett C. Incidence and outcome of bleeding before the 20th week of pregnancy: prospective study from general
practice. BMJ 1997;315:324.
[3] Rosevear S. Bleeding in early pregnancy. In: James DK, steer CP, Gonick B, editors. High risk pregnancy management
options. 2nd ed. London: Saunders WB; 2003.
[4] Shillito J, Walker JJ. Early pregnancy assessment units. Br J Hosp Med 1997;58:5059.
[5] Cacciatore B, Stenman UH, Ylstalo P. Comparison of abdominal and vaginal sonography in suspected ectopic pregnancy.
Obstet Gynecol 1989;73:7704.
[6] Bigrigg MA, Read MD. Management of women referred to early pregnancy assessment unit: care and cost effectiveness.
BMJ 1991;302:5779.
*[7] Jurkovic D, Mavrelos D. Catch me if you can: ultrasound diagnosis of ectopic pregnancy. Ultrasound Obstet Gynecol 2007;
30:17.
[8] Naftalin J, Jurkovic D. The endometrial-myometrial junction: a fresh look at a busy crossing. Ultrasound Obstet Gynecol
2009;34:111.
[9] Day A, Jurkovic D. The role of ultrasound in early pregnancy after assisted conception. In: Jauniaux E, Rizk B, editors.
Pregnancy after assisted reproductive technology. Cambridge: Cambridge University Press; 2012.
[10] Mehta T, Levine D, McArdle C. Lack of sensitivity of endometrial thickness in predicting the presence of an ectopic
pregnancy. J Ultrasound Med 1999;18:11722.
[11] Naftalin J, Hoo W, Pateman K, et al. How common is adenomyosis? A prospective study of prevalence using transvaginal
ultrasound in a gynaecology clinic. Hum Reprod 2012;27:34329.
[12] Lindsay D, Lovett I, Lyons E, et al. Yolk sac diameter and shape at endovaginal US: predictors of pregnancy outcome in the
rst trimester. Radiology 1992;183:1158.
[13] Rowling S, Langer J, Coleman B, et al. Sonography during early pregnancy: dependence of threshold and discriminatory
values on transvaginal transducer frequency. Am J Roentgenol 1999;172:9838.
[14] Robinson H, Fleming J. A critical evaluation of sonar crown-rump length measurement. Br J Obstet Gynecol 1975;82:70.
[15] ISOUG Bioeffects and Safety Committee. Safety Statement, 2000 reconrmed 2003. Ultrasound Obstet Gynecol 2003;21:
100.
[16] Doubilet P, Benson C. Embryonic heart rate in the early rst trimester: what rate is normal? J Ultrasound Med 1995;14:
4314.
[17] Schats R, Jansen CA, Wladimiroff JW. Embryonic heart activity: appearance and development in early human pregnancy.
Br J Obstet Gynecol 1990;97:98994.
[18] Prendeville W. Epidemiology of miscarriage. In: Grudzinskas JG, OBrien PMS, editors. Problems in early pregnancy:
advances in diagnosis and management. RCOG Press; 1997. pp. 115.
[19] Tummers P, De Sutter P, Dhont M. Risk of spontaneous abortion in singleton and twin pregnancies after IVF/ICSI. Hum
Reprod 2003;18:17203.
*[20] Jeve Y, Rana R, Bhide A, et al. Accuracy of rst-trimester ultrasound in the diagnosis of early embryonic demise: a sys-
tematic review. Ultrasound Obstet Gynecol 2011;38:48996.
[21] Reljic M. The signicance of crownrump length measurement for predicting adverse pregnancy outcome of threatened
abortion. Ultrasound Obstet Gynecol 2001;17:5102.
[22] Dickey R, Olar T, Taylor S, et al. Relationship of small gestational sac-crown-rump length differences to abortion and
abortus karyotypes. Obstet Gynecol 1992;79:5547.
[23] Nazari A, Check J, Epstein R, et al. Relationship of small-for-dates sac size to crown-rump length and spontaneous
abortion in patients with a known date of ovulation. Obstet Gynecol 1991;78:36973.
[24] Laboda L, Estroff J, Benacerraf B. First trimester bradycardia. A sign of impending fetal loss. J Ultrasound Med 1989;8:5613.
[25] Bora S, Bourne T, Bottomley C, et al. Twin growth discrepancy in early pregnancy. Ultrasound Obstet Gynecol 2009;34:
3842.
[26] Landy H, Weiner S, Corson S, et al. The vanishing twin: ultrasonographic assessment of fetal disappearance in the rst
trimester. Am J Obstet Gynecol 1986;155:149.
 J. Knez et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 28 (2014) 621636 635

[27] Jauniaux E, Elkazen N, Leroy F, et al. Clinical and morphological aspects of the vanishing twin phenomenon. Obstet
Gynecol 1988;72:577.
[28] Jurkovic D. Three-dimensional ultrasound in gynecology: a critical evaluation. Ultrasound Obstet Gynecol 2002;19:10917.
[29] Kurtz AB, Shlansky-Goldberg RD, Choi HY. Detection of retained products of conception following spontaneous abortion
in the rst trimester. J Ultrasound Med 1991;10:38795.
[30] Nielsen S, Hahlin M. Expectant management of rst-trimester spontaneous abortion. Lancet 1995;345:846.
[31] Sawyer E, Ofuasia E, Oli-Yebovi D, et al. The value of measuring endometrial thickness and volume on transvaginal
ultrasound scan for the diagnosis of incomplete miscarriage. Ultrasound Obstet Gynecol 2007;23:2059.
[32] Luise C, Jermy K, May C, et al. Expectant management of incomplete, spontaneous rst trimester miscarriage: outcome
according to initial ultrasound criteria and value of follow-up visits. Ultrasound Obstet Gynecol 2002;19:5802.
[33] Kirk E, Condous G, Van Calster B, et al. Rationalizing the follow-up of pregnancies of unknown location. Hum Reprod
2007;22:174450.
[34] Day A, Sawyer E, Mavrelos D, et al. Use of serum progesterone measurements to reduce need for follow-up in women
with pregnancies of unknown location. Ultrasound Obstet Gynecol 2009;33:70410.
[35] Fowler DJ, Lindsay I, Seckl MJ, et al. Routine pre-evacuation ultrasound diagnosis of hydatidiform mole: experience of
more than 1000 cases from a regional referral center. Ultrasound Obstet Gynecol 2006;27:5660.
[36] Kirk E, Papageorghiou AT, Condous G, et al. The accuracy of rst trimester ultrasound in the diagnosis of hydatidiform
mole. Ultrasound Obstet Gynecol 2007;29:705.
[37] Sebire NJ. The diagnosis of gestational trophoblastic disease in early pregnancy: implications for screening, counseling
and management. Ultrasound Obstet Gynecol 2005;25:4214.
[38] Fernandez H, Gervaise A. Ectopic pregnancies after infertility treatment: modern diagnosis and therapeutic strategy. Hum
Reprod Update 2004;10:50313.
*[39] Jurkovic D, Wilkinson H. Diagnosis and management of ectopic pregnancy. BMJ 2011;10:342.
[40] Bouyer J, Coste J, Fernandez H, et al. Sites of ectopic pregnancy: a 10 year population-based study of 1800 cases. Hum
Reprod 2002;17:322430.
[41] Walker J. Ectopic pregnancy. Clin Obstet Gynecol 2007;50:8999.
*[42] Varma R, Gupta J. Tubal ectopic pregnancy. Clin Evid (Online) 2009;4:116.
[43] Weckstein LN, Boucher AR, Tucker H, et al. Accurate diagnosis of early ectopic pregnancy. Obstet Gynecol 1985;65:3937.
[44] Dart RG, Kaplan B, Varaklis K. Predictive value of history and physical examination in patients with suspected ectopic
pregnancy. Ann Emerg Med 1999;33:28390.
*[45] Crochet JR, Bastian LA, Chireau MV. Does this women have an ectopic pregnancy? The rational clinical examination
systematic review. JAMA 2013;309:17229.
[46] Kobayashi M, Hellman LM, Fillisti LP. Ultrasound: an aid in the diagnosis of ectopic pregnancy. Am J Obstet Gynecol 1969;
103:113140.
[47] Condous G, Okaro E, Khalid A, et al. The accuracy of transvaginal sonography for the diagnosis of ectopic pregnancy prior
to surgery. Hum Reprod 2005;20:14049.
[48] Valenzano M, Anserini P, Remorgida V, et al. Transabdominal and transvaginal ultrasonographic diagnosis of ectopic
pregnancy: clinical implications. Gynecol Obstet Invest 1991;31:811.
*[49] Kirk E, Papageorghiou AT, Condous G, et al. The diagnostic effectiveness of an initial transvaginal scan in detecting ectopic
pregnancy. Hum Reprod 2007;22:28248.
*[50] Brown DL, Doubilet PM. Transvaginal sonography for diagnosing ectopic pregnancy. J Ultrasound Med 1994;13:25966.
[51] Kirk E, Daemen A, Papageorghiou AT, et al. Why are some ectopic pregnancies characterized as pregnancies of unknown
location at the initial transvaginal ultrasound examination? Acta Obstet Gynecol Scand 2008;87:11504.
[52] Timor-Tritsch I, Yeh M, Peisner D, et al. The use of transvaginal ultrasonography in the diagnosis of ectopic pregnancy. Am
J Obstet Gynecol 1989;161:15761.
[53] Jurkovic D, Bourne T, Jauniaux E, et al. Transvaginal color Doppler study of blood ow in ectopic pregnancies. Fertil Steril
1992;57:6873.
*[54] Barnhart K, van Mello NM, Bourne T, et al. Pregnancy of unknown location: a consensus statement of nomenclature,
denitions, and outcome. Fertil Steril 2011;95:85766.
[55] Nurmohamed L, Moretti ME, Schechter T, et al. Outcome following high-dose methotrexate in pregnancies misdiagnosed
as ectopic. Am J Obstet Gynecol 2011;205:533.e13.
[56] Benson CB, Doubilet PM, Peters HE, et al. Intrauterine uid with ectopic pregnancy: a reappraisal. J Ultrasound Med 2013;
32:38993.
[57] Fleischer AC, Pennell RG, McKee MS, et al. Ectopic pregnancy: features at transvaginal sonography. Radiology 1990;174:
3758.
[58] Nyberg DA, Hughes MP, Mack LA, et al. Extra-uterine ndings of ectopic pregnancy at transvaginal US: importance of
echogenic uid. Radiology 1991;178:8236.
[59] Wang R, Reynolds TA, West HH, et al. Use of a b-hCG discriminatory zone with bedside pelvic ultrasonography. Ann
Emerg Med 2011;58:1220.
*[60] Van Mello N, Mol F, Opmeer BC, et al. Diagnostic value of serum hCG on the outcome of pregnancy of unknown location:
a systematic review and meta-analysis. Hum Reprod Update 2012;18:60317.
[61] The Condential Enquiry into Maternal and Child Health (CEMACH). Saving mothers lives: reviewing maternal deaths to
make motherhood safer 20062008. The Eighth Report of the Condential Enquiries into Maternal Deaths in the United
Kingdom. BJOG 2011;118(Suppl. 1):1203.
[62] Lawrence A, Jurkovic D. Three-dimensional ultrasound diagnosis of interstitial pregnancy. Ultrasound Obstet Gynecol
1999;14:2923.
[63] Mavrelos D, Sawyer E, Helmy S, et al. Ultrasound diagnosis of ectopic pregnancy in the non-communicating horn of a
unicornuate uterus (cornual pregnancy). Ultrasound Obstet Gynecol 2007;30:76570.
[64] Jurkovic D, Hacket E, Campbell S. Diagnosis and treatment of early cervical pregnancy: a review and a report of two cases
treated conservatively. Ultrasound Obstet Gynecol 1996;8:37380.
636 J. Knez et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 28 (2014) 621636

[65] Raziel A, Schachter M, Mordechai E, et al. Ovarian pregnancy-a 12-year experience of 19 cases in one institution. Eur J
Obstet Gynecol Reprod Biol 2004;114:926.
[66] Bajekal N, Li TC. Fibroids, infertility and pregnancy wastage. Hum Reprod Update 2000;6:61420.
[67] Yazdany T, Bhatia NN, Nguyen JN. Urinary retention and voiding dysfunction in women with uterine leiomyoma: a case
series. J Reprod Med 2012;57:3849.
[68] Fleischer A, Shah D, Entman S. Sonographic evaluation of maternal disorders during pregnancy. Radiol Clin North Am
1990;28:518.
[69] Kemmann E, Ghazi D, Corsan G. Adnexal torsion in menotropin-induced pregnancies. Obstet Gynecol 1990;76:4036.
[70] McWilliams G, Hill M, Dietrich C. Gynecologic emergencies. Surg Clin North Am 2008;111:13139.
[71] Mashiach S, Bider D, Moran O, et al. Adnexal torsion of hyperstimulated ovaries in pregnancies after gonadotropin
therapy. Fertil Steril 1999;5:7680.
[72] Valsky DV, Esh-Broder E, Cohen SM, et al. Added value of the gray-scale whirlpool sign in the diagnosis of adnexal torsion.
Ultrasound Obstet Gynecol 2010;36:6304.
[73] Tamir IL, Bongard FS, Klein SR. Acute appendicitis in the pregnant patient. Am J Surg 1990;160:5716.
[74] He P, Viktrup L. The diagnosis of appendicitis during pregnancy and maternal and fetal outcome after appendectomy. Int
J Gynaecol Obstet 1999;65:12935.
[75] Lim HK, Bae SH, Seo GS. Diagnosis of acute appendicitis in pregnant women: value of sonography. AJR Am J Roentgenol
1992;159:53942.
[76] Israel GM, Malguria N, McCarthy S, et al. MRI vs ultrasound for suspected appendicitis during pregnancy. J Magn Reson
Imaging 2008;28:42833.
[77] Parulkar BG, Hopkins TB, Wollin MR, et al. Renal colic during pregnancy: a case for conservative treatment. J Urol 1998;
159:3658.
[78] Stothers L, Lee LM. Renal colic in pregnancy. J Urol 1992;148:13837.
[79] National Council on radiation protection and measurements (NCRP). Exposure criteria for medical diagnostic ultrasound:
I. Criteria based on thermal mechanisms. (Report No. 113). Bethesda, MD: National Council on Radiation Protection and
Measurements; 1992.
[80] Safety Group of the British Medical Ultrasound Society. Guidelines for the safe use of diagnostic ultrasound equipment.
Ultrasound 2010;18:529.
[81] Ang Jr ESB, Gluncic V, Duque A, et al. Prenatal exposure to ultrasound waves impacts neuronal migration in mice. Proc
Natl Acad Sci USA 2006;103:1290310.
[82] WFUMB. Safe use of Doppler ultrasound during 11 to 14 weeks scans (or earlier in pregnancy). Ultrasound Med Biol 2013;
39:373.