Introduction

Acute gynaecological symptoms such as pelvic pain and bleeding occur in pregnant women who have miscarriages, ectopic pregnancies and pregnancies of unknown location, or in women who are not pregnant with ovarian cyst accidents or acute pelvic inflammatory disease. The investigation of the acute gynaecological patient is moving away from the accident and emergency department and operating theatre, and into dedicated emergency gynaecology units (EGUs) located in an outpatient setting (Jones and Pearce 2009). EGUs offer an efficient way of organizing multidisciplinary services for women presenting with acute gynaecological symptoms. These changes have been driven by the demands of the patient and clinician to provide a rapid, accurate diagnosis, with the minimum of investigations and invasive procedures. They are underpinned by national standards and guidelines (Department of Health 2003, Royal College of Obstetricians and Gynaecologists 2006). Furthermore, changing from traditional care pathways to more cost-effective, patient-centred approaches to medical practice lies at the heart of modern health service management (Department of Health 2000, Jones 2008).

In the UK, the majority of gynaecological ultrasound scanning has traditionally been undertaken by radiographers and radiologists in a separate department on a separate day. Transvaginal ultrasonography (TVUS) is a pivotal investigation for the assessment of acute gynaecological patients. Therefore, in order to deliver modern acute gynaecological services, gynaecologists will have to learn these new skills (Jones 2005). TVUS probes, providing high-resolution images of the pelvic organs, have an established role in the characterization of adnexal masses, providing reliable and reproducible information regarding cyst type and probability of malignancy (Granberg et al 1989, 1990, Tailor et al 1997, Timmerman et al 1999a), particularly when they are used in combination with tumour markers (Moore et al 2008). They have superiority over transabdominal probes because of the higher resolution of pelvic anatomy. There is also no need for a full bladder, improving patient acceptability.

TVUS is a pivotal investigation in the delivery of an ‘Ambulatory Gynaecology Service’ (Jones 2008). This is a more comprehensive term which would include the assessment of patients with non-acute gynaecological symptoms such as menstrual disorders, postmenopausal bleeding and chronic pelvic pain. The discussion of these conditions falls outside the scope of this chapter.

The accurate identification of pathology on ultrasound will enable the clinician to plan the patient’s management effectively. The ultrasound diagnosis may determine if the patient is suitable for minimal access surgery or they may need a laparotomy and a multidisciplinary oncology team. In this way, preoperative ultrasound assessment of the patient will improve patient counselling and satisfaction with treatment.

This chapter deals with assessment of the patient presenting with acute symptoms, in whom gynaecological pathology is suspected. By adopting a problem-orientated approach, the aim is to provide a series of reproducible pathways allowing effective investigation of these patients. TVUS is central to all these diagnostic pathways but it should be regarded as an extension of the clinical examination; a normal scan will not exclude all underlying gynaecological conditions. The first part of this chapter will explore the place of TVUS in the differential diagnosis of acute pain and bleeding in early pregnancy. The second part of the chapter will give an overview of the role of pelvic ultrasound in the evaluation of acute pelvic pain and vaginal bleeding in women with a negative pregnancy test.

Resources for Services

The Emergency Gynaecological Unit

The provision of an EGU is now recognized as a gold standard service (Royal College of Obstetricians and Gynaecologists 2006). It provides an easily accessible outpatient area with facilities for ultrasound-based assessment of pregnancy duration, viability and location, This allows the clinician to provide informed management and counselling of the patient in a setting which maximizes her privacy and comfort. The EGU should have a dedicated area for the triage, assessment and initial management of patients presenting with acute gynaecological disorders, including early pregnancy complications. There should be a separate area where the patient can change, a designated waiting area with toilet facility, and bed/trolley spaces to accommodate women who need to recline. An initial assessment of the clinical condition can be made and recorded on a preformed history sheet. The cost benefits of the EPU are well established (Bigrigg and Read 1991) as admission can be avoided in approximately 40% of patients, with a further 20% requiring a shorter stay. Ideally, there should be a dedicated unit for the assessment and investigation of women with suspected complications of early pregnancy. This will be centred around an ultrasound scanning room, run by dedicated ultrasound practitioners. There should be a private area in which the patient can change and also a separate counselling room with access to a counselling service and outside line telephone.

Referral to the EPU may be by prior arrangement with the general practitioner or other consultant, or, ideally, a ‘walk-in’ self-referral system can operate. The latter will provide the patient with the means to contact a unit directly when problems arise, although it will make clinics busy. Again, the history can be collected on a proforma. This will highlight relevant gynaecological and obstetric history for the clinician, including factors predisposing to ectopic pregnancy or women in whom recurrent miscarriages have been identified, and will also facilitate audit and research. A number of computer databases now exist which aid data collection and facilitate reporting. This means that the patient can leave the clinic with a detailed report and follow-up plan, and a report can be distributed immediately to the referring clinician.

Many of the women who attend as gynaecological emergencies present with pelvic pain with or without vaginal bleeding. The differential diagnosis should be whether or not the symptoms have a gynaecological cause or some other underlying pathology, such as a gastrointestinal (e.g. constipation, appendicitis, diverticulities) or urological (e.g. urinary tract infection, stone) cause. If the symptoms are thought to be gynaecological in nature, the next decision is whether or not the woman is pregnant. If she is pregnant, pregnancy location and viability must be established. If the symptoms are not pregnancy related, are they chronic (>6 months’ duration) or acute? If they are acute, does she have an ovarian cyst accident (e.g. haemorrhage, rupture, torsion) or acute pelvic inflammatory disease?

Availability of rapid serum β-human chorionic gonadotrophin (β-hCG) assays will vary within units, but the judicious use and follow-up of serum changes in β-hCG is essential for a diagnosis where the pregnancy location is not readily identified. This may be achieved in a number of ways, but typically will be dependent upon a dedicated staff member to record and interpret results either via a computer database or log book. The patient is either contacted with further follow-up/intervention plans or the patient contacts the unit herself for these results.

Fortunately, the majority of women with pain or bleeding in early pregnancy will present subacutely; in all cases, however, an initial assessment of haemodynamic compromise must be made. Blood pressure, pulse, temperature and pulse oximetry should be recorded. Facilities for obtaining venous access and commencing intravenous volume replacement should be available.

Ultrasound Overview

TVUS should be regarded as an extension of a bimanual examination rather than a replacement. A recent prospective comparative trial of endovaginal sonographic bimanual examination versus traditional digital bimanual examination in non-pregnant women with lower abdominal pain has been reported (Tayal et al 2008). The study clearly demonstrated that vaginal examinations combined with TVUS improved confidence in key findings, such as ovarian and uterine size or position irrespective of the patient’s body mass index. The advantages of TVUS over the transabdominal approach are well documented (Gonzalez et al 1988, Mendelson et al 1988). The placement of the ultrasound probe closer to the pelvic organs means that higher frequency ultrasound transducers (6–7.5 MHz) can be used, which produce high-resolution images. TVUS assessment of ovarian volume and morphology correlates closely with subsequent operative findings at laparotomy (Rodriguez et al 1988).

No single ultrasound measurement of the different anatomical features in the first trimester has been shown to have a high predictive value for determining early pregnancy outcome, and Doppler studies are not helpful either (Jauniaux et al 2005). Despite this, high-resolution TVUS has transformed understanding of the pathophysiology and management of early pregnancy failure. The ultrasound findings in both a normal pregnancy and a failing pregnancy are well described (Dogra et al 2005). The intrauterine gestation sac can be visualized from approximately 4 weeks after the last menstrual period using a transvaginal transducer, which is approximately 2 weeks earlier than if using the transabdominal approach (Fossam et al 1988). This is dependent upon a regular 28-day cycle, and hence a more accurate approach to confirming pregnancy viability or failure is dependent upon either changes with time or the presence or absence of fetal cardiac activity. The diameter of the gestational sac increases at approximately 1 mm/day in early pregnancy (Nyberg et al 1985), and can be differentiated from the ‘pseudosac’ of ectopic pregnancy by its thick, echogenic rind surrounding the echo-lucent central chorionic sac and eccentric location to the endometrial midline. The presence of a normal intrauterine gestation sac is associated with β-hCG levels of >1000 IU. Detection at levels greater than this is dependent upon a number of factors, including type of probe and ultrasound machine used, the presence of leiomyomas, operator variations and multiple gestations (Nyberg et al 1985, Bernaschek et al 1988). The absence of a gestation sac should prompt the operator to look for an extrauterine gestation (including a close examination of the cornua and cervix as well as the adnexa). Visualization of the yolk sac is regarded as definitive evidence of an intrauterine pregnancy. The yolk sac can be visualized from 5 weeks’ gestation (Figure 6.1), and the early embryonic pole from approximately 6 weeks. First recognized as a thickening along the yolk sac, it is linear to begin with, subsequently becoming curved in nature. The embryonic growth rate is 1 mm/day. Cardiac activity starts at approximately 5 weeks after the last menstrual period. Cardiac activity not detected in embryos of more than 4 mm is associated with embryonic demise (Brown et al 1990, Levi et al 1990, Goldstein 1992). However, a repeat scan to confirm diagnosis is always indicated. Embryonic bradycardia can be associated with poor outcome (Doubilet et al 1999), and a follow-up scan is warranted to confirm viability.

Figure 6.1 Intrauterine gestation and yolk sac: 5 weeks’ gestation.

TVUS has an established role in the evaluation of adnexal masses. It provides an accurate assessment of ovarian morphology (Granberg et al 1989, 1990, Timmerman et al 1999b) and is a significant contributor to mathematical models being developed to assess ovarian tumours (Tailor et al 1997, Timmerman et al 1999a,c). There are several types of ovarian cyst that can be assessed using the recognition of characteristic morphological patterns (Jermy et al 2001). Endometriomas and benign cystic teratomas are two examples, accounting for over two-thirds of persistent adnexal masses in premenopausal women (Koonings et al 1989). These lesions can be particularly difficult to score using morphological scoring systems, and as angiogenesis is ubiquitous throughout the ovarian cycle, colour Doppler is of limited value (Alcazar et al 1997).

Clinical Conditions

Miscarriage

Once the diagnosis of intrauterine pregnancy has been established, the potential viability of the pregnancy will need to be addressed. The clinical value of the normal developmental timespan of the early embryonic and extraembryonic structures is its application in the diagnosis of pregnancy failure. What becomes more relevant is not the earliest point at which a structure can be seen (threshold), but the point at which a structure is always seen in a normally developing intrauterine pregnancy (discriminatory level), so its absence is diagnostic of pregnancy failure. Once an ectopic pregnancy has been excluded, all intrauterine pregnancies should be given the benefit of the doubt and serial scans should be performed to confirm a diagnosis. However, cardiac activity should always be visualized in an embryo measuring 6 mm or more (Figure 6.2).

Figure 6.2 Assessment of viability in early pregnancy. β-hCG, β-human chorionic gonadotrophin; FBC, full blood count; TVS, transvaginal sonography; CRL, crown–rump length; RPOC, retained products of conception; G&S, groups and save; FH, fetal heart.

Knowledge of early pregnancy anatomical landmarks allows one to follow fetal development within a normal pregnancy, and to establish safe guidelines in the diagnosis of a pathological pregnancy.

Terminology

• Missed miscarriage (early fetal demise): crown–rump length of at least 6 mm with no cardiac activity or no change in size on weekly serial scanning.

• Blighted ovum (early embryonic demise): mean diameter of gestational sac of at least 20 mm with no embryonic/extraembryonic structures present.

• Incomplete miscarriage: disrupted endometrial echo measuring >15 mm in the anteroposterior plane (Figure 6.3).

• Complete miscarriage: endometrial thickness of <15 mm measured in the anteroposterior plane associated with the cessation of heavy bleeding and pain.

Figure 6.3 Sagittal view of the uterus demonstrating retained products of conception.

Trouble shooting

• In practice, the presence of an intrauterine gestation usually excludes an ectopic pregnancy.

• Once ectopic pregnancy is excluded, early pregnancies should be given the benefit of the doubt; serial scans are of more value than serial β-hCG values.

• Cardiac activity should be seen with a crown–rump length of 6 mm or greater.

• A fetal pole/yolk sac should be seen with a gestation sac with a mean diameter of 20 mm.

Ectopic pregnancy

The exclusion of an ectopic gestation remains a primary goal for the clinician assessing a patient presenting with pain and/or bleeding in early pregnancy. Traditionally, the presence of an empty uterus with a positive urinary pregnancy test has meant hospital admission with either laparoscopy or follow-up scans until a diagnosis is made. However, the combined use of high-resolution vaginal probes and serum β-hCG assays has meant that patients with symptoms and clinical findings suggestive of an ectopic gestation will fall into one of three categories, as follows:

• an intrauterine pregnancy will be diagnosed on scan: to all extents and purposes, this will exclude an extrauterine location. The frequency of heterotopic pregnancy in spontaneous conceptions is estimated at between 1 : 10,000 and 1 : 50,000, but is as high as 1 : 100 in assisted conceptions (Ludwig et al 1999), and should be suspected if there are persistent symptoms;

• an extrauterine gestation or highly predictive features (e.g. haemoperitoneum) will be visualized by ultrasound (Figures 6.4–6.6); or

• no evidence of intra- or extrauterine gestation will be found on scan alone; a pregnancy of unknown location.

Figure 6.4 Sagittal section of the left adnexa demonstrating a tubal ectopic pregnancy. The crown–rump length is marked (12.1 mm).

Figure 6.5 Cornual ectopic pregnancy. The gestation sac is seen separately from the endometrial echo.

Figure 6.6 Sagittal view of the cervix containing a gestational sac, fetal pole and yolk sac. The proximal portion of the endometrial cavity contains free fluid and can be seen on the left side of the image.

Further management of the patient will be dependent upon her clinical status. Within the authors’ unit, there has been a significant reduction in patients presenting with acute haemodynamic compromise due to a ruptured ectopic pregnancy since the EPU was established. This has meant that more conservative methods of treatment can be used, whether surgical (laparoscopic salpingectomy, salpingotomy), medical (methotrexate) or expectant (Figure 6.7).

Figure 6.7 Assessment of an extrauterine pregnancy. β-hCG, β-human chorionic gonadotrophin; FBC, full blood count; TVS, transvaginal sonography; G&S, group and save.

All non-surgical methods of treating ectopic gestations will need intensive follow-up to ensure resolution of symptoms and β-hCG values. Patient compliance is central to these patients being treated on an outpatient basis, along with a dedicated EPU service (see Chapter 25, Ectopic pregnancy, for more information).

Pregnancy of unknown location

A proportion of women presenting with complications of early pregnancy will have no sonographic features of extra- or intrauterine pregnancy. This group will include patients who have a very early, continuing, intrauterine pregnancy, those who have an ectopic gestation and those who have a failing pregnancy, whether intra- or extrauterine. Rarely, a false-positive result may be due to a placental site tumour, for example of the ovary. Figure 6.8 demonstrates a suggested follow-up regime for women who fall into this category. Again, a coordinated EPU and rapid β-hCG assay service are essential.

Figure 6.8 Assessment of pregnancy of unknown location. β-hCG, β-human chorionic gonadotrophin; FBC, full blood count; TVS, transvaginal sonography; G&S, group and save.

The premenopausal patient

The importance of a complete clinical history and examination cannot be overemphasized, especially with the use of pelvic ultrasound. The premenopausal ovary and endometrium, as visualized with the transvaginal probe, are dynamic structures, exhibiting cyclical changes in morphology and volume throughout the cycle. A knowledge of the normal variations that can be exhibited is important, as the most common ovarian pathology is the functional cyst.

The presence of adnexal pathology in the patient presenting with pelvic pain may be a coincidence, and gentle use of the transvaginal probe to map the pain within the pelvis will help to indicate the structures giving rise to the pain. The overwhelming advantages of the transvaginal approach in the assessment of pelvic pain are not only the excellent diagnostic capabilities, but also that, in the presence of an acute abdomen, especially in women with pelvic inflammatory disease, this route is tolerated much better than a transabdominal approach with a full bladder.

Pelvic pain in association with a pelvic mass

In those patients presenting with pelvic pain in whom there is evidence of a pelvic mass on ultrasound, an assessment needs to be made not only of pain severity and type, but also the nature of the mass. TVUS has a proven track record in experienced hands in the differentiation of uterine myomas from adnexal masses, and also in the characterization of adnexal masses.

The prevalence of adnexal pathology among premenopausal women is high and the overwhelming majority of these lesions will be benign in nature. A large proportion of benign ovarian cysts will be functional and, if symptoms settle, may be managed expectantly.

A careful history and a clear knowledge of the day of the menstrual cycle will prompt the sonographer to the most likely cause of the pain. For example, acute-onset, midcycle pain may be indicative of a follicular or corpus luteal cyst accident. Haemorrhage into a corpus luteal cyst has characteristic sonographic findings (Figure 6.9). The condition tends to be self-limiting and often responds to non-steroidal anti-inflammatory analgesia. Surgery should be avoided if possible.

Figure 6.9 Haemorrhagic corpus luteal cyst. Fine strands of fibrin are seen within the cyst contents.

Although usually self-limiting, the cyclical occurrence of haemorrhagic corpus luteal cyst can be a source of persistent morbidity; if there are no contraindications, use of the combined oral contraceptive to suppress ovulation is beneficial. This is also indicated for women who have clotting factor deficiencies, such as von Willebrand’s disease. They may present with an acute abdomen, secondary to a haemoperitoneum, as a result of ovulation or corpus luteal haemorrhage or rupture.

Women with a history suggestive of endometriosis presenting with acute pain may have sonographic evidence of an endometrioma (Figure 6.10). These rarely undergo torsion, as they are often fixed within the pelvis, but may undergo rupture or acute haemorrrhage within the cyst. Rarely, they can become infected. Recent cyst rupture may be suggested by the presence of resolving clinical symptoms, with free fluid present in the pouch of Douglas, often with a collapsing irregular cyst wall.

Figure 6.10 Large ovarian endometrioma: characteristic ‘ground-glass’ appearance of the cyst contents.

A suggested follow-up regime for women diagnosed with a pelvic mass is shown in Figure 6.11. Intervention will be dictated by the resolution, or not, of the patient’s symptoms. If the symptoms resolve and there are no sonographic features of malignancy on the ultrasound, a repeat scan at 6 weeks should be performed to confirm cyst resolution.

Figure 6.11 Assessment of pelvic mass. FBC, full blood count.

Ovarian torsion is unusual with adnexal masses <5 cm (Nicholas and Julian 1985). However, there are no pathognomonic features specific to adnexal torsion, and a high degree of clinical suspicion is essential. The clinical history is of acute-onset, constant pain that does not respond to analgesia, often with nausea and vomiting and systemic upset. Of the persistent adnexal masses in premenopausal women, benign cystic teratomas (Figure 6.12) are more likely to undergo torsion than endometriomas. The central feature of ovarian torsion is the cessation of vascular supply. Colour Doppler has therefore been used to interrogate the adnexal mass suspected of undergoing torsion. It is likely, however, that even if flow can be visualized within the mass, despite clinical symptoms and signs of ovarian torsion, ovarian blood flow may still be compromised, as demonstrated by surgically proven ovarian torsion despite the detection of blood flow within the mass (Rasado et al 1992).

Figure 6.12 Benign cystic teratoma: acoustic shadowing is demonstrated.

Uterine leiomyomas

These may undergo torsion if pedunculated in nature, may prolapse through the cervix or may undergo degeneration, especially during pregnancy. In fibroid degeneration, the patient is often systemically unwell, with pyrexia, leucocytosis and generalized abdominal tenderness.

The ultrasound characteristics of leiomyomas are well documented and cystic areas can be visualized within the fibroid if it is degenerating. Advances in the management of uterine myomas have resulted in a need to provide accurate pretreatment information concerning their size, quantity and location. This is especially true with the increasing use of minimally invasive techniques of fibroid resection. The ultrasound appearances of myomas are varied. Before the menopause, they tend to be a well-defined heterogeneous or hypoechoic uterine mass. TVUS is used in conjunction with abdominal scanning to ensure that pedunculated subserosal fibroids are not missed.

Tubal pathology

Acute pelvic inflammatory disease, with hydrosalpinx, pyosalpinx or frank tubo-ovarian abscesses, tends to present with systemic upset, leucocytosis, unilateral progressing to bilateral pelvic pain, menstrual disturbances and vaginal discharge. The differentials will therefore include adnexal torsion, acute fibroid degeneration, urinary tract infection and appendicitis.

Acute inflammatory processes within the fallopian tubes tend to produce thick-walled, cystic structures, tender to the touch of the probe (Figure 6.13). However, a chronic hydrosalpinx will have the appearance of a thin-walled structure, not obviously tender on probing and often detected coincidentally (Timor-Tritsch et al 1998).

Figure 6.13 Tubo-ovarian abscess: an irregular, thick-walled multiloculated structure.

Acute urinary retention

A full bladder may sometimes be confused with an ovarian cyst. If there is any doubt and the patient is unable to void urine, a catheter should be passed to ensure the bladder is empty. Occasionally, a blood clot may be visible within the bladder.

Pelvic pain with no mass on scan

In the absence of pathology on scan, blood should be taken for leucocytosis and culture. Urinalysis should be performed and a complete infection screen, including endocervical and high vaginal swabs. Contact tracing should be performed via the genitourinary clinic. A transvaginal scan may reveal evidence of adhesions and loculated fluid, which may be indicative of previous pelvic or abdominal pathology.

Non-gynaecological pathology

It is often not possible to make a diagnosis based on clinical, ultrasound and serum findings alone. Medical, surgical and urological opinions should be sought where indicated, with early recourse to laparoscopy in those patients who have persistent or worsening symptoms.

The postmenopausal patient

Characterization of any adnexal mass is important within this age group as the risk of a mass being malignant is high (Figure 6.14). Unilocular cysts may be found in up to 20% of asymptomatic postmenopausal women. Numerous studies have shown that simple, unilocular cysts measuring <5 cm in diameter are associated with a very low risk of malignancy (Kroon and Andolf 1995). Blood should be taken for tumour markers and emergency laparotomy should be avoided if at all possible, to allow for adequate oncological work-up of the patient if indicated. Urinary retention must be excluded. Other chronic surgical and medical conditions are more predominant in the older age group, such as diverticulitis, constipation and urinary tract infections. Early recourse to advice from other specialties should be considered in women with pelvic or abdominal pain.

Figure 6.14 Ovarian endometroid adenocarcinoma: Stage Ia. Note the large papillary projection from the cyst wall.

Acute bleeding and a negative pregnancy test

This is most effectively divided into problems occurring in the premenopausal and postmenopausal patient, as the aetiology can be very different. A full history and clinical examination is essential, along with resuscitation of the patient.

The investigation of abnormal uterine bleeding will centre on an assessment of the endometrium (Figure 6.15). Undirected endometrial sampling alone has no role in the evaluation of abnormal uterine bleeding. It will miss focal lesions, such as polyps and fibroids. Whilst TVUS remains a cost-effective, non-invasive, well-tolerated technique for examining the pelvic organs, it is less specific than hysteroscopy when differentiating between endometrial polyps, myomas, carcinoma and hyperplasia. The addition of a negative contrast medium, such as saline, into the uterine cavity addresses this problem. An overwhelming quantity of data has shown that high-resolution TVUS with saline instillation is as predictive as hysteroscopy in the detection of endometrial pathology.

Figure 6.15 Assessment of acute vaginal bleeding. β-hCG, β-human chorionic gonadotrophin; FBC, full blood count; TVS, transvaginal sonography; COCP, combined oral contraceptive pill; GnRH, gonadotrophin-releasing hormone; TAH, total abdominal hysterectomy; NET, norethisterone acetate; EB, endometrial biopsy; ET, endometrial thickening; SHS, Saline Hysteroscopy Study.

The premenopausal patient

The main differentials within this group are genital tract disease, systemic disease and iatrogenic causes. When all these have been excluded, a diagnosis of dysfunctional uterine bleeding can be made. History, clinical examination and pelvic ultrasound will help to elucidate the cause. Disease of the genital tract in this age group will focus on benign rather than malignant conditions. Benign pelvic conditions will include fibroids, endometrial and cervical polyps, cervicitis, adenomyosis and endometriosis, along with pelvic infection and foreign bodies. Systemic problems contributing to abnormal uterine bleeding will include coagulation disorders, chronic liver and renal disease, and thyroid dysfunction. Iatrogenic causes will include anticoagulant therapy, intrauterine contraceptive devices and hormonal preparations. There needs to be heightened suspicion of an underlying systemic disease in younger patients presenting with heavy vaginal bleeding, as up to 20% (Kadir et al 1998) may have a coagulopathy. Screening for a coagulopathy is also advised in women with abnormal vaginal bleeding who fail medical or surgical therapy. The endometrial thickness on ultrasound will dictate the need for endometrial sampling, as will the patient’s history.

The primary goal with acute uterine bleeding will be to ensure cessation of bleeding, usually with a combination of therapies, such as antifibrinolytics, high-dose progestogens, gonadotrophin-releasing hormone analogues or the Mirena intrauterine system, until definitive management can be effected. Occasionally, urgent examination under anaesthesia with the introduction of a uterine cavity balloon is indicated to stop the bleeding. Interventional radiology with uterine artery embolization may have a role in acute management.

The postmenopausal patient

Abnormal vaginal bleeding in this age group should be attributed to malignancy until proven otherwise. Malignant tumours of the endometrium, cervix, vagina and vulva may all present with vaginal bleeding, as can ovarian tumours, such as granulosa cell tumours. It is wise to confirm that the bleeding is genital tract in origin, as occasionally haematuria and rectal bleeding may present as suspected postmenopausal vaginal bleeding.

Conclusion

The National Service Framework for Children, Young People and Maternity Services (Department of Health 2003) is clear in its aim of providing patient-centred care with the identification and appropriate management of relevant social, medical and psychiatric problems, with a one-stop assessment, diagnosis and management ethos as set out in the NHS Plan (Department of Health 2000). EGUs are excellent examples of how this aspiration can be translated into clinical practice. TVUS plays a pivotal role in assessment of the acute gynaecological patient; as such, it is the core investigating modality in the EGU. It complements a full clinical examination, affording a ‘view’ of the pelvic structures. Its integration into the gynaecology emergency service facilitates more rapid diagnosis in a number of gynaecological conditions. It also helps to exclude gynaecological pathology, ensuring prompt referral to other specialties and multidisciplinary teams. Central to its appropriate use will be training and supervision, with up-to-date protocols and regular audit, and awareness of the limitations of personnel and the equipment.