Abnormal uterine bleeding

Menstrual abnormality is one of the most common reasons for referral to the gynaecological outpatient department. Menometrorrhagia, dysmenorrhoea and intermenstrual bleeding are the most common. After the menarche, such problems are usually dysfunctional, and hysteroscopy is rarely indicated. In postmenopausal patients, bleeding is a worrying symptom and warrants urgent attention as the patient is anxious to exclude endometrial cancer. Outpatient hysteroscopy offers rapid diagnosis and reassurance. In women of reproductive age, abnormal uterine bleeding can be associated with hormonal disturbances or pathology such as fibroids and polyps. Cervical polyps may be associated with endometrial polyps in 26.7% of patients (Stamatellos et al 2007), and exclusion of endometrial pathology is important.

Infertility

Hysteroscopy is a reliable method to detect and potentially treat submucous fibroids, endometrial polyps, intrauterine adhesions and endometritis. Congenital anomalies are infrequent but are associated with infertility. Abnormalities of the endometrium and organic intrauterine pathologies are important causes of failed in-vitro fertilization/embryo transfer cycles. A recent meta-analysis has shown potential benefits of performing pretreatment hysteroscopy in patients being referred for IVF (El-Toukhy et al 2008).

Foreign body

Missing threads of an intrauterine contraceptive device (IUCD) or a deeply embedded coil may need to be removed under hysteroscopic guidance. Apart from anxiety, these patients may present with menstrual irregularities and pain.

Chronic pelvic pain

Patients with chronic pain can pose a difficult challenge to the gynaecologist. Obstructive uterine anomaly may be associated in 40% of these patients (Schifrin et al 1973). Concomitant hysteroscopy may reveal polyps, fibroids, adhesions and septate uterus which may contribute to the patient’s symptoms.

Infection

Hysteroscopy in the presence of acute pelvic infection can lead to spread of infection with the distension medium flowing through the tubes, spreading the infection to the peritoneal cavity. The only exception is when infection is secondary to a lost IUCD, which should be removed hysteroscopically under antibiotic cover.

Pregnancy

Pregnancy is another contraindication, although embryoscopy prior to 10 weeks of gestation has been performed as an aid to prenatal diagnosis. Pregnancy leads to increased uterine softening and vascularity, increasing the risks of bleeding, uterine perforation and miscarriage. Hysteroscopy during pregnancy may occasionally be performed to remove a coil where the tails are not visible.

Malignancy

Hysteroscopy is contraindicated in the presence of cervical cancer. However, it is a gold standard procedure in patients with endometrial cancer as it aids directed biopsy.

Bleeding

Diagnostic hysteroscopy may be performed in the presence of bleeding, but the view is likely to be poor. Hysteroscopy is best performed in the mid-to-late proliferative phase when the endometrium is thinnest. If this is difficult to arrange, preoperative hormonal treatment with progesterone can help to postpone or control bleeding.

Cardiopulmonary disorders

Patients with these medical conditions are at higher risk of anaesthetic when recognized hysteroscopic complications, such as gas embolism and fluid overload, occur. Outpatient hysteroscopy under local anaesthetic is therefore the treatment of choice for such patients.

Cervical stenosis

Patients who have a history of cervical surgery or difficult uterine entry in the past are at increased risk of cervical trauma, perforation and false passage during the performance of a hysteroscopy. Prostaglandins inserted 2 h before hysteroscopy may help to soften the cervix, allowing easy dilatation and entry into the uterine cavity, although the evidence is limited for their value in postmenopausal patients (Crane and Healey 2006).

Optical systems

Rigid telescopes

Panoramic hysteroscopy is performed with the help of a distension medium, and is common practice. The majority of hysteroscopists use rigid telescopes. Improved technology has allowed significant reduction in the outer diameter (as low as 2.5 mm, including the inflow sheath). The outer diameter of operative hysteroscopes using monopolar electrosurgery may be up to 10 mm. This includes the telescope with the working element, inflow sheath (inner) and outflow sheath (outer). Smaller-diameter operative scopes (Figure 2.3) measure 5 mm and consist of a 5 French operative channel communicating with the outflow sheath. This allows instruments of 5F diameter (e.g. the bipolar electrode, laser fibre, mechanical scissors and grasper) down the operative channel. The telescopes may be forward view (0°) or oblique view (angled scopes 12, 30, 70 or 90°).

Figure 2.3 Bettochhi hysteroscope.

Hysteroscopic sheaths

For diagnostic panoramic hysteroscopy, a single sheath for inflow is sufficient. This enables a smaller outer diameter and outpatient hysteroscopy without anaesthesia. Operative hysteroscopy needs separate sheaths for inflow and outflow. The inflow sheath carries the distension medium to the tip of the telescope, from where it is withdrawn via the outer sheath (Figure 2.4). Fluid circulates over the tip of the telescope to maintain a clear view.

Figure 2.4 Continuous flow concept.

Camera and stack system

The hysteroscopic image is visualized on a monitor with the help of a camera connected to a camera drive unit. The image clarity of a single chip camera is perfectly adequate. Special weighted cameras are available and facilitate orientation, but the same camera as is used for laparoscopy is equally appropriate. It is important to ‘white balance’ the camera system to ensure that the colours are displayed correctly (see Figure 2.2).

Distension medium

The anterior and posterior uterine walls are in apposition. In order to obtain a view, the uterine cavity has to be distended with gas or fluid. Carbon dioxide can be effective for diagnosis, and is delivered via a pressure reduction system or hysteroflator which is designed to give a maximum flow rate of 100 ml/min and a maximum pressure of 200 mmHg (note: a laproflator must not be used). An alternative is to use a liquid distension medium (e.g. normal saline or Hartmann’s solution) which is appropriate for diagnosis and some operative procedures. Monopolar electrosurgery requires a non-conductive distension medium such as 1.5% glycine. However, it should be remembered that once the solution becomes contaminated with blood, its function as an insulator is reduced. Fluid can be pressurized via a roller pump (with maximum flow rate of 500 ml/min and maximum pressure of 200 mmHg), gravity or a pressure infusion bag cuff system. For diagnosis, gravity or a 50-ml syringe connected to a giving set via a three-way tap provides very good control of the distension pressure. Dextran 70 (Hyskon) is no longer used due to anaphylactic reactions.

Clinical anatomy of the uterus

The uterine cavity is flat in its anterior/posterior dimension as, when not distended, the anterior and posterior endometrial surfaces are apposed to each other. The cervical canal is essentially round, but once the cavity is entered, the lateral dimension widens until it is at its broadest at the fundus. The cavity extends laterally towards the tubal ostia (see Figure 2.1). The fundal dome may be flat, concave or convex on its interior surface. Indeed, a proportion of uteri have a pronounced fundal convexity, a so-called ‘arcuate deformity’.

Endometrial preparation

A diagnostic hysteroscopic examination is best performed without medical preparation of the endometrium. The endometrium will be at its thinnest in the immediate postmenstrual phase, and at its most vascular in the premenstrual phase. Abnormal bleeding is a common indication for diagnostic hysteroscopy, and whilst it is often possible to time the examination such that bleeding is not occurring, light bleeding need not preclude successful examination of the endometrial cavity. When prolonged heavy bleeding occurs, norethisterone 5 mg three times daily for 7–10 days may be given prior to the examination. It is rarely necessary to use a gonadotrophin-releasing hormone (GnRH) analogue prior to a diagnostic procedure, but one to three doses are employed prior to endometrial ablation, or to shrink a vascular fibroid prior to operative hysteroscopic resection.

Anaesthesia

In the majority of cases, the smallest modern diagnostic hysteroscopes require no anaesthetic whatsoever. A crucial factor is good communication between the gynaecologist and the patient, with thorough preoperative counselling and the support of a skilled nurse who is able to complement the operator in explaining things to the patient. The role of local anaesthetic gels is open to question. They are inexpensive and certainly lubricate and cause the external os to open. They probably have little anaesthetic effect other than as a psychological adjunct to the gynaecologist’s reassurance.

Where facilities do not exist for outpatient hysteroscopy, the procedure is carried out in the operating theatre. Whilst a 5-mm rigid diagnostic hysteroscope can be passed through the cervix in the majority of women of reproductive age, this can prove extremely difficult in postmenopausal women and cervical dilatation is often needed. This requires a paracervical block, a regional anaesthetic or a general anaesthetic. The choice of anaesthetic method will, of course, be decided in conjunction with the anaesthetist and the patient. The choice will, in part, be determined by the patient’s state of health and whether any additional procedures are planned. It must be remembered, however, that those patients who are at risk of endometrial cancer in the postmenopausal period are often more obese and suffer from cardiovascular and airways disease. These render them at high risk from the anaesthetic point of view. Such patients provided a major stimulus to the establishment of outpatient diagnostic and, in some cases, therapeutic hysteroscopy services.

Positioning

Hysteroscopy can be performed on a general operating table equipped with lithotomy poles or Lloyd Davies supports. Care must be taken to avoid excessive pressure from the leg supports on neurovascular structures, and if there is limitation of joint movement owing to hip or knee pathology, the final position of the patient will have to be modified.

In the outpatient setting, a colposcopy chair is desirable. This allows a comfortable modified lithotomy position with minimum loss of dignity. It is essential that the patient should be kept covered until the examination is due to start.

Procedure

Inpatient/general anaesthetic

A pelvic examination is performed in order to determine the size and direction of the uterus. The vulva and vagina are cleaned with antiseptic solution, and the anterior lip of the cervix is grasped with a vulsellum forceps. The majority of rigid hysteroscopes offer a fore/oblique view ranging from 12 to 90°. By convention, the direction of the view is away from the light post. The camera system is attached to the scope with the camera orientated correctly and the light post either up or down such that the angled view is in the vertical plane. It is easier to follow the cervical canal if the scope is orientated to view along the direction of the canal (i.e. forwards and downwards in a retroverted retroflexed uterus). Crucial to obtaining a thorough hysteroscopic examination is an understanding of how rotation of the hysteroscope allows the area of uterine wall under inspection to be changed. The authors suggest holding the camera in one hand and maintaining the position of the camera fixed in relation to the vertical plane. Rotation of the scope by manipulating the light post with the other hand allows the view to be manipulated appropriately.

Insertion of the hysteroscope through the cervical canal should be performed under direct vision and, in the first instance, without cervical dilatation or the passage of a sound. This allows examination of the cervical canal and inspection of undamaged endometrium. Once instruments have been passed into the uterine cavity, they cause damage to and stripping of the endometrium, which can then give appearances suggestive of polyp formation. The image of the cervical canal during passage of the scope is, of course, dependent on the viewing angle of the scope. A 0° scope requires the cervical canal to be kept in the middle of the field of view during insertion to maintain the direction of travel of the scope parallel to the direction of the cervical canal. When an angled viewing scope is used, the position of the cervical canal in the field of view has to be offset in order to maintain the direction of travel of the hysteroscope parallel with the direction of the cervical canal. This is why orientation of the light post relative to the orientation of the camera is a crucial step in the assembly of the equipment, but it is often overlooked.

Once the cervical canal is passed, a panoramic view of the uterine cavity is obtained. Uterine distension at a flow rate of 40–60 cc/min with pressure between 40 and 80 mmHg achieves good visualization. The scope is advanced towards the fundus and rotated to allow inspection of the tubal ostia. The scope can then be withdrawn and readvanced whilst rotating it to enable systematic inspection of each uterine wall in turn.

Potential problems include blood accumulating in the cavity and obscuring the view. This can occur during diagnosis when the seal between the hysteroscope and the cervix is very tight, preventing outflow of distension medium. The passage of a dilator 1 mm greater than the outer diameter of the hysteroscope allows flow and clearance of the contaminating blood. If a large polyp is present in the endometrial cavity, it is possible to inspect the cavity without realizing that the polyp is there because the polyp fills the cavity and the scope has been passed beyond the tip of the polyp before inspection begins (Figure 2.5). Suspicion should be aroused if the endometrial surfaces are different in colour, and careful inspection of the panoramic view of the cavity during insertion and removal of the scope will avoid missing a large polyp as the tip of it will be seen. A thorough examination of the uterine cavity should allow inspection of both tubal ostia. A record of this in the operation notes demonstrates that the operator has obtained a good view.

Figure 2.5 (A) Transvaginal sonography showing polyp. (B) Sonohysterography with polyp clearly outlined.

Postoperative care

The majority of the recovery process after a diagnostic hysteroscopy relates to recovery from the anaesthetic. A small amount of vaginal bleeding is not unusual following hysteroscopy, and the patient should be warned about this. Occasionally, cramping period-like pains are experienced; these should settle within 48 h and respond to paracetamol or a non-steroidal anti-inflammatory drug such as mefenamic acid. Persistent pain and bleeding may suggest a complication such as endometritis, and the patient should seek medical help. The provision of a postprocedure information leaflet is good practice.

Hysteroscopy in outpatients

Provision of hysteroscopy combined with transvaginal ultrasound in outpatients provides a very efficient method of assessing women with pre- and postmenopausal bleeding. As mentioned above, many patients who require a hysteroscopy pose significant risk for general anaesthesia, and are best managed under local or no anaesthetic (Valli et al 1998). In addition, there are many advantages to the patient, including shorter time at the hospital and more rapid return to normal activity. There are also cost advantages to the hospital, as an outpatient clinic is clearly a less expensive environment than an operating theatre (Marsh et al 2004).

Choice of equipment

Choice of equipment will vary with the prior experience of the operator and the facilities available for disinfection/sterilization of the instruments. Rod lens hysteroscopes may be autoclaved, necessitating the provision of an adequate number of scopes for a session of activity. Fibreoptic hysteroscopes, whether rigid or flexible, cannot be autoclaved and must be sterilized with ethylene oxide or closed liquid disinfection systems.

In reality, there are disadvantages and advantages of every system. Rigid autoclavable scopes tend to be of larger diameter, necessitating cervical dilatation in a proportion of cases. The advantage is clarity of view and a fore/oblique view allowing easier examination of the cornua. Flexible scopes are delicate and more expensive, but allow steering through the cervical canal and angulation to allow full inspection of the uterine cavity without any anaesthesia (Kremer et al 1998) (Figure 2.6A). Fibreoptic rigid scopes can be of very small diameter (1.2 mm in a 2.5-mm diagnostic sheath (Figure 2.6B). They are 0° so viewing of the cornua is more difficult, but they are relatively easy to pass through stenosed postmenopausal cervices, leading to a very low failure rate. The Versascope system provides a disposable sheath which can be distended by the passage of a 5 French instrument. This means that a change of sheath is not required to convert a diagnostic procedure into an operative procedure.

Figure 2.6 (A) Flexible hysteroscope. (B) Rigid outpatient hysteroscope.

Training and learning curve

Hysteroscopy is a relatively simple surgical skill to learn, provided that the surgeon undergoes a structured training process. It is clear that familiarity with equipment and knowledge of the basic underlying principles should be imparted early in the training process. To this end, the basic skills course in obstetrics and gynaecology provides an introduction to these skills. Manipulating hysteroscopes and a camera system on models allows an early grasp of the basic principles of the orientation, and allows the trainee to progress rapidly to successful diagnostic hysteroscopy on anaesthetized patients. Outpatient diagnostic hysteroscopy requires considerable skill, confidence and familiarity with the equipment. It should be reserved for consultants and senior trainees with a specific interest.

It has been shown that achieving the operative hysteroscopy skills has a steep learning curve. A recent survey of trainees registered for the hysteroscopy module showed that the 1-year target for obtaining the special skills module was difficult to achieve, with the most evident cause being the inability to acquire the expected operative hysteroscopy standard within the intended time (Shoukrey et al 2008).

Therapeutic strategy for benign disease

Instrumentation for operative hysteroscopy

Operative hysteroscopy generally requires a larger instrument and a continuous flow sheath. In addition, a source of energy is used to perform the required manipulations.

Mechanical

Scissors and grasping forceps are passed down the operating channel of the hysteroscope, and can be used to cut the stalk of simple polyps, which are then grasped with forceps for removal (Figure 2.10). Lost foreign bodies such as IUCDs can be removed in a similar fashion.

Figure 2.10 Mechanical instruments for operative hysteroscopy.

Monopolar electrical

Monopolar electrosurgery requires flow of radiofrequency electric current from the active tip of the instrument through the patient to the large surface area return plate. The electrical effect depends on both current density and the waveform provided by the generator. Simple instruments for use with monopolar diathermy include a polyp snare and simple electrodes, which can be passed down a 5 French operating channel. A dedicated continuous flow hysteroscope is required for more sophisticated operative instruments, including a resection loop, knife, rollerball and cylinder (Figure 2.11).

Figure 2.11 (A) Working elements for operative hysteroscopy. (B) Assembled resectoscope. (C) Components of resectoscope.

Trancervical resection of the endometrium (TCRE) is a popular technique for endometrial destruction in which the loop is used to cut away the prepared endometrium down to the myometrium. Figure 2.12A demonstrates the loop in use, and the transverse myometrial fibres can be seen clearly. Movement of the loop within the hysteroscope sheath and withdrawal of the scope itself allow long strips of endometrium to be resected. Removal of the tissue strips maintains a clear view. Most operators start with the posterior wall in order that the resected tissue strips do not obscure the working area. Considerable training is needed to ensure safe use of this technique. Difficult areas to treat include the cornua where the myometrium is thin, and the fundus where the resection loop has to be bent forwards in order to achieve a cutting effect in front of the tip of the hysteroscope. Cornual endometrium may be ablated with a rollerball and combined with resection of the remaining endometrium (Figure 2.12B). A Scandinavian study (Furst 2007) reported long-term follow-up of TCRE. This showed 80% patient satisfaction, and 20% of patients required hysterectomy at the end of 10 years.

Figure 2.12 (A) Endometrial resection. (B) Rollerball ablation. (C) Laser ablation.

The same instrument is used for resection of submucous fibroids and fibroid polyps, whilst the knife is employed for the treatment of Asherman’s syndrome and uterine septae. The resectoscope system is inexpensive as it uses reuseable electrodes, which are autoclavable. A careful watch must be kept on the fluid balance during the procedure, as overload of 1.5% glycine solution may lead to significant complications. More than 1 l of solution lost into the patient poses significant risk, and the procedure should be abandoned if such levels of fluid loss are significantly exceeded. Laser destruction is an alternative (Figure 2.12C).

Bipolar electrosurgery

Versapoint (Figure 2.13A) is a 5 French electrode. It requires a dedicated electrosurgery generator, which pulses the electrical energy through the saline irrigation medium from the active electrode tip to the larger return sleeve (Figure 2.13B). This creates a vapour pocket around the active electrode, and when tissue enters the vapour pocket, the high current density causes vaporization of the tissue. This device has been used for vaporizing fibroids and polyps. It is suitable for use under local anaesthetic, particularly in conjunction with the Versascope. The Versapoint resection system, a larger version using similar technology, is suitable for the removal of larger fibroids under general anaesthetic (Figure 2.13C). Bipolar electrodes appear to have a safer profile compared with monopolar electrodes because of the unchanged serum sodium. Irrigant consumption has been reported to be significantly higher in the two bipolar groups, without any side-effects during or after the procedure. Furthermore, the TCRE loop appears to be superior to the Versapoint loop in terms of operating time and amount of tissue removed (Berg et al 2009).

Figure 2.13 (A) Bettochhi with Versapoint. (B) Versapoint spring electrode at polyp base. (C) Versapoint resection system.

Polypectomy and myomectomy

Hysteroscopic polypectomy and myomectomy are dependent on the experience of the operator and the size and position of the fibroid. Removal of the polyp can be as difficult as detaching it. If the polyp is small enough, grasping it with forceps, pulling it against the end of the scope and then withdrawing the whole scope can be effective. An alternative is to remove the detached polyp blindly using polyp forceps. This is often difficult and serves to illustrate the futility of blind techniques.

Fibroids under 2 cm in size can be resected in one sitting. For larger fibroids or those where more than half is intramural, a two-stage procedure may be necessary. In the first stage, the protruding portion of the fibroid is removed, followed by transhysteroscopic myolysis of the intramural portion (Donnez and Nisolle 1992). Subsequently, GnRH agonist therapy for 8 weeks shrinks the uterine cavity, extruding the intramural portion of the fibroid, which is then easily removed at the second stage to complete the myomectomy. Complete removal of the fibroid is associated with improved long-term results (Wamsteker et al 1993).

Uterine adhesions and septae

The division of dense uterine synechiae (Figure 2.14) and uterine septae requires considerable skill and is often performed under laparoscopic guidance, which ensures that thermal or electrical energy is not allowed to penetrate through the myometrium, damaging the adjacent bowel. Careful assessment prior to division of uterine septae is essential to prevent perforation at the fundus. The hysteroscopic appearances of a septate uterus and a uterus didelphys can be similar, and any attempt to resect the septum will clearly lead to disaster in the latter case. Careful ultrasound evaluation helps to differentiate between the two, but this has not yet replaced laparoscopic assessment. After adhesiolysis, an IUCD may be inserted for 3 months to prevent reformation of the adhesions. Similarly, postoperative hormone therapy may be initiated to help endometrial development and prevent further adhesion formation (Gordon et al 1995).

Figure 2.14 Division of intrauterine synechiae with scissors.

Operative hysteroscopy in the outpatient unit will be feasible in future. Authors have carried out transcervical resection procedures under local anaesthetic in daycare settings and have found a high level of patient satisfaction. A recent study has shown that this procedure is feasible with the availability of a mini-resectoscope (Papalampros et al 2009).

Thermal balloon ablation (Thermachoice, Gynaecare Ethicon Ltd, Somerville, NJ, USA)

Thermachoice is a silicone balloon catheter system which is inserted into the uterine cavity and inflated with 5% dextrose solution to achieve a pressure of 160–180 mmHg. The dextrose is heated to a temperature of 87° when the treatment cycle starts and continues for 8 min.

The procedure does not require any dilatation and is feasible to perform in an outpatient setting (Andersson and Mints 2007, Clark and Gupta 2004). The National Institute for Health and Clinical Excellence (NICE) has approved and validated the clinical safety and effectiveness of thermal ablation (see Figure 2.15).

Figure 2.15 Thermal balloon ablation system.

A large multicentre observational study showed a significant reduction in the severity and duration of menstrual loss (Amso et al 1998). Thermachoice achieved a reduction in menstrual blood loss that was comparable with the first-generation techniques. In a long-term multicentre study, Thermachoice showed an 86% probability of avoiding a hysterectomy at 4–6-year follow-up (Amso et al 2003).

Other thermal balloons used in ablation are Cavaterm and Thermablate.

Cryoablation (Her Option, Uterine Cryoablation Therapy System, Cryo-gen Inc., San Diego, CA, USA)

The principle of cryotherapy is used for cryoablation. Liquid nitrogen is used to produce extreme cooling, producing an iceball at the tip of the cryoprobe. The probe is moved from one cornua to the other and through the entire cavity in a series of freeze–thaw cycles. Ultrasound guidance is needed to monitor the position and the depth of the probe. Tissue necrosis of 9–12 mm is achieved using this method in a 4-min cycle. The procedure is recommended as quick, easy and feasible for outpatient settings. However, severe abdominal cramping and vaginal pain, prolonged tiredness and perimenopausal symptoms were some of the major side-effects and late complications.

Free fluid ablation (Hydrothermablator, Boston Scientific, Natick, MA, USA)

In this technique, circulating hot saline solution is instilled into the uterus under direct hysteroscopic guidance. The saline is heated to 90°C and circulated for 10 min. A randomized controlled trial comparing Hydrothermablator with rollerball ablation (Townsend et al 2003) showed similar success rates as other second-generation devices (see Figure 2.16).

Figure 2.16 Free fluid endometrial ablation system (Boston Scientific, Natwick, MA, USA).

Impedance bipolar radiofrequency ablation (NovaSure, Novacept, Palo Alto, CA, USA)

Bipolar radiofrequency under impedence control is used for endometrial ablation. The bipolar current provides shallower ablation in the cornual region and deeper ablation in the main cavity. The device is a three-dimensional conformable bipolar gold-plated mesh mounted on an expandable frame. After insertion, bipolar radiofrequency energy travels between the electrodes and initial electrical pulses vaporize the low-impedence endometrium. As the energy travels deeper into the muscle and begins to desiccate the superficial myometrium, the resistance grows. The system automatically terminates the procedure when an impedence of 50 ohm is reached, signifying adequate ablation.

The device can be inserted into the uterus with a cavity length between 4 and 12 mm. The mean ablation time is 90 s. Ablation is based on physical tissue characteristics and not on temperature, pressure or time. This is claimed to achieve high ablation success rates regardless of endometrial thickness. Hence, pretreatment is not necessary. The short duration of the procedure makes it feasible for use in the outpatient setting under local anaesthetic.

Efficacy was reported in two randomized controlled trials and two case series (Abbott et al 2003). Amenorrhoea rates in these four studies ranged from 41% to 59% at 12 months, with higher amenorrhoea rates in the impedance controlled procedure. Continuing menorrhagia rates ranged from 3.9% to 14% at 12 months. There was also evidence to suggest that the majority of patients were satisfied following the procedure, and that quality of life had improved. At 3-year follow-up (Gallinat 2004), hysterectomy was avoided in 97% of women, 65% of whom were amenorrhoeic. Five-year follow-up of a randomized controlled trial comparing NovaSure and Thermachoice endometrial ablation showed that bipolar radiofrequency ablation was superior to thermal ablation (Kleijn et al 2008).

Microwave endometrial ablation (MEA, Mircosulis Plc, Waterlooville, UK)

The MEA device utilizes low-power, high-frequency microwave energy to generate heat and destroy the endometrium. The probe is inserted into the uterine cavity and moved from side to side with the temperature maintained at 75–80°C to ablate the endometrium.

MEA was compared with TCRE in a well-designed randomized controlled trial (Cooper et al 1999), and was reported to have a significantly shorter mean operative time than TCRE (11.4 vs 15.0 min, P = 0.001), and comparable satisfaction and acceptability. Improved quality of life 1 year after treatment was reported for both groups. These results were sustained at 2-year follow-up. At 5-year follow-up, both techniques achieved significant and comparable improvements in menstrual symptoms and health-related quality of life (Cooper et al 2005). High rates of satisfaction with and acceptability of treatment were achieved following TCRE, but these rates were significantly lower than those following MEA. The long-term data after a further 5 years of follow-up, when combined with the trial’s operative findings and the known costs of both procedures, led the authors to conclude that MEA is a more effective and efficient treatment for heavy menstrual loss than TCRE (Sambrook et al 2009).

Complications of second-generation ablation devices

Minor immediate postoperative complications are common with second-generation ablation techniques and include pelvic pain, endometritis, urinary tract infection, nausea, vomiting, haematometra and pelvic inflammation. Reported serious complications included sepsis, adnexal/uterine necrosis, lower genital tract thermal injury and one death. Many of these resulted from non-compliance with the manufacturers’ instructions and other avoidable factors. A 2005 update of a Cochrane review (Lethaby et al 2005) concluded that equipment failure, nausea, vomiting and uterine cramping were more likely with second-generation ablation techniques but these were less likely to be associated with fluid overload, uterine perforation, cervical lacerations and haematometra than conventional ablation and resection techniques. Overall, they compared favourably with TCRE. Most of the second-generation devices (Cavaterm, Thermachoice, NovaSure, Cryoablation) require a relatively normal uterine cavity with a depth of less than 12 cm to complete the treatment satisfactorily. Free fluid ablation (Hydrothermablator) and MEA can be performed when the uterine cavity is irregular, although complete septate or bicornuate uteri are contraindications to undertaking the procedures. Pretreatment is not generally required.

Each device has its relative strengths and weaknesses, and no single device can be preferred over another in all circumstances. Clinicians should consider the safety and effectiveness figures, applicability to the outpatient environment, and evidence of quality-of-life improvement, applied on an individualized basis after full discussion with the patient and taking her preferences into account, as well as equipment availability and skill. Ongoing audit of a department’s success rates compared with the published figures ensures that patient selection and surgical strategy are appropriate. A survey of preferences and practices of endometrial ablation/resection for menorrhagia in UK showed that second-generation devices were preferred over first-generation devices. Thermal balloon ablation was the most favoured procedure (Deb et al 2008).

Essure system

A safe, simple and highly effective alternative to invasive procedures for interval tubal sterilization is now possible using a hysteroscopic approach. The Essure permanent birth control device (Conceptus Inc., San Carlos, CA, USA) is a combined metallic and fibre coil placed into the cornua hysteroscopically. Approval for its use has been granted by the US Food and Drug Administration (FDA) and NICE. The dynamic outer coil made of nickel titanium alloy includes an inner flexible coil (stainless steel). The device is delivered into the fallopian tube in a tightly wound or ‘low profile’ position, where the two coils are tightly opposed and held in place with a wire before being released in the tubal lumen. The rapidly expanding outer coil fills the tubal lumen (Figure 2.17), anchoring the device in place. The polyethylene terephthalate fibres induce a benign localized tissue response consisting of inflammation and fibrosis, which leads to obliteration and occlusion of the tubal lumen in approximately 3 months (Valle et al 2002, Abbott 2007). The procedure is preferably done in the proliferative stage of the menstrual cycle. The procedure can be performed under local anaesthetic (paracervical nerve block with lidocaine 1%) in the outpatient setting (Sinha et al 2007). Hysteroscopy is then performed and tubal access is established. During insertion, the microinsert is maintained in a ‘wound down’ configuration by a release catheter, which is enclosed in a transparent delivery catheter. A guide wire is attached to the inner coil to facilitate device placement. The device is then deployed in the proximal tubal ostium with a single-handed release mechanism, and the guide wire is detached and withdrawn. Follow-up pelvic X-ray or hysterosalpingogram is recommended 3 months after insertion to confirm correct placement of the device or tubal occlusion. An occlusion rate of 100% has been shown at the end of 1 year with no pregnancies reported (Kerin et al 2003).

Figure 2.17 Essure microinsert. This cross-sectional view of the uterus and fallopian tube shows the device placed across the uterotubal junction (external).

A comparison of the Essure permanent birth control device and laparoscopic sterilization (Duffy et al 2005) reported reduced hospital stay, less pain, better patient acceptability, higher patient satisfaction levels and fewer adverse events with the Essure procedure. A separate study of the direct costs associated with the Essure procedure compared with laparoscopic sterilization indicated that there is a cost advantage to Essure, although indirect costs were not considered in this model (Levie and Chudnoff 2005).

Although outpatient sterilization offers many potential advantages, uptake is currently affected by cost, additional follow-up and the 3-month delay in providing permanent fertility control. Long-term results are still awaited.

Alternative techniques and future developments

The Adiana procedure (Adiana Inc., Redwood City, CA, USA and Hologic, Inc., Bedford, MA, USA) involves bipolar ablation of the proximal fallopian tube followed by the placement of a synthetic matrix, subsequently leading to tubal occlusion. The device is currently being investigated in an FDA study in the USA, Australia and Mexico (Johns 2005) (Figure 2.18).

Figure 2.18 Adiana hysteroscopic delivery catheter.

Surgeon related

Inadequate training and lack of experience can lead to avoidable, but potentially major, complications, such as uterine perforation, haemorrhage and other visceral injuries.

Distension medium

There are different types of distension media used for hysteroscopy. Carbon dioxide may lead to gas embolism, which can lead to hypotension, tachycardia and arrhythmias, which can be fatal. Fortunately, this is a rare complication and should not occur provided the correct pressure insufflator is used.

Electrolyte imbalance from excessive absorption of non-electrolytic liquid distension medium such as 1.5% glycine can lead to hypervolaemia and hyponatraemia; a deficit of 1 l requires caution and surgery should be discontinued at 1.5 l. If the patient is asymptomatic with normal serum electrolytes and haematocrit, observation is sufficient. However, patients may develop bradycardia, hypertension, nausea, vomiting, headache, confusion, agitation and visual disturbances. This is due to water moving across the blood–brain barrier, leading to cerebral oedema. If hyponatraemia is present with haemodilution, a loop diuretic (e.g. frusemide) and restricted fluid intake may be helpful. This symptom complex has been described in the past as ‘post-TURP syndrome’ (TURP, transurethral resection of prostate). Careful assessment of fluid input and output is critical, particularly if glycine is used.

Haemorrhagic

Bleeding during hysteroscopy is caused by inadvertent trauma or by the operative steps disturbing the endometrium. Thick and vascular endometrium is likely to cause heavy bleeding. Preoperative endometrial thinning to decrease the vascularity is recommended practice. Bleeding can be controlled during surgery by coagulating the vessels under direct vision, or after surgery by inflating a Foley balloon in the uterine cavity with 5 ml of normal saline. The balloon is then deflated 6 h after surgery. If bleeding has settled, the catheter should be left in place for 1–2 h before discharging the patient. If not, the balloon can be left for a maximum of 24 h. If bleeding continues despite tamponade by the Foley balloon, hysterectomy may become necessary. In the MISTLETOE study, 1% of women undergoing endometrial resection required an emergency hysterectomy.

Bleeding is likely to occur during endometrial ablation, myomectomy, adhesiolysis and septum resection. Polyps are unlikely to bleed significantly. Arterial bleeding is pulsatile and should be stopped immediately using electrosurgery. The distension medium can suppress venous ooze. The cavity should be viewed after the pressure is reduced to check for haemostasis after operative hysteroscopy, and the hysteroscopist should always be prepared to control bleeding, which can happen unexpectedly.

Traumatic

At the beginning of any hysteroscopy, due care must be taken while grasping the cervical lip as good grip ensures that the tenaculum does not cut through. Dilatation of the cervix, if needed, is a blind procedure. If one is not careful, a false passage may be created and, occasionally, perforation of the cervical canal. This is more likely to occur when the cervix is scarred and stenosed following previous surgery, and some surgeons prefer to soften the cervix with vaginal prostaglandins 2 h before surgery. Rarely, the dilator can give way suddenly, resulting in perforation of the fundus.

Introducing the hysteroscope under direct vision, as described above, minimizes the chance of cervical trauma or uterine perforation during introduction of the scope.

Uterine perforation

Uterine perforation occurs in approximately 1% of women undergoing endometrial ablation (Lewis 1994). It can happen due to entry of the scope into a false passage, when the scope is against the fundus, or while using the energy sources without visualizing the electrode. It is suspected if excessive bleeding is noted before the procedure is commenced, if the scope goes freely into the uterine cavity, or if the fluid distension pressure stays low or decreases suddenly. Rarely, a loop of bowel or omentum may be seen.

The energy source should only be activated while withdrawing the active tip, keeping the working element constantly under vision, which decreases the risk of perforation.

The distension medium tends to stretch the uterine wall, which is normally 2–2.5 cm thick. This has to be understood while performing operative hysteroscopy to avoid myometrial damage and decrease the risk of perforation. Extra care should be taken while operating near the uterine cornu as this is the thinnest portion of the uterine wall. Cornual injuries are likely to be full thickness with an additional risk to bladder, bowel and pelvic vessels.

As soon as perforation is suspected, the procedure should be stopped. Diagnostic hysteroscopy may be attempted, but the visualization will be poor due to inadequate distension and bleeding. Laparoscopy will help in assessment of the perforation site. The majority of injuries can be managed conservatively. In the event of heavy bleeding, haemostasis should be achieved by coagulation or by suturing the perforation. Diagnostic hysteroscopy may be completed under laparoscopic control. If perforation occurs during an operative hysteroscopy, it is wise to defer the procedure.

Infections

Diagnostic hysteroscopy is less likely to cause infection. During operative hysteroscopy, blood and tissue debris becomes a focus of infection. Energy applied to tissue leads to necrotic tissue in the uterine cavity. Infection leads to blood-stained offensive discharge, which is accompanied by fever, generally feeling unwell and raised white cell count. A high degree of suspicion in postoperative patients with deteriorating symptoms is crucial for early diagnosis. Prophylactic antibiotics should be considered for many operative hysteroscopy procedures.

Anaesthetic and energy source related

These complications are rare but still need to be discussed with patients. Patients at high anaesthetic risk will need a thorough explanation of the risks involved, and should be offered the alternative of having the procedure under local anaesthetic.

Infertility

Neosynechiae formation after hysteroscopic surgery may lead to Asherman’s syndrome, and adhesions can reform after adhesiolysis. Infertility may persist despite treatment of the suspected cause.

Abnormal menstrual bleeding

Hysteroscopic surgery performed in the younger age group (40–45 years) is more likely to fail than that performed in older age groups. Counselling to ensure realistic postoperative expectations is crucial to improve patient satisfaction.

Pregnancy

Patients undergoing endometrial ablation should be made aware of the need to use contraception as sterility cannot be guaranteed.

Cancer

Cervical cancer is a contraindication for hysteroscopy, but endometrial cancer is not. Failure of endometrial ablation and lack of endometrial thinning despite agents (e.g. GnRH analogues and danazol) should raise suspicion. Inability to shrink a fibroid by preoperative GnRH analogue treatment is another cause for concern. There is no evidence of retrograde spread with the fluid or gaseous distension media, or of any adverse effects on the activity of the malignant cells. Similarly, histology of all specimens obtained at hysteroscopic surgery must be reviewed. Sarcomatous change in a fibroid occurs in less than 1% of patients, but must be ruled out. Endometrial chips obtained at resection may occasionally reveal malignancy.

Haematometra and pyometra

Collection of blood in the uterine cavity followed by infection can occur after hysteroscopic surgery due to cervical stenosis. This can be prevented by avoiding ablation near the internal os. Occasionally, haematometra may occur after commencement of hormone replacement therapy (Dwyer et al 1991).

Postablation sterilization syndrome

Endometrial ablation in women with a history of sterilization can occasionally lead to postoperative pain. Active endometrium between the scarred uterine cavity and the tubal block leads to haemorrhagic tubal distension during menses. This leads to stretching of the intramural and isthmic portion of the tube with resultant pain. The presentation of this condition can be similar to an ectopic pregnancy.

Consent issues

Thorough preoperative explanation and counselling, including detailed explanation of the pathophysiology of the underlying gynaecological condition, treatment options available and their risks and benefits, is essential to ensure a successful operative experience for the patient as well as the surgeon. This is best started during the outpatient consultation when the decision for hysteroscopy is taken, and followed by giving an information leaflet with a helpline telephone number to clarify any further doubts.