Endometrial ablation (EA) is a minimally invasive surgical intervention that aims to reduce heavy menstrual bleeding (HMB) by destroying functionally active endometrial glands within the endometrium and the superficial myometrium, including the deep basal glands. To prevent regeneration and stop menstruation this destruction should be to a depth of 5 mm. In the past, destruction of the endometrium required an operating hysteroscope. Endometrial tissue was either removed using an electrical cutting loop or destroyed by applying thermal energy to induce necrosis using an electrical ‘rollerball’ or laser fibre. These first-generation techniques have largely been superseded by second-generation techniques that comprise semi-automated global ablative systems using a variety of energy sources to thermally ablate the endometrium. These systems require less operator skill, are less likely to require general anaesthesia, are quicker to perform and offer enhanced safety with no loss in effectiveness.
Endometrial ablation (EA) is a minimally invasive surgical intervention that aims to reduce heavy menstrual bleeding (HMB) by destroying functionally active endometrial glands within the endometrium and the superficial myometrium, including the deep basal glands. To prevent regeneration and stop menstruation this destruction should be to a depth of 5 mm. In the past, destruction of the endometrium required an operating hysteroscope. Endometrial tissue was either removed using an electrical cutting loop or destroyed by applying thermal energy to induce necrosis using an electrical ‘rollerball’ or laser fibre. These first-generation techniques have largely been superseded by second-generation techniques that comprise semi-automated global ablative systems using a variety of energy sources to thermally ablate the endometrium. These systems require less operator skill, are less likely to require general anaesthesia, are quicker to perform and offer enhanced safety with no loss in effectiveness [1, 2].
EA is a quick day-case or outpatient ‘office’ procedure that allows women to conserve their uterus and avoid the morbidity and costs associated with hysterectomy. Indeed, the introduction of EA along with effective medical therapies such as the levonorgesterol-releasing intrauterine system (LNG-IUS) has led to a steady decline in the rates of hysterectomy as a treatment for HMB . This trend is likely to continue with the introduction of third-generation technologies that are being developed. These technologies use novel energy systems, increasing the speed of treatment and providing further miniaturisation and enhanced portability, thereby increasing their suitability for treatment in a more accessible outpatient or community-based setting. This chapter will discuss patient selection and available technologies, and consider the steps necessary to optimise clinical outcomes in terms of both safety and effectiveness.
Endometrial ablation is indicated in women with HMB who want to preserve their uterus but have no future fertility desires and who decline medical treatment or have found it to be unacceptable or unsuccessful . It is inappropriate for women who have not completed their family. Absolute contraindications for all techniques include surgeries that can lead to weakening of the myometrium, such as that following classical caesarean section or myomectomy. Chronic high-dose corticosteroid use can also be associated with reduced myometrial strength, and so caution should be advised in these women and ablative procedures probably avoided. In addition, it should not be performed if there is active pelvic infection at the time of the procedure or there has been pre-malignant or malignant endometrium (see Table 11.1).
There are circumstances in which EA should be undertaken with caution. For example, younger patients in general have a greater likelihood of needing additional treatment (hysterectomy or repeat ablation) within five years of the procedure. In North Carolina, post-ablation hysterectomy was found to be as high as 40% in women under 40 years of age, with the risk continuing for the eight years of follow-up, compared with less than 20% for women aged 45 to 49.9 years . A large UK study reported a retreatment rate of 27% within five years for women aged 35 years at treatment, following either first- or second-generation treatment; the rate for women aged 45 years or more was only 10.4% . Hysterectomy was the most common subsequent procedure (77%) particularly in younger women; repeat EA was performed in 23% . However, EA at any age remains an option, provided the possible need for further treatment is accepted, if the patient wishes to avoid hysterectomy either at all, or at that particular stage in her life.
Caution needs to be exercised if the woman has had previous, extensive abdominal surgery, or severe pelvic infection, where there is a risk of the bowel having become adherent to the serosal surface of the uterus. It is not expected that energy from any ablation device will heat the serosal uterine surface , but bowel damage can occur without uterine perforation with previous bowel surgery as the only risk factor . The underlying mechanism for ‘non-target’ thermal injuries is unclear but may arise from the ablative device becoming embedded within the myometrium. Partial perforation of the uterine wall allows approximation of the device to the serosal surface that may predispose to transmission of heat beyond the uterus.
The presence of chronic pelvic pain, either cyclical or constant, may reflect concurrent endometriosis or adenomyosis; these are conditions that will not be improved by EA, or may worsen after it. The 2018 NICE guideline on HMB advises additional investigations in these circumstances, initially with an ultrasound scan, though an MRI scan may be necessary, the results of which may help determine the most appropriate treatment . HMB with pain associated with adenomyosis or endometriosis may be better treated with a LNG-IUS. However, in the absence of such conditions, or even where they are known to exist, EA is still a viable option, being effective in around 60–70% of cases .
One common reason for further treatment after EA is the presence of pelvic pain as a new symptom or its failure to resolve. This may be accompanied by deep dyspareunia. New pain, often cyclical, may be a consequence of induced ‘iatrogenic’ adenomyosis or the development of a haematometra arising from pockets of regenerated or untreated endometrium within a scarred and restricted uterine cavity. Post-ablation sterilisation syndrome, where blood from regenerated or untreated endometrium, often around the cornua, is unable to escape and gradually increases over time, may also be the underlying cause of cyclical pain . Suppression of menstruation using high-dose, systemic progestogens or gonadotrophin-releasing analogues (GnRHa) can provide temporary relief of symptoms and indicate that a hysterectomy would permanently alleviate symptoms.
The myometrial thickness at the level of a caesarean section (CS) scar may be reduced and consideration should be given to measurement of the thickness by transvaginal ultrasound scan (TVS). When using a second-generation device the manufacturer’s information for users of that device should be followed, especially regarding pre-operative diagnostic workup. Most currently available devices do not stipulate a need to undertake a pre-procedural TVS, but a deficient myometrium overlying the lower segment scar and/or a fluid-filled cervical niche are more likely in women with more than one CS or following a complicated CS. CS scars should be investigated during the hysteroscopic examination preceding the EA, looking specifically for a substantial cervical niche or a more distal isthmic scar from a ‘high’ uterine incision. In such circumstances, alternative treatment, including the use of first-generation resection or ablation where the scar can be specifically avoided may need to be considered. Particular care should be taken when deploying second-generation devices after a CS to ensure that delivery of more proximal thermal energy into the cervical canal is avoided.
A particularly small uterine cavity (<4 cm) may preclude the use of any second-generation device as it may be too small to deploy at all or safely. In addition, substantially enlarged uterine cavities arising from uterine fibroids or adenomyosis, such that clinically the size is estimated to be more than that of a 12-week gravid uterus, are not suitable because the effectiveness of any device is likely to be compromised as it may not gain contact with all of the endometrium. Uterine sound lengths ≤8 cm are associated with better outcomes (i.e. higher satisfaction). When counselling women, clinicians should be aware that outcomes may be more variable with a larger uterus .
In general, EA with a second-generation device is suitable for women with HMB and a normal sized and shaped uterine cavity when complete endometrial coverage and depth of ablation can be safely and effectively achieved. In women with structural or acquired intrauterine anomalies, such as congenital malformations or substantial submucosal fibroids (>3 cm diameter), it may not be possible to deploy the second-generation EA device because of cavity distortion. First-generation EA remains indicated when combined with resection of a submucosal fibroid . Data suggest a trend towards lower satisfaction with an EA if conducted in the presence of an endometrial polyp or submucosal fibroid . Removal of these acquired lesions prior to EA is advisable.
Women with HMB can be treated empirically without the need for mandatory investigations because: (i) HMB is common, (ii) most pathology is benign, (iii) effective medical interventions can be initiated in primary care. However, while the accessibility and implementation of treatment within primary care may be enhanced, investigations to evaluate the uterus can allow the underlying aetiology of HMB to be ascertained. This knowledge facilitates tailored patient care, arguably improving treatment outcomes through recommending the most appropriate treatments . For example, identified submucosal fibroids can be removed hysteroscopically, endometrial hyperplasia treated with an LNG-IUS and large fibroids by hysterectomy.
11.3.1 Endometrial and Uterine Assessment
Some form of evaluation of the uterus and uterine cavity is advisable to help gauge the appropriateness of an EA and to exclude uterine pathologies that would contraindicate the procedure (Table 11.1). A significantly enlarged uterus, usually due to the presence of fibroids, is readily detectable on bimanual pelvic examination in most compliant women of a relatively normal body mass index (BMI). However, a pelvic ultrasound scan can more accurately identify an enlarged uterus caused by fibroids or adenomyosis, especially in obese women. These conditions, as discussed earlier, do not necessarily preclude an EA but alternative treatment options may be considered. Moreover, TVS can diagnose structural intracavity abnormalities such as congenital defects, endometrial polyps and submucous fibroids with good accuracy. In a woman without significant pelvic pain and a normal pelvic examination, an outpatient hysteroscopy (OPH) is preferable to a TVS because intracavity abnormalities can be detected with greater accuracy. In addition, OPH facilitates the taking of an endometrial biopsy, which can exclude endometrial hyperplasia and endometrial cancer. It should be noted that an endometrial biopsy is recommended regardless of whether an OPH is performed prior to an EA. The experience of an endometrial biopsy with or without an OPH may also help a woman decide whether she would wish to have endometrial ablation performed in an outpatient setting.
A pre-procedural hysteroscopic inspection of the uterine cavity has been recommended immediately prior to inserting a blind, second-generation ablative device . Thus, a hysteroscopy can be deferred until the time of the ablative procedure. However, the likelihood that an endometrial procedure is not feasible on the day of surgery is increased because of unanticipated problems accessing the uterine cavity or pathologies within it. Failure of adequate workup has greater implications when planning an EA under general anaesthesia from a convenience, safety and health service resource point of view. If an endometrial biopsy is only performed immediately prior to the EA, then it is generally considered acceptable to proceed as long as the patient is aware of the small chance of discovering endometrial disease not revealed by the hysteroscopy. The threshold to proceed without a biopsy should be higher in women with risk factors for endometrial disease such as obesity.
The important points to cover when advising women with HMB about endometrial ablation are summarised below. Provide written information, too, and direct women to where online information can be obtained, such as the RCOG website .
Women must have completed their family or not desire future fertility. However, they should be counselled that it is not a form of contraception and this should be discussed in clinic. The practitioner should establish what form of effective contraception they plan to continue to use post-procedure. Women should be informed that there is a pregnancy rate of 0.7% following endometrial ablation  and that such pregnancies may be more likely to be complicated by miscarriage, premature delivery or placental problems, including growth restriction and bleeding.
Women should be aware that the aim of treatment is to reduce HMB, thereby alleviating the burden of symptoms and enhancing HRQL. Amenorrhoea cannot be guaranteed and at one year post-treatment is of the order of 40%. Approximately 80–90% of women will be satisfied with the outcome of treatment at one year; in those dissatisfied it will usually be because of ongoing HMB, erratic bleeding, vaginal discharge or persisting or new period pain. At five years, the rates of further surgical intervention (repeat ablation or hysterectomy) are 16–25% [3, 4, 14] but may be higher or lower, dependant on prognostic variables such as age, uterine size and pathology, as previously discussed.
While it is not possible to individualise prognosis with a great deal of precision, women should be made aware of factors that may affect the likelihood of successful treatment to aid their decision-making. In particular, they should know that treatment outcomes, especially the reduction in HMB and incidence of amenorrhoea, appear to be better in older women (>45 years), women without fibroids and women with a smaller uterus (≤8 cm) . The impact of pre-existing pelvic pathologies (e.g. pelvic inflammatory disease, endometriosis or adenomyosis) and cyclical or non-cyclical pain on treatment outcomes is unclear.
Women should be advised what to expect and what to avoid during the first few days and weeks following EA. A watery, blood-stained vaginal discharge will occur for a few days, and then become clearer. However, some discharge may last for several weeks as the endometrium heals. This is not unusual and not a cause for concern, provided the discharge is not offensive and not associated with pelvic pain. Tampons are best avoided at this stage to reduce the risk of infection, and sanitary pads used instead.
Cramping, period-type pains are to be expected and regular analgesia, such as paracetamol or ibuprofen, is recommended for the first 48 hours; codeine-containing medication may be preferred by some women. Women should be made aware that worsening pain with fever, with or without persistent, foul-smelling discharge, may indicate a uterine infection and immediate medical attention should be sought as antibiotics will be needed.
Exercise and physical activity the day after the procedure will not cause harm and the usual activities, including work, can be resumed within a few days following an uncomplicated treatment. Sexual intercourse should be avoided until the vaginal discharge has resolved.
Women should be made aware of the potential complications arising from the procedure. Minor complications include genital tract infection (endometritis or urinary tract infection) and prolonged post-operative bleeding or vaginal discharge. More serious complications are rare and include significant genital tract trauma (uterine perforation requiring further medical or surgical intervention such as hysterectomy), damage to the bowel or urinary tract and septicaemia arising from endometritis .
11.5 First-Generation Procedures
First-generation techniques for EA refer to approaches to destroy the endometrium and superficial myometrium under direct vision through hysteroscopy. The most common techniques include transcervical resection of the endometrium (TCRE) or rollerball endometrial ablation (REA). These procedures are performed with a resectoscope, as described in Chapters 3 and 5. The rigid operative hysteroscope, typically around 4 mm in diameter with an offset 12–30° forward-oblique distal lens, is housed within inflow and outflow sheaths providing continuous irrigation of the uterine cavity to maintain an operative view. The hysteroscope accommodates a moveable electrode, such as a cutting loop for resection, rollerball for ablation and rollerbar for vaporisation. The forward-oblique distal lens affords a wide field of view, enabling magnified visualisation of the electrode. The outer diameter is normally 8–9 mm and so placement within the uterine cavity requires cervical dilatation; consequently TCRE and REA are usually performed under general or regional anaesthesia in an operating theatre.
Resectoscopes provide the gynaecologist with a highly versatile tool allowing TCRE, REA and removal of focal uterine anomalies such as endometrial polyps and submucosal fibroids (Chapter 12). Knife electrodes facilitate surgical division of a uterine septum (hysteroscopic metroplasty) (Chapter 14).
Resectoscopes conventionally utilised monopolar electrical circuits that require non-conducting fluid media, such as sorbitol or glycine, which are hypotonic. However, excessive fluid absorption intraoperatively can lead to serious complications arising from fluid overload and hyponatraemia (see Chapter 9). Advances in technologies have led to the development of equally effective bipolar resectoscopic systems with the safety advantage of using isotonic distension media (e.g. normal saline) with reduced risks of fluid overload and avoidance of hyponatraemia.
TCRE involves resecting the endometrium under vision with the use of a small, typically 3–5 mm, electrosurgical diathermy arch-shaped loop. Irrigation fluid enables a direct view of the cavity. Resection should begin from one tubal ostium and on the posterior wall of the uterine cavity. The operator should assess the depth of the resection and, if adequate, continue in a systematic fashion around the uterine cavity, making grooves from the fundus towards the isthmus and always moving the electrode towards the surgeon . The isthmus and uterine cornual areas, where the myometrium is thinnest, should be left intact. While a cutting electrode can be used throughout, once resection of the uterine body has been completed, most surgeons use a rollerball electrode to cauterise any untreated areas in the uterine cornua, fundus and isthmic areas (Figure 11.1).
The strips of endometrial tissue created from the resection should be removed at intervals from the cavity to allow a constant good field of view, and sent for histological analysis to exclude endometrial cancer or hyperplasia. An alternative technique is to remove the strips of tissue (‘chips’) after each cutting pass of the electrode. Although this has the advantage of avoiding blind uterine instrumentation to retrieve tissue, the disadvantage of multiple insertions of the resectoscope may prolong the procedure and risk air embolism (see Chapter 9). Hysteroscopic myomectomy (TCRF) can be performed during the same procedure (see Chapter 12) .
REA involves destruction of the endometrium and superficial myometrium by coagulation using a 2 mm rollerball electrode comprising a moveable ball or cylinder. Tissue is destroyed to a depth of 5 mm and this technique is therefore considered safer in areas of thin myometrium, such as in women with previous lower-segment CS . Successfully treated areas of the cavity can be identified because the endometrium becomes a pale yellow colour indicating that myometrial tissue has been coagulated. Curettings are not obtained with this technique, therefore a pre-operative endometrial biopsy is mandated to exclude endometrial disease.
An alternative technique is to use a miniature bipolar electrode (e.g. Versapoint®) to ablate the endometrium. However, given their small size, they are best used in women with cervical stenosis precluding cervical dilatation or after a previous EA where bleeding symptoms persist from a small amount of active, refractory endometrium within a constricted, post-ablative uterine cavity.
Complications of hysteroscopic surgery are discussed in detail in Chapter 9. The large-diameter electrical resectoscopes may give rise to mechanical trauma, also thermal injury to surrounding structures, bleeding and systemic complications arising from fluid overload. The likelihood of cervical trauma, false passage creation and uterine perforation can be minimised by attaining proficiency through adequate supervised training, attention to good technique, maintenance of good visualisation within the uterine cavity and a sufficient caseload to maintain skills. Intraoperative and post-operative bleeding can be minimised by adopting a gentle, atraumatic technique and avoiding resection too close to the cervical canal.
Fluid overload can lead to compromise of cardiorespiratory and central nervous systems, and even death, through intracellular fluid absorption and resultant oedema. Excessive fluid absorption should be suspected when there is a large unaccountable fluid deficit during the procedure, as indicated by the inflow of fluid being significantly more than the outflow. Sequelae are greater with non-isotonic, non-osmotic solutions, sorbitol or glycine, used with monopolar circuits, causing electrolyte imbalance with hypo-osmolar hyponatraemia. Strict fluid balance monitoring is essential, and the procedure must be terminated if the maximum recommended deficit of 1.5 l of non-isotonic fluid is reached , though the actual safe amount of fluid overload will depend upon patient co-morbidities. When using such solutions the flow rate and pressure of the fluid should be controlled through use of an infusion pump or pressure bag when hanging the fluid on an infusion pole . Fluid overload risks are less when using isotonic, isomolar media, such as 0.9% normal saline, as used with bipolar resection. The maximum fluid deficit advised in this circumstance is 2.5 l, but again may need to be less if the patient has certain co-morbidities .
First-generation ablation technique complications were reported by the MISTLETOE study of 10 000 women who underwent first-generation endometrial ablation in England and Wales . Serious complications included emergency hysterectomy in 6.6 per 1000 procedures and a mortality rate of 0.26/1000.
A Cochrane systematic review by Lethaby et al., compared complication rates between first- and second-generation techniques (Table 11.2) . Perioperative and post-operative complication rates were lower with second-generation devices, but this was not statistically significant . Perioperative complication rates are higher in those undergoing a repeat ablation procedure .