INTRODUCTION — Patients with abnormal uterine bleeding, including heavy menstrual bleeding, have a variety of therapeutic options. Endometrial ablation has become an increasingly popular treatment since it is minimally invasive and successful ablation avoids chronic use of medications. In a national study in England from 2003 to 2006, 60 percent of all surgical procedures performed for menorrhagia were endometrial ablations [1].

Endometrial ablation is the surgical destruction of the uterine lining. This can be accomplished under hysteroscopic visualization, using resectoscopic instruments to ablate or resect the endometrium. It can also be performed with a non-resectoscopic ablation device, which is inserted into the uterine cavity and delivers energy to uniformly destroy the uterine lining. Non-resectoscopic endometrial ablation techniques are more widely practiced than resectoscopic ablation, since they require less specialized training and often have a shorter operative time [2]. Endometrial ablation is not appropriate for patients with endometrial hyperplasia or cancer, or those who wish to preserve their fertility.

This topic review will provide an overview of endometrial ablation. Techniques for non-resectoscopic endometrial ablation and resectoscopic endometrial ablation or resection, as well as other management options for abnormal uterine bleeding, are discussed separately. (See "Endometrial ablation: Non-resectoscopic techniques" and "Endometrial ablation or resection: Resectoscopic techniques" and "Abnormal uterine bleeding: Management in premenopausal patients".)

TERMINOLOGY — Terminology used to describe endometrial ablation varies. In this topic review, the following terms will be used:

●Endometrial ablation – Surgical destruction of the endometrium. The endometrium must be destroyed or resected to the level of the basalis (figure 1), which is approximately 4 to 6 mm deep, depending upon the stage of the menstrual cycle. In this topic review, endometrial ablation will refer to either non-resectoscopic ablation or resectoscopic ablation, or resection of the endometrium.

●Non-resectoscopic endometrial ablation – Endometrial ablation is performed with a disposable device which is inserted into the uterine cavity and delivers energy to uniformly destroy the uterine lining. Non-resectoscopic technologies approved and available for use in the United States by the US Food and Drug Administration (FDA) are bipolar radiofrequency (NovaSure), cryotherapy (Her Option, Cerene), circulating hot water (Hydro ThermAblator), and a combined thermal and bipolar radiofrequency (Minerva). These techniques are also referred to as global or second-generation ablation.

Other non-resectoscopic technologies, including the hot liquid-filled balloon (ThermaChoice) and microwave (Microwave Endometrial Ablation), are not clinically available in the United States.

●Resectoscopic endometrial ablation – Endometrial ablation or resection performed under hysteroscopic visualization with resectoscopic electrosurgical instruments (eg, rollerball, wire loop, vaporizing electrode) or with laser. Endometrial resection is often referred to as transcervical resection of the endometrium. Resectoscopic ablation methods are also referred to as standard or first-generation ablation. The term hysteroscopic ablation is also sometimes used for these methods; however, this designation would also include the circulating hot water non-resectoscopic technique, since it involves instilling water through a hysteroscope.

PATIENT SELECTION

Indications — The primary indication for endometrial ablation is treatment of ovulatory menorrhagia in premenopausal patients. Ablation in patients with anovulatory bleeding or postmenopausal patients is discussed below. (See 'Risk factors for endometrial cancer' below.)

Ablation is usually used to treat patients with chronic menorrhagia. It may also be used for acute abnormal uterine bleeding in hemodynamically stable patients in whom medical therapy is contraindicated or unsuccessful. (See "Managing an episode of acute uterine bleeding", section on 'Surgical management'.)

Endometrial ablation is not indicated for patients who wish to be amenorrheic for nonmedical reasons; in fact, many patients do not develop amenorrhea after ablation [3]. (See 'Improvement in bleeding symptoms' below and "Hormonal contraception for suppression of menstruation".)

Contraindications — Endometrial ablation is contraindicated in patients with the following conditions or characteristics [3,4]:

●Pregnancy

●Known or suspected complex or atypical endometrial hyperplasia or cancer

●Desire to preserve fertility

●Active pelvic infection

●Intrauterine device (IUD) in place

●Previous transmyometrial uterine surgery

Endometrial ablation is also relatively contraindicated in patients who are postmenopausal, have congenital uterine anomalies (eg, bicornuate uterus), have a uterine cavity length that is greater than 10 to 12 cm, or have severe myometrial thinning.

Endometrial hyperplasia or cancer — Heavy menstrual bleeding (HMB) is an indication for endometrial ablation but may also signify the presence of complex or atypical endometrial hyperplasia or cancer. Therefore, endometrial sampling should be performed in all patients prior to undergoing endometrial ablation. (See 'Endometrial sampling' below.)

Endometrial ablation is not a treatment for endometrial hyperplasia or cancer and may interfere with subsequent evaluation of the endometrium (see 'Risk factors for endometrial cancer' below). Thus, endometrial ablation is not appropriate in patients with these conditions. This position is consistent with the recommendation of the American College of Obstetricians and Gynecologists (ACOG) that patients with endometrial hyperplasia or uterine cancer should not undergo endometrial ablation [4].

Desire to preserve fertility — Pregnancy is contraindicated following endometrial ablation since the destruction of the endometrium may result in abnormal implantation or placentation [5,6]. (See 'Pregnancy after ablation' below.)

Endometrial ablation does not prevent pregnancy. Patients must be counseled that contraception will be necessary following the procedure. A systematic review of 274 pregnancies after endometrial ablation found that 80 to 90 percent of pregnancies occurred in patients who did not use contraception [7]. Some patients choose to undergo concomitant ablation and sterilization. (See 'Sterilization' below.)

Uterine anomalies — Endometrial ablation should generally be avoided in patients with congenital anomalies of the uterus (eg, septum, unicornuate uterus). There are case reports of non-resectoscopic ablation in patients with a bicornuate uterus, but this is not standard practice [8,9]. Some surgeons with advanced hysteroscopic skills may be able to safely and successfully perform ablation in patients with these abnormalities using resectoscopic techniques or with non-resectoscopic devices such as the hydrothermablator.

Acute retro- or anteflexion or retro- or anteversion of the uterus are not contraindications to endometrial ablation. In patients with these uterine characteristics, incomplete ablation may result if the fundus cannot be reached with a non-resectoscopic ablation device. Thus, resectoscopic ablation may be preferable in this population.

Myometrial thinning following uterine surgery — Endometrial ablation in patients with thinning of the myometrium may result in injury to the adjacent viscera (bladder, bowel) if the energy used for ablation passes through the uterine tissue [4,10]. There are no data to establish the minimum myometrial thickness required to prevent such complications. Because non-resectoscopic ablation typically ablates 4 to 6 mm into tissue, our practice is to avoid performing non-resectoscopic ablation in cases where there is known thin tissue, such as when an isthmocele has been identified in the lower uterine segment.

Attenuation of the myometrium is usually due to a prior uterine surgery (eg, cesarean delivery, myomectomy). There is no evidence that prior cesarean delivery increases the risk of complications associated with endometrial ablation [11]. As an example, one retrospective cohort study found no difference in operative complication or efficacy in patients who had one or more cesarean deliveries compared with other patients [12]. Myometrial thickness was not measured.

When microwave ablation was first introduced in Europe, there were reports of thermal injury to adjacent structures in patients with no uterine perforation [13]. For this reason, instructions for use of the manufacturer of the Microwave Endometrial Ablation system advise a minimum myometrial thickness of 10 mm as documented by ultrasound. While measurement of myometrial thickness is not a requirement for the other four non-resectoscopic ablation devices, surgeons should use clinical judgment to assess the risk of thermal injury when offering non-resectoscopic endometrial ablation to a patient with a history of cesarean delivery or transmural myomectomy. Evaluation of myometrial thickness is discussed below. (See 'Uterine evaluation' below.)

Postmenopausal patients — We suggest not performing endometrial ablation in postmenopausal patients. Endometrial ablation has been described in this population, mostly in patients with persistent bleeding while on postmenopausal hormone therapy. Prospective series have reported no cases of endometrial cancer in patients who underwent endometrial ablation after menopause; however, these studies included fewer than 50 subjects and most patients were followed for less than five years [14].

Excluding malignancy is the primary objective in the diagnostic evaluation of patients with postmenopausal bleeding. As discussed below, diagnosis of endometrial cancer may be more difficult following endometrial ablation (see 'Risk factors for endometrial cancer' below). Bleeding in postmenopausal patients due to benign causes is often managed expectantly or the specific etiology is treated (eg, removal of an endometrial polyp, use of a continuous rather than a cyclic regimen for hormone therapy). Given these other options, it does not appear that the benefit of endometrial ablation outweighs the risk of a missed diagnosis of endometrial cancer in postmenopausal patients. (See "Approach to the patient with postmenopausal uterine bleeding".)

United States professional organizations also do not support the use of endometrial ablation in postmenopausal patients. The American Society for Reproductive Medicine advised that endometrial ablation is not indicated for postmenopausal patients and ACOG states that it has not been rigorously studied in this population [4,15].

Special considerations — In patients with certain characteristics, endometrial ablation may not be as safe or effective as other therapeutic options. These include patients with risk factors for endometrial cancer or with certain uterine characteristics or pathology (eg, submucosal leiomyomas, adenomyosis, large cavity size).

Risk factors for endometrial cancer — The safety of endometrial ablation has not been well studied in patients who do not currently have endometrial neoplasia, but are at an increased risk of developing endometrial cancer (table 1). The concern is whether endometrial scarring caused by ablation would obscure symptoms and decrease the efficacy of endometrial sampling. Ablation itself does not appear to increase the risk of subsequent endometrial cancer [16,17]. (See "Endometrial carcinoma: Epidemiology, risk factors, and prevention", section on 'Risk factors'.)

Endometrial ablation introduces obstacles to the detection and diagnosis of endometrial neoplasia. Ablation results in intrauterine adhesions that may prevent or delay the bleeding that is the cardinal symptom of endometrial cancer and/or interfere with endometrial sampling [17-31]. In addition, histologic evaluation of endometrial samples from patients who have undergone ablation may be difficult [29-33]. There are no data regarding the sonographic appearance of the endometrium in postmenopausal patients following endometrial ablation. (See "Overview of the evaluation of the endometrium for malignant or premalignant disease".)

There is no evidence of an increased risk of endometrial cancer after ablation. There are few data to address this issue, and the best available data are from a population-based study of 1997 to 2014 nationwide data from Finland that compared 5484 patients who underwent endometrial ablation with controls; mean follow-up was 7.3 years [34]. The age range was 30 to 49 years and mean age at ablation was 42.4±4.4 years. In the ablation group, only three patients developed endometrial cancer (0.05 percent) and the incidence was lower than among controls (standardized incidence ratio 0.56, 95% CI 0.12-1.64). Two of the cases were early stage and one was unknown stage. The overall rate of endometrial hyperplasia was not reported, but among the 1086 patients who underwent hysterectomy after ablation, hyperplasia was the indication in 20 (1.8 percent). These patients were likely selected according to usual indications and contraindications for the procedure, and thus this does not change the advice to avoid ablation in patients with risk factors for endometrial cancer. It is somewhat reassuring that the endometrial cancers were detected at an early stage, but there were so few events that more study is needed regarding whether ablation results in a delay in diagnosis.

Many of the other reports of endometrial cancer after endometrial ablation have been in patients with chronic anovulation and/or endometrial hyperplasia [18,20,35]. Patients with these conditions are also often obese or have diabetes and/or hypertension and thus may have contraindications to some medical therapies (eg, steroid hormones, antifibrinolytics) and be at increased risk of complications if they undergo major surgery. We suggest that progestin supplementation be offered to prevent hyperplasia until ovulation is reestablished or menopause occurs in the presence of chronic anovulatory states.

Correction of anovulation is the optimal approach to treatment for these patients. In patients in whom this is not possible, other treatment options (eg, progestins) may be preferable to endometrial ablation. (See "Abnormal uterine bleeding: Management in premenopausal patients".)

There are few data regarding endometrial ablation in patients taking tamoxifen, and there are no studies with long-term follow-up [36-38]. We suggest not performing endometrial ablation in patients who are taking tamoxifen. Such patients are at an increased risk of endometrial cancer, and ablation may interfere with the later diagnosis of this disorder. In addition, endometrial ablation is not necessary for tamoxifen-induced bleeding, which will resolve after the cessation of tamoxifen therapy. (See "Abnormal uterine bleeding and uterine pathology in patients on tamoxifen therapy".)

Patients with Lynch syndrome (hereditary nonpolyposis colorectal cancer) are at a markedly increased risk of endometrial cancer and are advised to undergo hysterectomy after they have completed childbearing. Thus, abnormal uterine bleeding in this population should not be treated with endometrial ablation. (See "Lynch syndrome (hereditary nonpolyposis colorectal cancer): Screening and prevention of endometrial and ovarian cancer".)

In the absence of high-quality data, the decision to perform endometrial ablation in patients with risk factors for endometrial cancer depends mainly upon three factors: the degree to which endometrial ablation may limit future diagnosis of malignancy, the magnitude of the risk of endometrial cancer, and the feasibility of other therapeutic options.

Obesity — There are few data about endometrial ablation in patients with obesity. In one study, 968 patients underwent endometrial ablation for abnormal uterine bleeding (AUB) with an etiology that was endometrial (AUB-E) or ovulatory dysfunction (AUB-O) (figure 2) [39]. There was no difference in treatment failure or the need for a subsequent gynecologic procedure between groups during three-year follow-up with the exception of patients with obesity, with higher failure rates in the AUB-E group (16.7 versus 9.8 percent).

Leiomyomas or polyps — Endometrial ablation in patients with intracavitary (submucosal or intramural myomas that protrude into the uterine cavity) fibroids is controversial. Fibroids are a common etiology of abnormal uterine bleeding. Non-resectoscopic ablation is a technically less challenging procedure than hysteroscopic myomectomy, and surgeons may desire to perform ablation for patients with an isolated intracavitary fibroid. However, there is a concern that the presence of a lesion protruding into the uterine cavity may interfere with the function of some ablation devices.

Endometrial ablation does not treat the fibroid itself; ablation procedures do not remove or shrink fibroids. One study found no histologic effect on leiomyomas following bipolar radiofrequency endometrial ablation [40].

Non-resectoscopic techniques are not explicitly approved by the FDA for use with submucosal fibroids, although this patient population was included in the approval studies for some of the devices [41,42]. Thus, when performing non-resectoscopic ablation in patients with intracavitary fibroids, the surgeon should obtain informed consent regarding off-label use of the device.

Comparative data show that non-resectoscopic devices are less effective in patients with intracavitary fibroids than in those with a normal uterine cavity [43-46]. As an example, in a retrospective cohort study of patients who underwent Hydro ThermAblator procedures, patients with submucosal fibroids (1 to 4 cm) compared with controls had significantly higher rates of menorrhagia (12 versus 3 percent) and hysterectomy (12 versus 1 percent) at an average of 5.8 years postoperatively [44].

By contrast, a prospective series of patients with fibroids up to 3 cm who underwent bipolar radiofrequency ablation reported that, at one-year follow-up, 95 percent had improvement in uterine bleeding [47]. In addition, hot liquid filled balloon ablation was used successfully in 45 patients with submucosal fibroids ≤3 cm in a randomized trial comparing balloon versus rollerball ablation [4,48].

Further study is needed to establish which of the non-resectoscopic devices can be used in patients with intracavitary fibroids and appropriate criteria for such use.

Resectoscopic endometrial ablation can usually be performed in patients with small intracavitary fibroids, or alternatively, a concomitant hysteroscopic myomectomy may be appropriate. (See 'Hysteroscopic myomectomy' below.)

Endometrial polyps are typically removed prior to endometrial ablation. However, patients with endometrial polyps were included in the FDA approval studies for the Microwave Endometrial Ablation device, NovaSure, and ThermaChoice hot liquid filled balloon [42,49,50]. (See "Endometrial polyps".)

Cavity size — In patients with a large uterus on pelvic examination (>10 weeks size), it is prudent to sound the uterus to assess cavity depth prior to planning an endometrial ablation. Each non-resectoscopic endometrial ablation device can accommodate a different minimum (4 to 6 cm) and maximum (10 to 12 cm) uterine cavity size. The cavity sizes for each device are listed in the table. Ablation may be incomplete in patients with a cavity larger than the maximum.

Cavity size is less important for resectoscopic ablation, but operative duration, irrigation fluid absorption, and potential blood loss make these methods impractical in patients with a uterus larger than 12 weeks size. Successful improvement of uterine bleeding symptoms has been reported following partial ablation with resectoscopic ablation or resection, but not for non-resectoscopic techniques [51,52].

Adenomyosis — Adenomyosis (uterine adenomyomatosis) is not a contraindication to endometrial ablation, although patients with this condition may have a lower treatment success rate. A retrospective study of 816 patients who were followed for five years after endometrial ablation reported that patients with a preoperative ultrasound suggestive of adenomyosis had a 1.7-fold increased risk of subsequent hysterectomy or repeat endometrial ablation [53].

Parity — Nulliparity is not a contraindication to endometrial ablation. Grand multiparity appears to be a risk factor for treatment failure. In the retrospective study of 816 patients described in the preceding paragraph, patients with a parity of ≥5 compared with <5 had a sixfold increased risk of subsequent repeat endometrial ablation or hysterectomy [53].

Bleeding disorders or anticoagulation — Coagulopathy due to bleeding disorders or anticoagulant medications is a potential etiology of HMB. Most methods of endometrial ablation are appropriate for use in such patients. A retrospective cohort study compared 41 patients with coagulopathy to 111 patients with normal coagulation, all of whom underwent non-resectoscopic endometrial ablation [54]. Perioperative complications were infrequent in both groups; hematometra occurred in one patient with coagulopathy and in none of the controls. An average of two years after the procedure, rates of subsequent hysterectomy were similar for patients with and without coagulopathy (5 and 7 percent).

Endometrial resection should be avoided in patients who have a bleeding disorder or who are taking anticoagulants, and a rollerball ablation or a non-resectoscopic ablation should be considered. In our practice, we do not stop anticoagulation in patients undergoing endometrial ablation as the risk of bleeding is very low. We order an international normalized ratio (INR) on patients taking warfarin approximately one week prior to surgery and then the day of surgery to be sure they are not super-therapeutic. We prefer patients to have an INR under 4.0.

PREOPERATIVE EVALUATION

Informed consent — Patients who are considering endometrial ablation should be counseled about other medical, interventional radiologic, and surgical options for treatment (see "Abnormal uterine bleeding: Management in premenopausal patients"). It is important to advise patients that a successful result is most likely to be a reduction in the volume of uterine bleeding, and amenorrhea is not guaranteed. In addition, endometrial ablation does not regulate bleeding in patients with irregular patterns. The risks of persistent or recurrent heavy uterine bleeding and of surgical complications should be reviewed. This discussion should be documented on the surgical consent form and in the medical record. (See 'Outcome' below and 'Complications' below.)

Evaluation — Preoperative evaluation, including perioperative risk assessment, is the same for endometrial ablation as for hysteroscopy. Pregnancy testing is performed; cervical cultures are appropriate if cervicitis is suspected. (See "Overview of hysteroscopy", section on 'Preoperative evaluation and preparation'.)

Uterine evaluation

Endometrial sampling — Endometrial sampling is performed in all patients prior to endometrial ablation to exclude endometrial hyperplasia or cancer. Ideally, this should be performed with enough time to receive the results and cancel the procedure, if neoplasia is found. However, if sampling has not yet been performed by the day of the procedure, it should be done just prior to the ablation.

Assessment of the uterus — The uterine cavity should be assessed for the presence of intracavitary myomas, endometrial polyps, or other abnormalities (eg, uterine septum) that may interfere with endometrial ablation (see 'Leiomyomas or polyps' above and 'Uterine anomalies' above). Some lesions can be removed hysteroscopically on the day of the procedure prior to performing the ablation. Patients with a history of transmural uterine surgery (eg, cesarean delivery, myomectomy) should be evaluated for myometrial thinning. (See 'Myometrial thinning following uterine surgery' above.)

We use saline infusion sonography (SIS) or office hysteroscopy to assess the uterine cavity. TVUS alone can identify uterine lesions, but does not define the contour of the cavity. In patients in whom myometrial thinning is suspected, we use SIS. Alternatively, office hysteroscopy may be combined with transvaginal ultrasound (TVUS).

In patients with a large uterus on pelvic examination (>10 weeks size), it is prudent to sound the uterus to assess cavity depth prior to planning an endometrial ablation. (See 'Cavity size' above.)

PREOPERATIVE PREPARATION — Cervical preparation is the same as for hysteroscopy. If an intrauterine contraceptive device is in place, it should be removed. (See "Overview of hysteroscopy", section on 'Preoperative evaluation and preparation'.)

Endometrial preparation — Endometrial pretreatment with hormonal agents is used by most surgeons prior to resectoscopic ablation (with the exception of endomyometrial resection) and is also advised by the manufacturers of most non-resectoscopic ablation devices [41,42,50,55]. The goal of endometrial preparation is to thin the endometrium to facilitate tissue destruction.

Hormonal suppression with a gonadotropin-releasing hormone (GnRH) agonist (eg, intramuscular leuprolide 3.75 mg/month) is the most studied method of endometrial preparation prior to endometrial ablation. Hormonal pretreatment should be initiated 30 to 60 days prior to the procedure [15]. However, due to the expense, side effects, and delay of surgery, we choose to use ablation techniques that do not require GnRH agonist preoperative thinning (eg, radiofrequency ablation).

The effectiveness of other hormonal agents for endometrial preparation prior to endometrial ablation is uncertain. Use of progestins (eg, oral medroxyprogesterone acetate [MPA] 15 mg daily) instead of GnRH agonists offers the advantage of fewer adverse effects (eg, menopausal symptoms). Randomized trials comparing progestins or danazol with GnRH agonists have found no difference in postablation improvement in uterine bleeding; however, these trials lacked statistical power [56,57]. We prefer not to use danazol, since it is poorly tolerated by most patients.

There are few data regarding endometrial preparation with estrogen-progestin contraceptives or GnRH antagonists. Nonhormonal methods of preparation include: performing the procedure during the follicular phase of the menstrual cycle and uterine curettage [58].

The benefits of hormonal suppression are well established for resectoscopic endometrial ablation, but not for non-resectoscopic ablation techniques. In a meta-analysis of 12 randomized trials of mostly resectoscopic ablation or resection (one trial used hot liquid filled balloon ablation), patients in the GnRH agonist pretreatment group compared with placebo or no pretreatment had a shorter operative duration (an average of 4.8 minutes shorter) [3]. At one year after surgery, patients in the GnRH agonist groups versus the no treatment groups had significantly lower rates of moderate or heavy menstrual bleeding (RR 0.7, 95% CI 0.6-0.9) and higher rates of amenorrhea (RR 1.6, 95% CI 1.2-2.0).

Prior to non-resectoscopic endometrial ablation, current practice regarding endometrial preparation is based on the device manufacturers' instructions. Pretreatment with GnRH agonists is advised by the manufacturers for most devices, with the following exceptions. Bipolar radiofrequency ablation (NovaSure) does not require pretreatment since the device adjusts to differing tissue depths [59]. The combined thermal and bipolar radiofrequency device (Minerva) does not require that endometrium be thinned prior to the procedure, and the procedure may be performed during either the proliferative or secretory phase of the cycle. The cryoablation device (Cerene) allows for the use of medications to thin the endometrial lining; alternatively, the procedure may be performed in the early proliferative phase of the menstrual cycle. (See "Endometrial ablation: Non-resectoscopic techniques".)

Methods that do not require hormonal suppression offer the advantages of avoiding the one- to two-month pretreatment time and drug-related adverse effects (eg, menopausal symptoms). Randomized trials and observational data suggest that hot liquid balloon and microwave ablation may be performed successfully using alternative methods (eg, curettage or timing during the follicular phase of the menstrual cycle) [60-63]. As an example, five-year follow-up from a randomized trial of microwave endometrial ablation showed that patients in whom the procedure was scheduled after menses compared with those who had preprocedure hormonal suppression had no significant differences in rates of amenorrhea (84 and 87 percent) or hysterectomy (11 and 6 percent) [63]. Further study of this issue is needed.

OPERATIVE SETUP

Antibiotic prophylaxis — Antibiotics are not routinely administered prior to endometrial ablation for prevention of surgical site infection or endocarditis since postablation infection is rare. (See "Overview of hysteroscopy", section on 'Preoperative evaluation and preparation'.)

Anesthesia — Resectoscopic ablation is usually performed with regional or general anesthesia. While there are reports of performing this type of ablation using local anesthesia, in our clinical experience, it is not well tolerated.

Non-resectoscopic endometrial ablation can be performed using local, regional, or general anesthesia [15,64,65]. For procedures under local anesthesia, some surgeons also offer patients oral or intravenous conscious sedation (see "Procedural sedation in adults outside the operating room"). General principles of anesthesia for endometrial ablation are the same as for operative hysteroscopy procedures. (See "Overview of hysteroscopy", section on 'Pain management'.)

Administration of an oral nonsteroidal anti-inflammatory drug (NSAID) at least one hour preoperatively is recommended by most non-resectoscopic device manufacturers [15,59,66-69]. NSAIDs may be useful to inhibit uterine contractions; however, regarding pain control in patients receiving local anesthesia, randomized trial data show that NSAIDs given prior to operative hysteroscopy reduce postoperative, but not intraoperative, pain. (See "Overview of hysteroscopy", section on 'Pain management'.)

Local anesthesia with sedation was compared with general anesthesia in a randomized trial of 191 patients undergoing microwave endometrial ablation [70]. Nine percent of procedures that started under local anesthesia were converted to general anesthesia due to patient discomfort. There were no significant differences between anesthetic groups regarding postoperative nausea, use of opiates, or recovery time. While tolerability was high overall, significantly fewer patients who were assigned to local versus general anesthesia reported that the procedure was generally or totally acceptable (85 versus 97 percent).

Office versus operating room procedures — Resectoscopic endometrial ablation is performed in the operating room, since it typically requires regional or general anesthesia and close monitoring of irrigation fluid absorption. For laser techniques, the expense and size of the equipment preclude office use.

Non-resectoscopic endometrial ablation has become an accepted office-based procedure [71,72]. Endometrial ablation performed in the office offers the advantages of a familiar setting for the patient, easier scheduling, and lower procedural costs. The disadvantages of the office setting are the inability to switch to general anesthesia if the patient is not tolerating the procedure and the need to transfer the patient to an operating room if complications occur.

In order to be candidates for office surgery, patients should be at low risk for surgical or anesthesia complications. Physicians who offer ablation in the office need to be trained in administering and responding to complications of conscious sedation; assisting personnel must also be trained appropriately [73]. Emergency equipment must be available (eg, updated emergency cart, pulse oximeter, oxygen, wall suction) and protocols to handle rare complications such as excessive bleeding, uterine perforation, allergic reactions, or severe vagal reactions. (See "Procedural sedation in adults outside the operating room".)

Concomitant procedures

Hysteroscopic myomectomy — In patients who have a submucosal leiomyoma, some surgeons perform a hysteroscopic myomectomy followed by endometrial ablation [74,75]. There are no safety data on the use of non-resectoscopic techniques in patients having a concurrent hysteroscopic myomectomy. In fact, prior uterine surgery is considered a contraindication for certain non-resectoscopic devices. If myomectomy is performed followed by ablation, we use morcellation or endometrial resection techniques. Moreover, it is uncertain whether this procedure is more effective at improving uterine bleeding symptoms than myomectomy alone.

Based on the available data, it appears that a combined procedure neither increases nor decreases the rate of subsequent hysterectomy [76,77]. The only comparative study that evaluated myomectomy alone versus in combination with ablation was a retrospective cohort study of 177 patients with submucosal fibroids and abnormal uterine bleeding; the combination of hysteroscopic myomectomy with endometrial ablation (rollerball or rollerbarrel method) or resection was associated with a significantly greater improvement in bleeding symptoms than myomectomy alone (96 versus 81 percent) [77]. Further study is needed to establish whether a combined procedure impacts control of bleeding symptoms or operative complications.

If a combined procedure is performed, it is prudent to perform the myomectomy followed by ablation or resection. Available studies have used resectoscopic methods, such as rollerball ablation or endometrial resection, rather than the non-resectoscopic ablation methods that are more commonly used [78]. However, resectoscopic methods may be preferable in this context. This is because they are performed under direct visualization, and the operator can avoid desiccating or resecting the endometrium or exposed myometrium that were attached to the myoma. This may avoid the spread of energy through areas of thin myometrium. In addition, non-resectoscopic devices have not been approved by the United States Food and Drug Administration for concomitant use with hysteroscopic myomectomy; the manufacturer of the microwave endometrial ablation system, though not clinically available in the United States, states that surgery that thins the myometrium should not be performed concurrently with ablation [67]. Hysteroscopic myomectomy is discussed separately. (See "Uterine fibroids (leiomyomas): Hysteroscopic myomectomy".)

Sterilization — Concomitant endometrial ablation and hysteroscopic sterilization had been performed when hysteroscopic sterilization devices were available. The primary manufacturer of these devices stopped production in 2018. (See "Hysteroscopic female permanent contraception".)

Laparoscopic tubal occlusion can be performed concurrent with the ablation procedure, but there appears to be an increased risk of symptomatic cornual hematometra. (See 'Postablation tubal sterilization syndrome' below.)

Intrauterine device insertion — There are few data regarding the placement of an intrauterine device (IUD) concomitant with or following endometrial ablation, thus the safety of this combination of procedures is not well established. Decisions regarding use of an IUD following endometrial ablation should be individualized.

Transcervical resection of the endometrium (TCRE) with concomitant insertion of a 52 mg levonorgestrel (LNG) IUD appears to be safe and to result in a high rate of amenorrhea, based upon several small observational studies [79-81]. The largest study was a retrospective cohort study of 53 patients who underwent insertion of an LNG IUD on completion of endometrial resection and confirmation of adenomyosis; a control group of 42 patients underwent endometrial resection only [81]. All patients received a prophylactic antibiotic postoperatively. At one-year follow-up, patients in the LNG IUD group had higher rates of amenorrhea (100 versus 9 percent) and resolution of dysmenorrhea (90 versus 20 percent). There were no complications in the IUD group that resulted in removal of the device. None of the patients in the LNG IUD group had a subsequent procedure to control uterine bleeding compared with eight patients in the control group.

Radiofrequency endometrial ablation and concomitant insertion of the 52 mg LNG IUD appears to result in higher amenorrhea rates, but data are limited. In a retrospective study of 246 patients who underwent either radiofrequency ablation plus insertion of the 52 mg LNG IUD or radiofrequency ablation alone, both the rates of amenorrhea and remission of dysmenorrhea were significantly higher in the IUD group [82]. Amenorrhea rates at 6, 12, and 24 months were 78 versus 46 percent, 85 versus 53 percent, and 87 versus 58 percent for patients with and without an IUD, respectively. The rates of dysmenorrhea remission were 100 versus 70 percent, 100 versus 64 percent, and 100 versus 64 percent.

COMPLICATIONS — The most common complications associated with endometrial ablation are uterine perforation, hemorrhage, hematometra, and pelvic infection. Rates of complications of endometrial ablation reported in a meta-analysis of 21 randomized trials are listed in the table (table 2) and discussed below for each type of complication [83]. Other complications of endometrial ablation that are the same as those for operative hysteroscopy are discussed separately. (See "Overview of hysteroscopy", section on 'Complications'.)

Major complications are rare. In the United States, the FDA has a voluntary reporting system for non-resectoscopic ablation complications. From 1990 to 2011, bowel injury (128 cases reported from 2005 to 2011) was the complication most commonly reported to the Manufacturer and User Facility Device (MAUDE) database [84-86]. Circulating hot water ablation, in particular, was associated with burns to the vagina and perineum. Other major complications that were rarely reported included: urinary tract injury, perioperative hysterectomy, carbon dioxide embolism, necrotizing fasciitis, and death. It is not possible to calculate the rate of complications from these data, since the total number of procedures performed is not known.

Uterine perforation — Uterine perforation has been reported in 0.3 percent of non-resectoscopic endometrial ablation procedures and 1.3 percent of resectoscopic ablations or resections [87]. The evaluation and management of uterine perforation are discussed separately. (See "Uterine perforation during gynecologic procedures".)

Hemorrhage — Perioperative hemorrhage has been reported in 1.2 percent of patients undergoing non-resectoscopic ablation and 3.0 percent of those undergoing resectoscopic ablation [87]. Evaluation and management of hemorrhage are the same as for hysteroscopic procedures. (See "Overview of hysteroscopy", section on 'Hemorrhage'.)

Hematometra — Intrauterine scarring is an expected result of endometrial ablation. When areas of endometrium are adherent and there is endometrial bleeding behind the occlusion, hematometra will occur. The reported rate of hematometra in non-resectoscopic ablation is 0.9 percent and is 2.4 percent in resectoscopic ablation [87].

Hematometra should be suspected in a patient with a history of an endometrial ablation who presents with amenorrhea and cyclic pain, even remote from the procedure [88]. This was supported by a study that found an association between postablation pain and hematometra; hematometra was present in 5 of 19 patients who underwent postablation hysterectomy with pain as the indication [89]. Hematometra can be diagnosed with the visualization of pockets of echogenic fluid in the uterine cavity on transvaginal ultrasound.

Uterine cavity occlusion is likely to occur in the cervical canal or at the most narrow portion of the endometrial cavity, the lower uterine segment. Thus, ablation of the cervix or cervico-uterine junction should be avoided. Complete ablation of the uterine fundus, cornua, and tubal ostia will also help avoid accumulation of blood from active endometrium.

Cervical stenosis is treated with cervical dilation. Adhesions in the mid- to upper uterine cavity can be lysed under hysteroscopic guidance. For persistent pain caused by hematometra despite minimally invasive treatment, hysterectomy may be indicated. (See "Intrauterine adhesions: Clinical manifestation and diagnosis".)

Postablation tubal sterilization syndrome — Some patients who have undergone tubal ligation prior to endometrial ablation experience cyclic or intermittent pelvic pain. The proposed etiologies of this are: (1) bleeding from active endometrium that is trapped in the uterine cornua and/or (2) uterine contracture and intrauterine scarring. The incidence of this complication is as high as 10 percent in some reports [90]. Prevention, diagnosis, and management of these symptoms are the same as for other presentations of hematometra. In addition, some surgeons assess patients with these symptoms laparoscopically and excise the tubal stumps to prevent the distension of the proximal tubal segments during menses.

Postablation tubal sterilization syndrome (PATSS) was initially reported in 1993 as patients presenting with unilateral or bilateral pelvic pain and vaginal spotting who had previously undergone tubal sterilization and endometrial ablation [91]. The patients had a common theme of endometrial cavity scarring with one or both swollen proximal fallopian tubes. The incidence of PATSS is estimated to be between 6 to 8 percent, and usually develops two to three years after endometrial ablation [92].

The diagnosis of PATSS is initially suspected clinically in patients with cyclic cramping with or without menses with a history of endometrial ablation and tubal sterilization. Ultrasound has not been reliably sensitive at diagnosing PATSS. The confirmatory diagnosis is made surgically; however, magnetic resonance imaging (MRI) during times of symptomatic cramping may be useful using T2 weighted images looking for corneal blood. The definitive treatment of PATSS is hysterectomy [93]. Hysteroscopic lysis of adhesions is possible but difficult to perform in the cornual regions, and is not reliably effective.

Central hematometra is most likely to occur when the cervical canal is damaged at the time of endometrial ablation and has an incidence of 1 to 3 percent [94]. The typical presentation is cyclic pelvic pain. Central hematometra can usually be seen on imaging (ultrasound or MRI) at the time of cyclic pain and can usually be successfully treated with cervical dilatation. In some cases, hysteroscopic adhesiolysis may be necessary.

Pelvic infection — Fever occurs in the immediate postoperative period in approximately 1 percent of patients who have undergone endometrial ablation [87]. In the meta-analysis, the incidence of infectious complications included: endometritis (1.4 to 2.0 percent); myometritis (0 to 0.9 percent); pelvic inflammatory disease (1.1 percent); and pelvic abscess (0 to 1.1 percent) [87]. Evaluation and management of these infections are discussed separately. (See "Endometritis unrelated to pregnancy" and "Pelvic inflammatory disease: Treatment in adults and adolescents" and "Epidemiology, clinical manifestations, and diagnosis of tubo-ovarian abscess".)

OUTCOME — The overall rate of subsequent surgery following all types of endometrial ablation ranges from 17 to 25 percent [34,87,95]. The outcomes for non-resectoscopic and resectoscopic endometrial ablation are discussed here. Separate data for each of these types of ablation are given, where available. Outcomes for specific procedures (eg, rollerball ablation, cryoablation) are discussed separately. (See "Endometrial ablation: Non-resectoscopic techniques" and "Endometrial ablation or resection: Resectoscopic techniques".)

Non-resectoscopic versus resectoscopic ablation — In our practice, we perform resectoscopic ablation if the shape or size of the uterine cavity will not accommodate a non-resectoscopic device, if a concomitant procedure such as submucous fibroid resection is planned, or if the patient has had multiple cesarean deliveries to avoid ablating over the hysterotomy scar.

Non-resectoscopic and resectoscopic endometrial ablation result in comparable rates of amenorrhea and patient satisfaction [87,96]. Resectoscopic endometrial ablation is associated with more frequent use of general anesthesia, a longer operative duration, and an increased risk of some surgical complications (eg, irrigation fluid overload). However, the risk of fluid overload can be significantly reduced if a bipolar resectoscopic device that will allow the use of saline is chosen. Technical skill requirements are another potential barrier to the use of resectoscopic ablation since, currently, many surgeons are less experienced with these techniques. Resectoscopic techniques are less costly per procedure than non-resectoscopic ablation. In 2010, the cost of disposable equipment for resectoscopic procedures was approximately USD $125 to $150, while the range for non-resectoscopic techniques was USD $850 to $1300.

In current practice, non-resectoscopic endometrial ablation is performed in most cases. Resectoscopic ablation is a reasonable option for surgeons with the required hysteroscopic skills. Resectoscopic ablation or therapy other than endometrial ablation may be optimal for patients with acute flexion or version of the uterus that does not allow a non-resectoscopic ablation device to reach the uterine fundus or for patients undergoing repeat endometrial ablation. (See 'Uterine anomalies' above and 'Repeat endometrial ablation' below.)

The differences between non-resectoscopic and resectoscopic ablation were best demonstrated in a meta-analysis of 21 randomized trials (table 2) [87]. Findings are discussed in the sections that follow.

Improvement in bleeding symptoms — Patients may experience irregular bleeding following endometrial ablation, so the success of the procedure cannot be determined until 8 to 12 weeks postoperatively. Most patients with successful ablation will have a reduction in uterine blood flow, but not amenorrhea. There is variability across studies in whether menstrual blood loss was measured and the method used (eg, alkaline hematin method, pictorial blood assessment chart) [97-99]. Thus, amenorrhea, patient satisfaction, and subsequent surgery are more commonly used to assess treatment success.

One example of the available data regarding reduction in menstrual blood flow is a randomized trial in which 279 patients were assigned to Her Option cryoablation or rollerball ablation [100]. At 12-month follow-up, rates of normal menstrual flow (defined as a pictorial blood assessment chart score of <75) were comparable for cryoablation and rollerball (85 and 89 percent).

Regarding amenorrhea, in the meta-analysis of 21 trials, both non-resectoscopic and resectoscopic ablation resulted in similar rates of amenorrhea at one year (37 versus 38 percent, odds ratio [OR] 0.9, 95% CI 0.6-1.4) and two to five years (53 versus 48 percent, OR 1.3, 95% CI 0.6-2.9) [87]. The Minerva device was not included in the meta-analysis, but in a multi-center randomized trial (n = 153), a one-year amenorrhea rate of 71.7 percent was reported compared with 49 percent in patients who underwent rollerball ablation [101].

Patient satisfaction — Patient satisfaction rates were high for both types of ablation at one year (91 versus 88 percent, OR 1.2, 95% CI 0.9-1.7) and two to five years (93 versus 87 percent, OR 1.4, 95% CI 0.6-3.5) [87].

Subsequent surgery — In the meta-analysis of randomized trials, non-resectoscopic and resectoscopic ablation were associated with similar rates of additional surgical treatment for abnormal uterine bleeding at two to five years postoperatively for any type of surgery (including hysterectomy) (21 versus 25 percent, OR 0.9, 95% CI 0.6-1.4) and for hysterectomy alone (14 versus 19 percent, OR 0.8, 95% CI 0.5-1.3) [87].

Similarly, a large retrospective study (n = 114,910) found a significantly lower likelihood of subsequent surgery within five years following radiofrequency compared with resectoscopic ablation (hazard ratio 0.69, 95% CI 0.63-0.76) [95].

Patients who are less than 45 years old at the time of endometrial ablation appear to be at a higher risk of treatment failure than older patients. This was illustrated in two retrospective studies (n = 816 and 3681) of patients followed for five to eight years after endometrial ablation [53,95,102]. The risk of subsequent hysterectomy or repeat ablation was twofold in patients less than 45 years old versus those 45 years or older (54 versus 27 percent in one study [102]).

Complication rates — The following perioperative complications were significantly less likely following non-resectoscopic compared with resectoscopic ablation: irrigation fluid overload (0 versus 0.3 percent, OR 0.2, 95% CI 0.04-0.8), hematometra (0.9 versus 2.4 percent, OR 0.3, 95% CI 0.1-0.9), and cervical laceration (0.2 versus 2.2 percent, OR 0.2, 95% CI 0.1-0.6) [87]. In addition, the rate of uterine perforation was lower, though this was not statistically significant (0.3 versus 1.3 percent, OR 0.3, 95% CI 0.1-1.0).

Other types of complications for which the frequency did not differ significantly between types of ablation included: hemorrhage (1.2 versus 3.0 percent, OR 0.7, 95% CI 0.3-1.9) and endometritis (2.0 versus 1.4 percent, OR 1.3, 95% CI 0.4-3.6).

Anesthesia and technical factors — An advantage of non-resectoscopic compared with resectoscopic ablation is the greater likelihood of using local rather than general anesthesia (61 versus 21 percent, OR 6.4, 95% CI 3.0-13.7) and shorter operative duration (an average of 15 minutes shorter) [87].

Disadvantages of non-resectoscopic compared with resectoscopic ablation were a higher incidence of equipment failure (9.1 versus 1.6 percent), nausea/vomiting (19 versus 7.7 percent), and uterine cramping (38 versus 33 percent).

In addition to the factors listed above, resectoscopic ablation requires more technical skill than non-resectoscopic ablation. It is useful for a gynecologist who performs endometrial ablation to be trained in both types of ablation.

Cost — Non-resectoscopic endometrial ablation devices are generally more expensive than the resectoscopic instruments used for rollerball ablation or wire loop resection. In addition, non-resectoscopic procedures are more commonly performed under local anesthesia and/or in an office setting, which have lower procedural costs. Cost-utility model studies from the United Kingdom have reported that microwave and hot liquid filled balloon compared with resectoscopic ablation are less expensive and accrue more quality adjusted life years [103,104]. It is important to note that cost-utility calculations may vary in different health care settings.

Endometrial ablation versus other treatments

Medical therapy — Endometrial resection was compared with medical therapy in a randomized trial that assigned 187 patients with heavy menstrual bleeding (HMB) to endometrial resection or oral medication (at least three cycles, specific medication varied) [105]. At four-month follow-up, days of heavy bleeding were significantly less in patients treated with endometrial resection versus medication (0.8 versus 3.2 days). At five-year follow-up, the rate of hysterectomy was similar in patients initially randomized to resection or medical therapy (18 and 19 percent); among patients in the original medical therapy group, 10 percent still used medical treatment and 59 percent had surgical treatment other than hysterectomy. Further high-quality data are needed to compare medical therapy with non-resectoscopic endometrial ablation.

Levonorgestrel IUD — The decision to use the levonorgestrel intrauterine device or endometrial ablation depends upon a patient's preferences regarding treatment factors such as plans for fertility and contraception, convenience, and risks of anesthesia. Comparison between these two treatment options is discussed in detail separately. (See "Abnormal uterine bleeding: Management in premenopausal patients", section on 'Levonorgestrel intrauterine device'.)

Hysterectomy — Hysterectomy is the definitive management option for HMB. However, for patients who desire surgical therapy, do not desire future pregnancy, and have a reliable form of contraception, we offer endometrial ablation as an alternative. In general, rates of patient satisfaction are high with both methods [83,106]. While endometrial ablation compared with hysterectomy has a shorter operative time, recovery time, and is associated with a lower risk of sepsis, blood transfusion, and hematoma, the risk of reoperation for recurrent symptoms is higher [106-108].

In a meta-analysis including 10 randomized trials and over 1900 patients comparing endometrial resection and ablation (any type) with hysterectomy (open, minimally invasive, or unspecified) for HMB, patients undergoing an endometrial procedure compared with hysterectomy had a [106]:

●Higher risk of requiring a second surgery for treatment failure within one year (3 to 13 percent versus 0 to 0.4 percent, eight trials). Treatment failures in the hysterectomy group resulted from patients with continued bleeding after supracervical hysterectomy. No patients required reoperation for treatment failure after total hysterectomy (ie, removal of the uterus and cervix).

●Shorter return to normal activity (mean difference [MD] -11 to -21 days, two trials).

●Similar quality of life at two years, but results favored hysterectomy.

●Lower risk of sepsis, blood transfusion, and hematoma when compared with open and unspecified hysterectomy; the risk was similar when compared with minimally invasive hysterectomy.

This meta-analysis reported on short-term outcomes; a higher incidence of reoperation has been reported in large observation studies and in a randomized trial not included in this meta-analysis that followed patients long term. In this randomized trial including 153 patients with abnormal uterine bleeding undergoing either endometrial ablation or laparoscopic supracervical hysterectomy and followed for a median of 14 years, the risk of reoperation for symptom recurrence was 29 percent for patients treated with endometrial ablation compared with none in the laparoscopy group (relative risk [RR] 0.39, 95% CI 1.20-1.61) [108]. (See 'Subsequent surgery' above.)

In general, studies have found no detrimental effects of hysterectomy on sexual function. However, some data suggest that these effects occur more frequently following hysterectomy than endometrial ablation. This was illustrated in a population-based prospective cohort study of over 9500 patients who underwent either endometrial resection or hysterectomy (total or subtotal) [109]. Oophorectomy versus no oophorectomy at time of hysterectomy was associated with a greater adverse effect on sexual function; therefore, data are included only for patients whose ovaries were conserved. At five-year follow-up, patients who underwent hysterectomy without oophorectomy versus endometrial resection had significantly higher age- and hormone therapy-adjusted rates of severe effects on sexual measures: loss of libido (OR 1.4, 95% CI 1.2-1.7), difficult sexual arousal (OR 1.4, 95% CI 1.2-1.6), and vaginal dryness (OR 1.5, 95% CI 1.2-1.9). (See "Hysterectomy: Selection of surgical route (benign indications)".)

FOLLOW-UP

Routine postoperative care — The most common postoperative side effects of endometrial ablation are cramping and vaginal discharge. Light vaginal bleeding or pink-tinged discharge is often present for two to three days. Uterine cramping may persist for 24 to 72 hours. Most patients can resume normal activities in one to three days.

We give patients routine post-hysteroscopy instructions. We counsel patients to call if they have severe or persistent pelvic pain or vaginal bleeding. (See "Patient education: Care after gynecologic surgery (Beyond the Basics)".)

In addition, patients who have undergone endometrial ablation must be counseled about the need for contraception.

Recurrent abnormal uterine bleeding

Prevention — Some investigators have evaluated the use of postablation/resection progestin therapy as a way to decrease ablation failure rates. The definition of success varies, and may include amenorrhea, relief of bothersome bleeding symptoms, and/or patient satisfaction.

Use of a progestin after endometrial resection has been evaluated in a randomized trial and found to increase short-term amenorrhea rates. The trial assigned 71 patients with ovulatory menorrhagia to hysteroscopic endometrial resection combined with medroxyprogesterone acetate (MPA; 150 mg intramuscularly, single dose at the time of surgery) or resection alone [110]. At one-year follow-up, patients in the progestin group had significantly higher rates of amenorrhea (40 versus 26 percent) and lower rates of subsequent ablation/resection or hysterectomy (3 versus 21 percent).

Observational studies have also found that oral and intramuscular medroxyprogesterone following resection was associated with higher rates of amenorrhea and patient satisfaction [111,112].

The intrauterine device (IUD) that releases levonorgestrel 20 mcg/day (Mirena, Liletta; referred to as LNG 52) has been studied in patients following endometrial ablation [113]. In an observational study of 105 patients with a mean follow-up of 25 months, 96 percent were satisfied with the treatment. A prospective study of 43 patients with adenomyosis that compared endometrial resection combined with the LNG 52 IUD versus LNG 52 alone found a significant reduction in menstrual flow in the combined group at 3, 6, and 12 months posttreatment [114]. Endometrial resection with LNG IUD was associated with a significant reduction of number of days bleeding, as well as an improved amenorrhea rate (40 percent) compared with endometrial resection alone (26 percent).

These studies suggest that the addition of progestin therapy may reduce bleeding after endometrial ablation or resection in the short term. Further studies are needed to verify longer term benefits.

Evaluation — Patients who have undergone endometrial ablation may have irregular uterine bleeding or spotting. However, such symptoms should not be presumed to be a normal outcome of ablation. Patients who develop irregular, excessive, or prolonged uterine bleeding following endometrial ablation should undergo imaging and endometrial sampling to evaluate the etiology of the bleeding and exclude endometrial neoplasia.

When patients report heavy uterine bleeding remote from the ablation, the clinician is faced with the concern for the possibility of new pathology, with the challenge of sampling an endometrium which may be difficult to access due to postsurgical adhesions. In a study of 303 patients deemed to require endometrial sampling after endometrial ablation, the failure rate for obtaining bleeding assessment was 40 percent [115].

Repeat endometrial ablation — For some patients in whom abnormal uterine bleeding recurs after ablation, a repeat ablation may be beneficial. Rates of re-ablation of approximately 5 to 20 percent have been reported within three to five years of the initial procedure [105,116-121]. Although repeat non-resectoscopic ablation procedures have been reported, we use only resectoscopic ablation for repeat procedures [120]. This is because the uterine cavity is likely to be distorted and hysteroscopic visualization makes it possible to identify and treat remaining areas of endometrium [27,28]. Complication rates (eg, uterine perforation, excessive fluid absorption, genital tract burns) were higher in repeat compared with primary ablations in a retrospective cohort study [116].

Pregnancy after ablation — Despite destruction of the endometrium, pregnancy occurs in 0.7 percent of patients who have undergone endometrial ablation; conceptions have been reported to occur a median of 1.5 years after ablation, with a range of 3 weeks to 13 years [5,7]. While successful pregnancies have been reported, there appears to be greater risk of complications in pregnancies that follow ablation [122-125]. Data are limited regarding obstetric outcomes; a review of 74 cases reported the following: pregnancy termination (38 percent), ectopic pregnancy (3 percent), first-trimester miscarriage (12 percent), second-trimester pregnancy loss (5 percent), preterm delivery (20 percent), and term delivery (20 percent) [5]. Another literature review of 70 cases reported rates of abnormal placentation (26 percent), malpresentation (39 percent), and perinatal mortality (13 percent); 71 percent of patients underwent cesarean delivery [126]. Further study is needed to evaluate the risk of obstetric complications in postablation pregnancies.

Use of postmenopausal hormone therapy — Patients who have had an endometrial ablation and later choose to use postmenopausal hormone therapy should be given progestins along with estrogen for endometrial cancer protection of any untreated endometrial tissue. (See "Menopausal hormone therapy: Benefits and risks", section on 'Protective effect of progestins'.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Gynecologic surgery".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topic (see "Patient education: Endometrial ablation (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Endometrial ablation is the surgical destruction of the endometrium for the treatment of abnormal uterine bleeding. Resectoscopic ablation is performed under hysteroscopic visualization, using resectoscopic instruments to ablate or resect the endometrium. Non-resectoscopic ablation is performed with a disposable device which is inserted into the uterine cavity and delivers energy to uniformly destroy the uterine lining. (See 'Terminology' above and 'Indications' above.)

●Pregnancy is contraindicated following endometrial ablation, but ablation does not prevent pregnancy. Contraception is still necessary after ablation in sexually active patients. (See 'Desire to preserve fertility' above.)

●Endometrial ablation is not appropriate in patients with complex and atypical endometrial hyperplasia or cancer. Endometrial sampling should be performed in all patients prior to ablation. (See 'Endometrial hyperplasia or cancer' above and 'Endometrial sampling' above.)

●We suggest not performing endometrial ablation in postmenopausal patients (Grade 2C).

●Some, but not all, of the non-resectoscopic ablation devices have been evaluated for use in patients with small submucosal fibroids. It appears that endometrial ablation in these patients is less effective than in patients with a normal uterine cavity. Fibroids are typically removed prior to performing resectoscopic ablation or resection. (See 'Leiomyomas or polyps' above.)

●For patients undergoing endometrial ablation with most non-resectoscopic devices, we suggest endometrial preparation rather than no preparation (Grade 2C). The exceptions are bipolar radiofrequency ablation (NovaSure), the combined thermal and bipolar radiofrequency device (Minerva), and the cryothermal ablation device (Cerene), which do not require endometrial preparation. Either hormonal suppression or uterine curettage can be used for endometrial preparation for hot liquid filled balloon ablation. (See 'Endometrial preparation' above.)

●For patients undergoing most types of resectoscopic endometrial ablation, we recommend endometrial preparation over no endometrial preparation (Grade 1A). Endometrial preparation is not necessary prior to endometrial resection. (See 'Endometrial preparation' above.)

●For endometrial preparation with hormonal suppression prior to endometrial ablation, we suggest gonadotropin-releasing agonists over other agents (Grade 2C). Pretreatment with oral contraceptives is used by some surgeons. (See 'Endometrial preparation' above.)

●The choice of resectoscopic or non-resectoscopic ablation must be individualized to the patient and the surgeon's familiarity with the procedure. Both approaches to endometrial ablation result in comparable rates of amenorrhea and patient satisfaction. Resectoscopic endometrial ablation is associated with more frequent use of general anesthesia, a longer operative duration, and an increased risk of some surgical complications (eg, irrigation fluid overload). Resectoscopic techniques are less costly per procedure and may be optimal for patients with acute flexion or version of the uterus that does not allow a non-resectoscopic ablation device to reach the uterine fundus or for patients undergoing repeat endometrial ablation. (See 'Non-resectoscopic versus resectoscopic ablation' above.)

●Resectoscopic ablation is usually performed with regional or general anesthesia. Non-resectoscopic endometrial ablation can be performed using local, regional, or general anesthesia. (See 'Anesthesia' above.)

●Patients with chronic anovulation often find chronic medical therapy cumbersome due to daily or weekly dosing. In such patients, we suggest the levonorgestrel intrauterine device (IUD) rather than endometrial ablation (Grade 2B). Endometrial ablation is a reasonable choice in patients who do not desire future pregnancy and/or wish to avoid using or changing an IUD. (See 'Levonorgestrel IUD' above.)

●For most patients who desire surgical therapy for abnormal uterine bleeding and who do not desire future pregnancy, we offer both endometrial ablation and hysterectomy. Endometrial ablation has shorter operative and recovery times while hysterectomy is a reasonable option for patients who desire definitive therapy and who are willing to accept the longer surgical and recovery times and possibly an increased risk of sepsis, blood transfusion, and hematoma. (See 'Hysterectomy' above.)

●Patients who have had an endometrial ablation and later choose to use postmenopausal hormone therapy require endometrial protection with progestins. (See 'Use of postmenopausal hormone therapy' above.)