Surgical Alternatives to Hysterectomy for Abnormal Uterine Bleeding

Minnesota Health Technology Advisory Committee

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Minnesota Health Technology Advisory Committee. Minnesota Health Technology Assessments [Internet]. St. Paul (MN): Minnesota Department of Health; 1995-2001.

This publication is provided for historical reference only and the information may be out of date.

This publication is provided for historical reference only and the information may be out of date.

Cover of Minnesota Health Technology Assessments

Minnesota Health Technology Assessments [Internet].

Show details

Contents

< Prev Next >

Surgical Alternatives to Hysterectomy for Abnormal Uterine Bleeding

Created: June 2000.

Executive Summary

The purpose of this report is to present physicians with several surgical alternatives to hysterectomy for women experiencing menorrhagia and dysfunctional uterine bleeding. Approximately 8700 hysterectomies are performed annually in Minnesota.²⁸ Due to its invasive nature, significant morbidity and mortality, long recovery period, and associated health care costs several surgical alternatives to hysterectomy were developed in the 1980s. These surgical techniques are called endometrial ablation. Endometrial ablation is a type of surgical procedure in which the endometrium, the uterine lining, is removed or destroyed by electrosurgery or the application of heat via laser energy or thermal fluids.

The patient selection criteria for endometrial ablation includes: blood loss >80 ml per cycle; bleeding for longer than 8 days; blood loss sufficient to cause anemia; or blood loss or symptoms that interfere with normal activities; drug treatment failed, was contraindicated, was refused; uterine size <12 weeks gestation, and uterine cavity <12 cm in length; all other causes of excessive menstrual bleeding have been excluded including cancer, precancer, and uterine lesions; and childbearing is complete.

Several endometrial ablation techniques have been developed. Transcervical resection of the endometrium (TCRE), involves the destruction of the endometrium with a resectoscope, a telescopic instrument with a wire loop or a rollerball at its tip. An electrical current delivered to the tip of the instrument cuts or coagulates the endometrium. During endometrial laser ablation (ELA), thermal energy produced by a neodymium-yttrium-aluminum-garnet (Nd:YAG) laser destroys the endometrium. TCRE and ELA are guided by hysteroscopy. A newer alternative for endometrial ablation is thermal balloon ablation of the endometrium (TBEA), a procedure that does not require the use of a hysteroscope. A latex balloon is inserted through the vagina into the cervix and is filled with heated fluid that coagulates the endometrium. An advantage of endometrial ablation is that it does not involve the removal of the uterus and thus, the operative and recovery times are shorter, and the complication rates are lower than for hysterectomy. In this evaluation the majority of studies reviewed compared one or more endometrial ablation techniques with hysterectomy, the surgical removal of the uterus.

In the two randomized controlled trials that compared outcomes for menorrhagia patients following endometrial ablation (laser ablation or endometrial resection) or hysterectomy at 1 year and at 4 years, the patient selection criteria included dysfunctional uterine bleeding, candidate for hysterectomy, age <50 years old, uterine size <10 weeks gestation, and no endometrial cancer or precancer. In the first study the length of hospitalization and duration of postoperative pain, and the time to return to normal activities were significantly lower in the endometrial ablation group compared with the hysterectomy group. A higher percentage of patients in the hysterectomy group experienced major complications compared with the ablation group (5.2% versus 1%). The incidence of post-operative infection was significantly lower in the ablation group. At 1 year, 22% of the endometrial ablation patients were amenorrheic, 62% were hypomenorrheic, 3% were unchanged, 11.5% required repeat ablation, and 14% required a hysterectomy for persistent menorrhagia. The patient satisfaction rate was significantly higher in the hysterectomy group, but between 70% and 90% of the women were satisfied with the outcome of endometrial ablation at the 1 year follow-up even though an appreciable proportion of women required a second procedure.⁴⁸ When the patients were followed-up at 4 years (74% of the original group responded), 34% of the endometrial ablation group required additional surgery (ablation and/or hysterectomy) compared with 0% in the hysterectomy group. The percentage of ablation patients requiring hysterectomy increased by 4% to 18%, those requiring repeat ablation increased to 12% from 11.5%, and those requiring both was 3%. A higher percentage of ablation patients were amenorrheic at the later follow-up (45% versus 22%). Satisfaction rates were high for both groups, 80% in the ablation group versus 89% in the hysterectomy group, the difference reflecting retreatment. However, a significantly higher percentage of hysterectomy patients reported having improved general health compared with the ablation group (67% versus 38%).

Cost analyses have demonstrated that endometrial ablation by TCRE has lower procedural costs than hysterectomy. The direct and indirect costs of TCRE were significantly lower than those for hysterectomy. However, the difference in the costs of endometrial ablation and hysterectomy narrow when repeat ablation procedures, or even hysterectomy, are required. Questions remain about whether endometrial ablation lowers the threshold for intervention in women with less severe symptoms, thereby increasing its use and offsetting any cost savings.

Conclusions

For carefully selected patients with symptomatic, objectively confirmed menorrhagia and uterine size <12 week's gestation (or 10 to 12 cm in length), endometrial ablation by TCRE, ELA, or TBEA is a safe alternative to hysterectomy.

Endometrial ablation does not always result in amenorrhea, results may be limited simply to a reduction in the amount of menstrual blood loss each month.

Patients undergoing endometrial ablation should be completed with childbearing. However, since the possibility of pregnancy remains after endometrial ablation, sexually active women of reproductive age should make every effort to prevent pregnancy.

Endometrial ablation is associated with a low rate of complications and few major complications.

It is possible, that after endometrial ablation, there may be difficulty in diagnosing endometrial cancer if it occurs in residual tissue. The effects of estrogen replacement therapy in postmenopausal women on residual endometrium are unknown.

Presently, there is no objective, standardized measurement of menorrhagia, making comparison of treatments for abnormal uterine bleeding difficult.

Total direct and indirect costs of endometrial ablation are significantly lower than those for hysterectomy. However, if future re-treatment or additional procedures are needed, the difference in cost between endometrial ablation and hysterectomy diminishes.

Recommendations

The medical community should establish guidelines to objectively measure menorrhagia.

Additional data from well-designed, randomized controlled trials comparing the different techniques of endometrial ablation with drug therapies, hysterectomy, and each other should be undertaken to determine the most effective, safest, and least costly alternative to hysterectomy.

Physicians should discuss risks and benefits of all options, including endometrial ablation techniques and hysterectomy, with each patient in order to choose the procedure which will best suit each individual patient's needs.

Because the uterus is still in place following endometrial ablation, the possibility of endometrial cancer still exists. There is also a concern that the diagnostic evaluation for endometrial cancer in women who have undergone endometrial ablation may be more difficult. Therefore, all women presenting with suspicious symptoms should be evaluated.

Medical Background

Menorrhagia, heavy bleeding during the menstrual period, is common among women of reproductive age. In an international study by the World Health Organization, menorrhagia was found to affect 1011 out of 5322 women surveyed (19%) from 14 different countries. A benign condition, menorrhagia is profuse menstrual bleeding lasting longer than seven days, or greater than 80 ml. Two-thirds of the women with this level of blood loss during menstruation will display iron-deficiency anemia. Menorrhagia causes decreased hemoglobin, hematocrit, and serum iron levels. Menorrhagia is associated with underlying uterine lesions such as fibroids, polyps, cancer, endometriosis, adenomyosis (endometriosis within the endometrium), inflammation, or infection; with an anatomical abnormality; or with a systemic illness such as coagulopathy, hypothyroidism, hyperthyroidism, diabetes, systemic lupus erythematosus, leukemia, or cirrhosis: or it can be idiopathic. Menorrhagia may also be linked with a hormonal imbalance, pregnancy, hormone therapy, or use of an intrauterine contraceptive device or because of certain medications. Without an underlying lesion or condition, the heavy bleeding is known as dysfunctional uterine bleeding that can occur during either ovulatory or anovulatory cycles between the onset of menstruation and menopause.^1-15

Uterine bleeding may be dysfunctional if the patient reports the passage of blood clots, flooding, or socially embarrassing bleeding; and/or the use of one or two super or maxi sanitary pads requiring a change every one half to 2 hours. The degree of menstrual blood loss is usually assessed by the clinical history. Prolonged or irregular bleeding is often documented on a bleeding calendar. Blood loss can also be assessed by the alkaline hematin method in which hemoglobin extracted from sanitary pads is measured. A laboratory diagnosis of iron deficiency anemia is the only available clinical test to detect excessive bleeding. Patients are questioned about dysmenorrhea and premenstrual symptoms.^8,11,16 In the absence of an objective assessment of menstrual blood loss, what constitutes excessive bleeding is a subjective matter. Factors that affect the perceptions of an individual patient include age, cultural background, activity levels, personal hygiene, symptoms, and psychological state. Other factors that may lead a woman to seek treatment include the effects of bleeding on her sexual and psychosocial functioning, work performance, and knowledge of therapies for menorrhagia.¹¹

Patients with abnormal uterine bleeding undergo a series of tests to attempt to determine its cause. These include a complete physical and pelvic examination, and medical history with questions regarding general health, medication use, contraception, bleeding patterns, and cycle length. Blood tests are done to evaluate blood counts and hormone levels, and to rule out pregnancy. Other tests that may be done include cervical cytology, colposcopy, and biopsy, endometrial biopsy, dilation and curettage, pelvic ultrasound, diagnostic hysteroscopy, hysterosalpingography, and laparoscopy. Once the cause(s) of the bleeding has been determined, treatment is selected. The treatment of any underlying lesions or conditions may be sufficient to control excessive bleeding. For example, cervical and endometrial polyps can be resected, and uterine fibroids can be excised during operative hysteroscopy or myomectomy. Adenomyosis, endometriosis within the endometrial wall, is also managed surgically. The objectives of treatment are to control or eliminate bleeding, and to prevent its recurrence by correcting associated disorders while preserving fertility in women who wish to have children.^1,2,6,17,18

If no underlying lesions or conditions can be identified as the cause of abnormal bleeding, drug therapy is generally recommended as the first approach to treatment. Candidates for drug therapy include patients whose bleeding interferes with normal activities or causes anemia, those who wish to avoid surgery, and those wishing to maintain their fertility. The drugs, used alone or in combination for a period of at least one month, include: nonsteroidal anti-inflammatory agents; hormonal therapy with estrogen, progestins, oral contraceptives, danazol; and gonadotropin releasing hormone (GnRH) agonists. While these medications can reduce the menstrual flow by 20% to 100%, they are not without side effects. Side effects include headaches, nausea, weight gain, mood changes, bone loss, and in addition drugs are not always effective. Patient compliance, tolerance of side effects, and long-term adverse effects are also important considerations. Dilation and curettage, a procedure in which the cervix is dilated and some of the uterine lining is removed, stabilizes bleeding in some patients; however, the effects are not always long lasting. If the patient does not respond or becomes refractory to drug treatment, if it is contraindicated, or if it results in intolerable side effects, surgery is considered. Surgical treatments include hysterectomy and endometrial ablation.^{2,4,6,8,9,11,13,14,19-27}

Surgical Treatments

Hysterectomy

Hysterectomy, the removal of the uterus, is a common procedure. More than 25% of women in the United States will undergo this surgery before the age of 60. A total hysterectomy, which is less common, includes the removal of the ovaries and fallopian tubes as well as the uterus. In Minnesota, 8712 women in 1997, and 8689 women in 1998, had either abdominal or vaginal hysterectomy as a principal, inpatient procedure.²⁸ Based on data from the Centers for Disease Control and Prevention's National Hospital Discharge Survey (NHDS), an annual probability sample of discharges from non-federal, short-stay hospitals in the United States, an estimated 8.6 million women age 15 years or older had either an abdominal or vaginal hysterectomy between 1980 and 1993. The average annual rate of hysterectomy during this time period was estimated to be 613,000. The rate decreased between 1980 (7.1 surgeries per 1000 women) and 1987 (6.6 per 1000 women). The average annual rate was 5.5 per 1000 women between 1988 and 1993. The hysterectomy rate was highest every year for women ages 40 to 44 years, and lowest for women ages 15 to 24 years.^23,29-32

The most common indication for hysterectomy was uterine fibroids (leiomyomas). From 1988 to 1993, 62% of hysterectomies among black women, 29% among white women, and 45% among women of other racial groups were for leiomyomas. In the absence of a diagnosis of leiomyoma, endometriosis, endometrial hyperplasia, prolapse, or cancer, the hysterectomy rates for menstrual and menopausal bleeding from 1988 to 1990 for women between 25 and 54 years were 2.8 to 5.9 per 10,000 women. Although only 4.5% of hysterectomies performed for women of all ages were for a primary diagnosis of menstrual or menopausal bleeding, bleeding was the secondary diagnosis for 27% of women with leiomyomas, endometriosis, endometrial hyperplasia, prolapse, or cancer.^23,29-32 Other researchers have estimated that from 6% to 18% of hysterectomies in the United States are performed for dysfunctional uterine bleeding.^33,34 Although hysterectomy cures menorrhagia and results in a high rate of patient satisfaction, it is a major surgical procedure. When performed for benign and nonobstetric causes, hysterectomy has a 40% morbidity rate, and a mortality rate of approximately 10 per 10,000 surgeries. Its invasive nature, significant morbidity and mortality, long recovery period, associated health care costs, and the possibility of further complications including adhesions, injury to the bowel and/or bladder, fever and infections, post-operative bleeding and wound dehiscence have led to the development of alternative treatments for menorrhagia.^{7,9,11,14,18,20,21,23,30,34-38}

Endometrial Ablation

In the late 1980s, endometrial ablation was introduced into clinical practice as a less invasive alternative to hysterectomy for menorrhagia. Endometrial ablation, which is performed under general, regional, or local anesthesia, is a surgical procedure in which the endometrium, the uterine lining, is removed or destroyed. Endometrial ablation can be performed by electro surgery or the application of heat via laser energy or thermal fluids. Electro surgery and laser surgery are performed during hysteroscopy, this is a procedure in which a thin lighted tube is inserted through the vagina and cervix in order to visualize the inside of the uterus.^{7,9,11,14,18,20,21,23,30,34-38}

An advantage of endometrial ablation is that it does not involve the removal of the uterus, the operative and recovery times are shorter, and the complication rates are lower than for hysterectomy. It has been estimated that approximately 32,400 hysteroscopic endometrial ablations are performed annually in the United States.³⁹ There are several endometrial ablation techniques including transcervical resection of the endometrium (TCRE), endometrial laser ablation (ELA), and thermal balloon ablation of the endometrium (TBEA). Regardless of the technique used, the goal of endometrial ablation is to induce amenorrhea, the complete absence of the menstrual flow; hypomenorrhea, a marked decrease in the amount of menstrual blood loss; or eumenorrhea, normal menses. This is accomplished through the removal or destruction of the glands in the basal layer of the endometrium and the superficial myometrium thereby inhibiting its regeneration. A disadvantage of endometrial ablation is the possibility of persistent or recurrent menorrhagia requiring repeat ablation or eventual hysterectomy. Up to 21% of patients eventually required a hysterectomy while up to 36.7% required repeat ablation following the initial ablation procedure. Since ablation of the endometrium is most effective when the tissue is at its thinnest, the procedure is usually performed in the late menstrual or early proliferative phases of the menstrual cycle, after endometrial curettage, or after thinning with pharmacological agents such as Danazol, hormones, or GnRH agonists that suppress endometrial proliferation.^{4,8,9,11,20-22,36,40,41}

Patient Selection Criteria

Endometrial ablation is not appropriate for a woman with normal menstrual blood loss who simply finds her period to be an inconvenience.¹¹ While endometrial ablation may improve symptoms of dysmenorrhea, premenstrual symptoms, or pelvic pain, it is unclear whether the surgery should be offered to patients with these symptoms in the absence of documented menorrhagia.^11,18

Assuming that all other potential causes of excessive menstrual bleeding that could be treated medically or by alternative forms of surgery have been excluded, and that drug therapy has been tried, failed, refused, or is contraindicated, the major patient selection criteria for endometrial ablation include:^{2,3,9,11,18,42}

Menorrhagia or heavy menstrual bleeding that is defined as blood loss >80 ml per cycle; bleeding for longer than 8 days; blood loss sufficient to cause anemia; or blood loss or symptoms that interfere with normal activities.
Drug treatment failed, was contraindicated, was refused, or produced undesirable or intolerable adverse effects.
Uterine size <12 weeks gestation, and uterine cavity <12 cm in length.
Absence of endometrial or cervical cancer or precancer, and uterine lesions requiring other forms of surgery.

No underlying uterine lesions requiring other types of surgery.

Childbearing has been completed.

Absolute contraindications to endometrial ablation include:^3,11

Current or planned pregnancy.
Presence of endometrial or cervical cancer or precancer, or atypical endometrium.
Uterine size >12 cm or >12 weeks gestation.

Patient Evaluation

Several methods are used to assess quality of life before and after surgery. The Short Form 36 (SF36) is a 36-item questionnaire used to measure subjective health. It has eight multi-item dimensions covering physical and social functioning, physical limitations, emotional problems, mental health, vitality, pain, and general health perception. The scores for each dimension range from 0 (poor health) to 100 (good health). Quality of life is also assessed by the visual analog scale of the EuroQol© health questionnaire. Patients rate their current health status on a 0 (worst possible state) to 100 (best possible state) scale. Psychological status is also assessed by the Hospital Anxiety and Depression Scale.^16,43

Pre-operative Preparation

Just after menstruation, the endometrium is approximately 1 mm thick, while in the late secretory phase it is > 10 mm thick. Endometrial ablation is most effective when the endometrium is < 3 to 4 mm in thickness. Thus, to ensure optimal results, the surgery should be performed in the late menstrual or early proliferative phase of the menstrual cycle, or following pharmacological endometrial thinning or curettage. To produce a thin, atrophic endometrium prior to surgery allowing for more efficient ablation of the tissue and less bleeding, and to increase hemoglobin levels, patients may be given anti-estrogenic drugs such as Danazol (600 to 800 mg daily for 3 to 12 weeks), or GnRH agonists such as Goserelin (single 3.6 mg dose injected subcutaneously) or leuprolide acetate (two 3.75 mg doses injected intramuscularly) administered 3 to 4 weeks prior to surgery.^{5,9,11,18,20,40,41} A preoperative pelvic ultrasound is obtained to evaluate endometrial thickness.^18,36 A laminaria tent may be placed into the cervix the day before surgery to facilitate dilation.^11,18 Appropriate preparation of the endometrium shortens surgical time, reduces excessive fluid absorption, and improves operative results and safety.¹⁸

The instruments used to perform TCRE and ELA include a light source and hysteroscope, a uterine distention system, and the energy source with ablative attachments (thermal balloon ablation of the endometrium is performed without these instruments). A video camera and display monitor may also be used to enhance visualization of the uterine cavity. In order for the surgeon to adequately visualize the entire uterine cavity, it must be distended with liquid. Operative hysteroscopy usually utilizes low-viscosity media which continuously flows into and out of the uterus, clearing out surgical debris and blood, and improving the surgeon's field of view. An inflow pressure of 80 to 110 mm Hg ensures optimal distention and continuous irrigation. An automated irrigation delivery system may be used, or a bag of distending medium is hung at 1 meter above the patient and produces distention through gravity pressure. Distention media used during endometrial ablation include hypotonic agents such as glycine (1.5%), sorbitol (3%), or mannitol; isotonic agents such as normal saline; or the high-viscosity agent, dextran 70 (a viscous solution of 32% dextrose). Infused and collected fluid volumes are measured every 5 minutes.^10,11,18

General, regional (spinal and epidural anesthesia), or local anesthesia with or without sedation are used. Lidocaine is the agent most frequently used for a paracervical block. A Foley catheter is inserted to monitor urinary output. Aseptic precautions are taken, the hysteroscope is inserted through the cervix, and the sheath is flushed with distention media. The hysteroscope and sheath are directed into the uterine cavity, and the cavity is thoroughly examined before ablation begins.^10,11,18

Post-operative Follow-up

Post-operative pain and cramping are treated with anti-prostaglandins, nonsteroidal anti-inflammatory agents, or paracervical block with long-acting bupivacaine. Oral analgesics are also prescribed. Intravenous diuretics may be administered to promote fluid clearance. Vaginal bleeding following endometrial ablation lasts approximately nine days. This is followed by a watery, bloody discharge which lasts for an average of 14 days. Since the possibility of pregnancy remains after endometrial ablation, sexually active women of reproductive age should make every effort to prevent pregnancy.^11,18,44

Endometrial Ablation Techniques

Transcervical Resection of the Endometrium (TCRE)

During TCRE, the endometrium is destroyed with a resectoscope, a telescopic instrument with a wire loop or a rollerball at its tip. An electrical current delivered to the tip of the instrument cuts or coagulates the endometrium.

Procedure

TCRE is performed with the patient under general or regional anesthesia. Electro cautery ablation of the endometrium, using a standard urological resectoscope, is performed with a wire loop and/or a rollerball (a freely-moving barrel on a resectoscope). As the loop electrode of the resectoscope is moved forward, the tissue in front of it is cut when the cells vaporize. Due to the difficulty in reaching the cornual regions of the uterus with the wire loop, a rollerball electrode may be introduced to thermally destroy that part of the endometrium. The goal of TCRE is to ablate the basal layer of the endometrium as well as the first few millimeters of the myometrium in order to ensure endometrial destruction. The first incision resects to a depth of approximately 3 mm, exposing the base of the endometrial glands. The second incision resects 2 to 3 mm of myometrium. Resection begins at the fundal areas and moves downward towards the internal os. Pieces of resected tissue may be removed by curettage or with polyps forceps. During rollerball ablation, a 2 to 4 mm barrel-shaped electrode is rolled over the endometrium at a speed of 10 to 15 mm per second. The basal layer is reached via a 3 to 4 mm depth of ablation at a coagulating power of 50 to 80 watts. If present, submucous fibroids or polyps may be resected with the wire loop. The total operating time is approximately 30 to 60 minutes.^5,11,18

The patient selection criteria for TCRE includes: symptomatic menorrhagia unresponsive to drug treatment, candidate for hysterectomy, uterine size <12 weeks gestation, negative cervical cytology, no endometrial cancer or precancer, no adnexal tumors, and patients who have completed childbearing. The exclusion criteria includes: uterine size >12 weeks gestation, or the presence of endometrial or cervical cancer or precancer, pelvic inflammatory disease, endometriosis, pathology necessitating hysterectomy, stress incontinence, moderate-to-severe uterine prolapse, blood clotting disorder, intrauterine device or anticoagulant use, unstable medical condition, fibroids >3 cm to >5 cm in diameter or >50% intramural extension, and intermenstrual or postcoital bleeding.^{16,23,29,35,44}

Findings

In the studies on TCRE, from 13% to 64% of the patients were amenorrheic, 20% to 76% had reduced menstrual blood loss, 2.7% to 16% were treatment failures, 0% to 16% required repeat ablation, and 0% to 18.2% required a hysterectomy during follow-up.^{16,23,29,35,44} In one study the amenorrhea rate declined from 46% at 1 year to 21% at 3 years.²³ Longer term follow-up of patients from one study showed that treatment failure rates increased from 11% at 4 months to 21% at 2.8 years.^20,48 In addition, the rate of repeat ablation increased from 7% to 10%, and the rate of hysterectomy increased from 4% to 18.2% during the longer follow-up period. A high percentage of TCRE patients complained of persistent bleeding, pain, or worsening symptoms. At the later follow-ups more TCRE patients than hysterectomy patients were found to require time off of work.^16,29 In these studies, the length of hospitalization and recovery time were lower. The need for analgesia and time to return to normal activities were reduced for the TCRE patients compared with the hysterectomy patients. The TCRE group experienced fewer complications; however, more hysterectomy patients expressed satisfaction with their treatment.^{16,23,29,35,44} None of the studies systematically analyzed prognostic variables that may have affected treatment outcome.

Complications

Complications during TCRE resection such as fluid overload (1% to 3%), uterine perforation (0.9% to 4%), and hemorrhage are likely due to surgeon inexperience, and use of the loop technique as opposed to the rollerball technique. If resection is deeper than approximately 4 mm, injury can result to the pelvic viscera. This is most likely to occur at the isthmus and cornua where myometrial thickness leaves little margin for error. The risk of fluid overload and hemorrhage are increased during resection due to the extensive transection of endometrial blood vessels. While intra- or post-operative hemorrhage is uncommon (1%), it can occur particularly if there is deep myometrial injury. Inspection of the newly-resected uterine cavity and coagulation of bleeding points ensures hemostasis and reduces the risk of hemorrhage.^11,29,44 Other reported complications include: pelvic pain/peritonitis (1%); cervical tear (0.9% to 3%); blood transfusion (1% to 2%); pyrexia (2%); pelvic infection (2%); pelvic hematoma (1%); hematometra (0.9%); endometritis (0.9%); hydrosalpinx (0.9%); need for postoperative analgesia (16%), and backache (4%).^{13,23,2,29,44}

Endometrial Laser Ablation (ELA)

During ELA, thermal energy produced by a neodymium-yttrium-aluminum-garnet (Nd:YAG) laser destroys the endometrium.

Procedure

ELA is performed as an inpatient or outpatient procedure under general or regional anesthesia. Distention media, usually normal saline, is delivered into the uterus by a peristaltic pump or sphygmomanometer, or under gravity, and uterine pressure is maintained between 80 and 100 mm Hg. The amount of media is measured so that the development of fluid overload can be monitored. The hysteroscope is placed, and the endometrium is visualized. The laser is placed through the operating channel of the hysteroscope and activated at a power level of approximately 60 W to 80 W delivered by a 600 µm bare quartz fiber. There are two techniques used for ELA, the touch or drag technique, and the non-touch or blanching technique. During the touch technique, the laser tip is lightly applied to the endometrial surface and gently swept across the uterine cavity. For the non-touch technique, the laser tip is brought within 1 to 5 mm of the endometrial surface but does not touch it. Compared with the touch technique, this technique reduces the potential for fluid absorption since the blood and lymphatic vessels are coagulated as opposed to being cut open. Some physicians use a combination of the two techniques. Ablation begins at the cornual and fundal areas with the delivery of short, 5 to 10 second bursts of laser energy. The anterior wall is ablated to the level of the cervical os followed by ablation of the lateral and posterior walls. The Nd:YAG laser coagulates and denatures the endometrium to a depth of approximately 4 to 6 mm. As coagulation proceeds, tissue damage is minimized since the laser energy absorbed decreases, and the amount that is reflected increases. Fluid balance is calculated every 5 minutes and intravenous furosemide is given if the fluid absorption exceeds 1500 ml. ELA takes approximately 30 to 180 minutes. If successful, post-ablation histology several months later will reveal a single layer of simple cuboidal epithelium devoid of endometrial glands.^{5,11,40,43,45}

The patient selection criteria for ELA includes: severe menorrhagia unresponsive to drug treatment, candidate for hysterectomy, no endometrial cancer or precancer, uterine size <12 weeks gestation (one studies specified uterine size <10 weeks gestation), and patients who have completed childbearing. The exclusion criteria includes: uterine size >10 or >12 weeks gestation, and the presence of endometrial cancer or precancer. One study excluded patients with fibroids >2 cm in diameter, while one excluded patients with large submucous fibroids unsuitable for hysteroscopic surgery.^{36,37,40,41,43,45} One study did not exclude patients whose uteri were >10 weeks gestation or >10 cm in size.⁴⁰

Findings

In the clinical studies on ELA, from 6% to 81% of the patients became amenorrheic, 19% to 66.2% had reduced menstrual blood loss, 0% to 16.6% were treatment failures (20% to 32% of those after repeat ablation), 4.5% to 14.3% required repeat ablation (0.4% required a third ablation), and 1.8% to 21% required a hysterectomy during follow-up. Complication rates were generally low.^{36,37,40,41,43,45} In one study a higher rate of amenorrhea post-endometrial laser ablation was observed in older women compared with younger women (97% versus 6%).³⁶ In the same study, adenomyosis was diagnosed after hysterectomy in 81.8% of the patients who required the surgery after ELA. In addition, treatment failure was more likely after repeat ablation (32%) compared with after primary ablation (7%). In another study 56% of patients who required hysterectomy after failing ELA were found to have underlying conditions including endometriosis, adenomyosis, or leiomyomas.⁴⁵ In another study multi variate analysis showed that the rate of hysterectomy was significantly reduced in patients with underlying conditions (polyps, fibroids, or uterine malformations), and significantly increased in patients who required repeat ablations.³⁷ The same study showed with a survival curve analysis that 21% of patients required hysterectomy over a follow-up time of 6.5 years.³⁷

Complications

Fluid overload is the most serious complication associated with ELA (0.9% to 9%) and occurs when there is excessive absorption of distention medium through large exposed blood vessels in the myometrium. Fluid overload associated with the development of hyponatremia and pulmonary edema (0.4%) can lead to delayed extubation and continued ventilation, a need for treatment with diuretics, and neurological complications.^5,36,40

A sheathed quartz tip laser fiber with a gas-cooled tip or a gaseous distending medium must never be used during ELA due to the possibility of potentially fatal embolization.⁵

Other potentially serious complications associated with ELA include uterine perforation (0.2%), hemorrhage, and thermal damage to surrounding tissues and structures. Other complications include the development of hematometra or pyometra as a result of scarring and obliteration of the endocervical canal coupled with continued bleeding from residual endometrium.^5,40 Other reported complications include: balloon tamponade (0.3% to 5%); infection (0.2% to 0.6%); urinary tract infection (2%); pelvic sepsis (3%); endometritis/myometritis (0.2%); laser burns at the cornua (1.1%); small bowel obstruction (0.6%); blood transfusion (0.2%); side effects from danazol pretreatment (6% to 40%); side effects from goserelin pretreatment (25% to 40%); pregnancy (0.5%); and hematometra (0.5%).^{36,37,40,43,45} The lack of endometrial tissue for histopathological examination following laser ablation has caused some concern due to the fact that it can lead to a missed diagnosis of endometrial cancer. Thorough preoperative endometrial sampling and analysis helps to reduce the likelihood of missing an existing cancer but the detection of endometrial cancers occurring in residual endometrium is more problematical.^5,45

Thermal Balloon Ablation of the Endometrium (TBEA)

A newer alternative for endometrial ablation is thermal balloon ablation of the endometrium (TBEA), a procedure that does not require the use of a hysteroscope. A latex balloon is inserted through the vagina into the cervix and is filled with heated fluid that coagulates the endometrium.

Procedure

TBEA is performed under general or local (paracervical) block anesthesia with or without intravenous sedation. Following the administration of anesthesia, a latex balloon attached to a uterine catheter is inserted through the vagina into the uterus, and filled with a small volume of sterile 5% dextrose and water to conform to the shape of the uterus. A heating element located inside the balloon raises the temperature of the fluid to 87° C for 8 minutes. This results in the thermal coagulation of 3 to 5 mm of the endometrium by heat transfer through the balloon. A controller connected to the catheter monitors and displays catheter pressure, regulates temperature, and controls therapy time. The entire procedure can be performed in less than 1 hour on an outpatient basis if local anesthesia and intravenous sedation are used. Mild to moderate uterine cramping may occur during the first day post-ablation but otherwise, recovery is generally uncomplicated. Vaginal discharge or spotting may occur during the first week, and patients generally resume normal activities within 24 hours of treatment.^13,24

The patient selection criteria for TBEA includes: premenopausal women with severe menorrhagia unresponsive to or unwilling to continue drug treatment, candidate for hysterectomy or endometrial ablation, no endometrial cancer or precancer, negative cervical cytology, uterine size <12 cm in size, and completed childbearing. Exclusion criteria includes: uterine cavity >12 cm in size, endometrial cancer or precancer, submucous fibroids, uterine polyps, adnexal pathology, septate uterus, pelvic inflammatory disease, undiagnosed genital bleeding, pregnancy or desire to become pregnant, genital tract infection or cancer, and latex allergy.^{13,14,20,22,24,46} Two studies excluded patients who had previously undergone endometrial ablation.^13,24 In one study, patients who wanted to eliminate menses were advised to undergo hysterectomy.²⁰

Findings

In the clinical studies on TBEA, from 3% to 55% of the menorrhagia patients were amenorrheic, from 38.9% to 76% had reduced menstrual blood loss, from 0% to 19.8% were treatment failures, from 0% to 36% required repeat ablation, and from 0% to 11.1% required a hysterectomy during follow-up. There was a low rate of complications following TBEA.^{13,14,20,22,24,46} In one study, the treatment failure rate increased from 17% at 6 months to 43% at 12 to 18 months.²⁰ In one study it was stated that age, parity, uterine size, and operative technique were not related to outcome, but no details or statistical analyses of these variables were provided.²⁰ In another study logistic regression analysis revealed that treatment failure was significantly related to higher uterine volume, preoperative curettage, and a higher number of pads used prior to surgery.²⁴ The older the age the higher the likelihood of treatment success. In the study that compared thermal balloon ablation of the endometrium (TBEA) with transcervical resection of the endometrium (TCRE), a significantly higher percentage of TCRE patients were amenorrheic at follow-up although improvement in menorrhagia and patient satisfaction rates were similar between groups; however, TBEA had a lower complication rate. No differences in treatment outcome were found (menstrual status, mean days of flow, patient satisfaction, or hysterectomy rate) between patients who were and were not pre-treated with endometrial thinning agents prior to thermal balloon ablation of the endometrium.^13,46

Complications

While the likelihood of pregnancy after TBEA is low, it is still possible if all endometrial tissue is not ablated. While there are limited data on the pregnancy following TBEA, such pregnancies may be risky to both mother and child. Women who undergo this procedure should be advised to use effective contraception.⁴⁷

Thermal balloon ablation of the endometrium (TBEA), the following complications were reported: uterine cramps (5.6% to 23.3%); abdominal cramps (5%); cystitis (0.3%); endometritis (2% to 2.3%); hematometra (0.7%); pain requiring hospitalization (0.3%); and urinary tract infection (0.8%).^{13,14,20,24,46}

Endometrial Ablation Compared to Hysterectomy

In the two randomized controlled trials that compared outcomes for menorrhagia patients following endometrial ablation (laser ablation or endometrial resection) or hysterectomy at 1 year and at 4 years, the patient selection criteria included dysfunctional uterine bleeding, candidate for hysterectomy, age <50 years old, uterine size <10 weeks gestation, and no endometrial cancer or precancer.^26,48 In the first study the length of hospitalization and duration of postoperative pain, and the time to return to normal activities were significantly lower in the endometrial ablation group compared with the hysterectomy group.⁴⁸ A higher percentage of patients in the hysterectomy group experienced major complications compared with the ablation group (5.2% versus 1%). The incidence of post-operative infection was significantly lower in the ablation group. At 1 year, 22% of the endometrial ablation patients were amenorrheic, 62% were hypomenorrheic, 3% were unchanged, 11.5% required repeat ablation, and 14% required a hysterectomy for persistent menorrhagia. The patient satisfaction rate was significantly higher in the hysterectomy group, but between 70% and 90% of the women were satisfied with the outcome of endometrial ablation, at follow-up of six and twelve months, even though an appreciable proportion of women required a second procedure.⁴⁸ When the patients were followed-up at 4 years (74% of the original group responded),²⁶ 34% of the endometrial ablation group required additional surgery (ablation and/or hysterectomy) compared with 0% in the hysterectomy group. The percentage of ablation patients requiring hysterectomy increased to 18%, those requiring ablation increased to 12%, and those requiring both was 3%. A higher percentage of ablation patients were amenorrheic at the later follow-up (45% versus 22%). Satisfaction rates were high for both groups, 80% in the ablation group versus 89% in the hysterectomy group, the difference reflecting retreatment. However, a significantly higher percentage of hysterectomy patients reported having improved general health compared with the ablation group (67% versus 38%).²⁶

Endometrial Ablation Compared to Drug Treatment

In the randomized controlled trial that compared transcervical resection of the endometrium (TCRE) with drug treatment for menorrhagia,²⁷ the patient selection criteria included: menorrhagia, uterine size <10 weeks gestation, normal endometrium, and completed childbearing. In this study, a significantly higher percentage of TCRE patients were amenorrheic compared with the drug treatment group (37% versus 3%), and although the mean bleeding score decreased in both groups, it was significantly lower in the TCRE group. Significantly more patients in the drug treatment group had unchanged or persistent menorrhagia compared with the TCRE group (52% versus 8%), and significantly more TCRE patients were satisfied with their treatment (76% versus 27%).²⁷

Safety

Safety issues with endometrial ablation include:¹⁰

Distention-Media Complications

The media used to distend the uterine cavity during hysteroscopy can cause fluid overload, allergic reactions, and other toxic reactions. Fluid overload is associated with prolonged operating times, the use of high distending pressures, and the resection of large myomas at surgery. Absorption of excess amounts of hypotonic distending solution results in increased central venous pressure and hyponatremia that, if untreated, lead to pulmonary edema, hypotension, cerebral edema, and potentially fatal cardiovascular collapse. To avoid these complications, intra operative fluid use must be strictly monitored during the procedure, and the patient observed for signs and symptoms of fluid overload. Most physicians will stop a hysteroscopic procedure at a 1 to 1.5 liter fluid deficit when using glycine, sorbitol, or mannitol to avoid hyponatremic hypovolemia; a lower threshold may be required for older patients or those with pre-existing cardiovascular problems. Dextran 70 has been associated with fluid overload, pulmonary edema, intra vascular coagulopathy, renal insufficiency or failure, and rhabdomyolysis, as well as anaphylactoid reactions. Glycine can result in hyperammonemic encephalopathy, and transient blurred vision and blindness.^10,11,18

Mechanical Complications

A rare, potentially fatal, complication of operative hysteroscopy is air embolism. This may occur when venous sinuses on the surface of the endometrium are opened and exposed to ambient air. Factors that pre-dispose to air embolism include the Trendelenburg position, cervical trauma, and exposure of a dilated cervix to ambient air. Other mechanical complications associated with hysteroscopy include cervical laceration and uterine perforation. Large uterine perforations may require laparotomy to rule out extensive injury or bleeding. Untreated damage to the bowel can be fatal. If the ablation extends too deeply into the myometrium and major blood vessels are severed, uterine hemorrhage can occur. Hysterectomy may be required to achieve hemostasis.^10,18

Thermal Complications

Thermal complications are possible when electrical or laser energy are used. While uncommon, thermal bowel injury has been reported to occur during Nd:YAG laser surgery and during rollerball coagulation.¹⁰

Endometrial Cancer

It is possible that cancer could develop in residual endometrium remaining after ablation. A cancer developing in an ablated and scarred uterine cavity may be difficult to diagnose. The effects of hormone replacement therapy upon residual endometrium in postmenopausal women is unknown. Thus, some physicians prescribe combined estrogen and progestin replacement therapy to postmenopausal patients in the event that some endometrial tissue persists.^{5,7,8,11,18,29}

Pregnancy

Since the goal of endometrial ablation is the complete destruction of the endometrium, theoretically, pregnancy should not be possible after surgery. However, pregnancy is possible after ablation, even in women with amenorrhea. If pregnancy occurs in these patients, it can be difficult to diagnose. There are inherent risks involved in a pregnancy occurring in an ablated uterus. The risk of ectopic pregnancy is increased, and even if the embryo successfully implants in the endometrium, possible complications include a decreased blood supply to the uterus resulting in fetal growth impairment, and impaired placentation.¹⁸

Issues of Controversy

Assessment of Menorrhagia and Menstrual Blood Loss

The diagnosis of menorrhagia is inexact and excessive blood loss is defined in various ways, most of which are subjective. It is difficult to objectively measure excessive menstrual blood loss; only 40% of women with this complaint have it confirmed by objective means. Patients may be questioned about the passage of blood clots, or flooding despite the use of adequate measures. However, the amount of menstrual blood loss does not necessarily correlate with the number of tampons or sanitary napkins used or with the duration of bleeding. Pad counting is a subjective process that relies on accurate reporting by the patient. Anemia is an objective sign of menorrhagia, and normal menstrual blood loss is defined as an amount <80 ml per cycle. However, some women with blood loss in the normal range complain of menorrhagia, and conversely, some women with confirmed menorrhagia perceive their blood loss as light or moderate. Thus, some patients may be treated for perceived excess bleeding that is viewed as inconvenient rather than for objectively diagnosed menorrhagia.^5,7,8,24

Over-use of Endometrial Ablation

There is some concern that the less invasive nature and easier recovery from endometrial ablation compared with hysterectomy may increase the overall rates of surgery since it may lower the threshold for treatment of menorrhagia; women with milder cases that could be managed with more conservative means may undergo ablation.^7,37,48

Outcome Measures

While the induction of amenorrhea is commonly used to assess the outcome of endometrial ablation, its use as a measure of treatment efficacy is limited if information on the patients' menopausal status is not provided since it would be difficult to determine whether menstrual bleeding stopped as a result of the treatment or menopause.³⁷ While amenorrhea is often the primary outcome measure after endometrial resection, the usual outcome is hypomenorrhea. Hysterectomy is the only treatment that guarantees amenorrhea.^11,13,18,26

The varying definitions of menstrual status or flow used throughout the literature can affect data interpretation. For example, if having a discharge at menses is classified as amenorrhea, the rates of amenorrhea induced by a particular treatment will be higher than if discharge at menses is not included in this outcome category.^43,48

Patient Satisfaction and Menstrual Blood Loss

A higher subjective failure rate has been reported for women who believed they had heavy menstrual blood loss (18%) but who in fact had normal periods compared with women who had a documented blood loss of >80 ml per cycle (9%).⁴⁹ A 50% reduction in menstrual blood loss may indicate treatment success for one patient and failure for another. This is why the patient's assessment of the treatment results may be a more important indicator of success than the induction of amenorrhea.^24,45

Costs And Cost-effectiveness

Two studies comparing the costs in U.S. dollars of endometrial ablation with the costs of hysterectomy were found during searches of the medical literature.^4,25 The study methodologies and findings are summarized in this section. One study compared the cost-effectiveness of outpatient endometrial ablation by TCRE with inpatient hysterectomy for dysfunctional uterine bleeding in patients enrolled in a national managed healthcare organization in the United States. The data were derived from claims information collected between 1990 to mid-1992. The findings are presented in terms of the cost of a procedure to the health plan as opposed to the fee charged by the health professional or institution associated with treatment. The data were based on claims made for the provision of services associated with endometrial ablation and hysterectomy during the period beginning 180 days before treatment, and continuing for up to 365 days afterwards. This period was subdivided into a pretreatment period (the time ranging from 180 days to 5 days before treatment), a periprocedural period (5 days before to 5 days after treatment), and a postprocedure period (from 6 to 365 days after treatment).⁴

A review of submitted claims revealed 85 patients who had undergone TCRE (mean age, 39.3 years) for dysfunctional uterine bleeding. These were matched by treatment date, health plan, and diagnostic indication with 255 patients who had undergone hysterectomy (mean age, 39.9 years) during the specified time period. Among the patients who underwent TCRE, 17 received danazol and 12 received leuprolide acetate for endometrial thinning prior to surgery. Among the 255 hysterectomy patients, 178 had abdominal surgery and 77 had vaginal hysterectomies.⁴

The findings of the cost comparison study on hysterectomy and TCRE are presented in Table 1. Costs were calculated for all TCRE patients, and also after excluding data on 13 TCRE patients who required retreatment by repeat TCRE or hysterectomy after primary TCRE.

Table

Table 1: Comparative Health Care Costs for All Patients Treated for Dysfunctional Uterine Bleeding with Endometrial Ablation or Hysterectomy.

Table

Table 2: Comparison of Total Direct and Indirect Costs Associated with TCRE and Hysterectomy.

Whether expressed as the mean per patient cost to health plan, or the mean cost to health plan per patient day, the total cost of a hysterectomy for menorrhagia for all patients, including endometrial ablation patients who required retreatment, was higher than the cost of endometrial ablation [$6412 for hysterectomy versus $3749 for endometrial ablation and $13.02 for hysterectomy and $8.09 for endometrial ablation (P=0.0001), respectively]. The only times that costs for hysterectomy were lower than costs for endometrial ablation were during the postprocedure period [mean per patient cost to health plan of $174 for hysterectomy versus $722 for endometrial ablation, and mean cost to health plan per patient day of $0.56 for hysterectomy versus $2.53 for endometrial ablation (P=0.002)].⁴

Since the higher postprocedure cost of endometrial ablation may be related to the costs of retreatment during the 12 months after treatment, the second analysis compares the costs of the two procedures excluding patients who required retreatment by endometrial ablation or hysterectomy. In this comparison, the mean per patient cost to health plan and mean cost to health plan per patient day remained higher for hysterectomy [$6412 versus $3185 and $13.02 versus $7.02 (P=0.0001), respectively]; however, the mean per patient postprocedure cost to health plan for endometrial ablation decreased from $722 to $194, a figure that approximated the cost of hysterectomy, i.e., $174. The mean postprocedure cost to health plan per patient day also decreased from $2.53 to $0.70, a cost that was not significantly different from the cost of hysterectomy ($0.56) (P=0.37).⁴

In this analysis, which analyzed costs related to outpatient endometrial ablation and hysterectomy for menorrhagia for 6 months prior to treatment, for the perioperative period and for up to 1 year after treatment, the mean cost per patient treated was similar for the two treatments during the 6 months prior to surgery. The cost of hysterectomy, including hospitalization and other costs, was two times that of endometrial ablation, while the cost for following and retreating endometrial ablation cases was four times the cost of hysterectomy. Nevertheless, due to the higher cost of inpatient treatment for hysterectomy compared with outpatient treatment for endometrial ablation, the total cost of 18 months of care for hysterectomy patients was higher than the costs of care for endometrial ablation patients, even when the analysis included patients who required retreatment after ablation. When patients who required retreatment after endometrial ablation were excluded (i.e., the treatment failures), the postprocedure costs for ablation were not significantly different from the costs for hysterectomy suggesting that a reduction in the number of treatment failures may improve the cost-effectiveness of endometrial ablation.⁴ It should be noted that in this analysis, the costs of endometrial ablation were for outpatient procedures; in the studies analyzed in this report, the TCRE patients were hospitalized.

In a prospective case series with historical controls, long-term treatment costs between TCRE and hysterectomy for menorrhagia were compared.²⁵ During 1992 to 1994, 72 patients with menorrhagia who failed drug treatment underwent TCRE; 8 patients >50 years old were excluded from the study group, thus the study group consisted of 64 TCRE patients (mean age, 41; range, 27 to 50 years). The control group was selected from a chart review of 403 patients who underwent hysterectomy between 1990 and 1992. Patients with uterine size >14 weeks gestation, uterine weight >300 g, and who had vaginal repairs, endometriosis, uterine prolapse, ovarian pathology, or tumors, or who were >51 years old were excluded. Thus, 46 hysterectomy patients (mean age, 41; range, 26 to 51 years) served as the control group. The TCRE patients underwent preoperative hysteroscopy and endometrial biopsy, and received a single 3.75 mg dose of leuprolide acetate for endometrial thinning at 4 weeks prior to surgery. Endometrial ablation consisted of wire loop resection and rollerball coagulation. Submucous myomas were removed by resection. In 1997, a questionnaire was sent to all of the patients who underwent TCRE to assess their menstrual status, pain, and overall satisfaction with the procedure. Charts were reviewed to obtain data on intraoperative and postoperative complications, and the need for surgery. Hysterectomy patients were interviewed by telephone to determine the length of convalescence.²⁵

The total direct cost associated with the surgical treatment of menorrhagia was defined as the total of the surgeon's fee, hospital charges, anesthesia fees, cost of leuprolide, and office charges for follow-up gynecological care for 3 years after the initial procedure. The charges were obtained from Saint Vincent Hospital and Fallon Clinic (Worcester, Massachusetts) databases. Professional fees were obtained from group practices. So that the cost of treatment failure was included in this analysis, charges associated with additional treatments following a failed TCRE were determined and added to the total cost of that case. Charges associated with readmission for postoperative complications or gynecological problems during the 3-year follow-up were included in the total cost. Indirect costs included those related to lost productivity during the postoperative recovery period. These varied depending upon the length of recovery, annual earnings for working women, annual value of housekeeping services for homemakers, and percentage of the female population in a given age range who were part of the labor force or who were homemakers. Published methods and assumptions were used to determine the indirect costs associated with each surgical treatment.²⁵

Although the two groups were similar with respect to age and parity, the hysterectomy patients weighed less (P=0.005), and had a larger uterine size (P=0.006). The mean operating time (38 minutes versus 107 minutes), mean length of hospitalization (0.5 days versus 2.8 days), and the mean recovery period (5 days versus 32 days) were significantly lower in the TCRE group compared with the hysterectomy group (P<0.001). The percentage of patients admitted for additional treatment or surgery was similar between groups (12.5% for TCRE versus 10.9% for hysterectomy) (P=NS). The incidence of intra operative complications was similar between the two groups (1.6% for TCRE versus 4.3% for hysterectomy) (P=NS); however, the hysterectomy group had a higher incidence of postoperative complications (21.7% versus 6.3%) (P=0.02).²⁵

Among the 63 TCRE patients with follow-up data on menstrual status (mean follow-up, 48.5 months; range, 36 to 68 months), 49% were amenorrheic, 30% were hypomenorrheic, and 8% were eumenorrheic. Eight (12%) patients were treatment failures (mean time to failure, 15 months; range, 6 to 29 months). Two of the 64 (3.1%) patients had repeat TCRE, and 6 (9.3%) had a hysterectomy. Among the 8 patients who failed TCRE, 5 had intramural fibroids, 2 had adenomyosis, and 1 had a normal uterus at repeat surgery. Overall, 85% of the TCRE patients were satisfied with their outcome and dysmenorrhea improved in 85% of the patients. The two patients who had repeat ablation were amenorrheic at 20 and 47 months after surgery. Five (11%) patients in the hysterectomy group required follow-up treatment or surgeries during the 3-year follow-up.²⁵

The mean direct costs for TCRE and hysterectomy were $5434 and $8417, respectively, and the mean indirect costs for TCRE and hysterectomy were $525 and $3360, respectively. The total direct and indirect costs per case were $5959 for TCRE and $11,777 for hysterectomy (P < 0.001). TCRE resulted in a savings of $5818 per case compared with hysterectomy.²⁵ In this study, the operating time, length of hospitalization, recovery time, and postoperative complication rate were significantly lower for TCRE than for hysterectomy. Thus, the direct and indirect costs of TCRE were also significantly lower than those for hysterectomy.²⁵

The cost of TBEA is estimated at $2500 to $3000, depending on the need for general anesthesia and the setting in which the procedure is performed (office versus hospital).⁵⁰ No information on the costs of ELA for menorrhagia was found, and no studies were identified that systematically compared the costs of TCRE, ELA, and TBEA or with one exception, that compared the costs of drug treatments for menorrhagia with endometrial ablation or hysterectomy.

Conclusions

Endometrial ablation does not always result in amenorrhea, results may be limited simply to a reduction in the amount of menstrual blood loss each month.

Endometrial ablation is associated with a low rate of complications and few major complications.

It is possible, that after endometrial ablation, there may be difficulty in diagnosing endometrial cancer if it occurs in residual tissue.

The effects of estrogen replacement therapy in postmenopausal women on residual endometrium are unknown.

Presently, there is no objective, standardized measurement of menorrhagia, making comparison of treatments for abnormal uterine bleeding difficult. Total direct and indirect costs of endometrial ablation are significantly lower than those for hysterectomy. However, if future retreatment or additional procedures are needed, the difference in cost between endometrial ablation and hysterectomy diminishes.

Recommendations

The medical community should establish guidelines to objectively measure menorrhagia.

Appendix I: Methodology

Evidence evaluated for this report was obtained from searches of the PreMEDLINE, MEDLINE, EMBASE, Current Contents, and HealthSTAR databases that spanned the years 1966 to June 1999. Search terms included endometrial ablation, endometrial resection, hysterectomy, resectoscopy, laser surgery, and thermal balloon ablation, combined with uterine bleeding, menorrhagia, cost, and cost-effectiveness as keywords, subject words, and title words. The searches were also limited to the English language, and human subjects. An attempt was made to identify randomized controlled trials comparing the efficacy and safety of TCRE, ELA, and TBEA with hysterectomy, with each other, or with standard drug treatment, and in which > 50 patients were followed for >6 months. If such studies were unavailable, an attempt was made to identify randomized controlled trials or prospective case series that met the criteria for follow-up time and/or sample size. Additional articles were identified during manual searches of the citations of the retrieved articles. Since several randomized controlled trials or prospective case series were found for each surgical technique, retrospective case series and case reports were excluded from this evaluation.

Information was also obtained from the U.S. Food and Drug Administration Web site, the Health Care Financing Administration, the American Medical Association, the American College of Obstetricians and Gynecologists, and the American Association of Gynecologic Laparoscopists.

Appendix II: Government Agencies And Professional Organizations Guidelines

American College of Obstetricians and Gynecologists (ACOG): In 1994, ACOG published a criteria set for physicians to use in determining the appropriateness of endometrial ablation for the treatment of abnormal uterine bleeding in women of reproductive age.⁴

The criteria set states that each numbered and lettered item must be present:⁴

Confirmation of indication
1. History of excessive uterine bleeding evidenced by either of the following:
  1. Profuse bleeding or repetitive periods lasting > 8 days.
  2. Anemia due to acute or chronic blood loss.
2. Failure to find uterine or cervical pathology on physical examination that would cause abnormal bleeding.
3. Laboratory data
  1. No endometrial neoplasia
  2. No malignancy found in cervical cytology specimens
4. No finding of remediable cause by hysteroscopy
Actions prior to procedure
1. Document counseling of patient regarding completion of childbearing.
2. Consider patient's medical and psychologic risks concerning consequences of ablation and alternative modes of therapy.
3. Determine that hormone treatment (estrogen-progesterone) was not successful.
4. Rule out a bleeding diathesis or use of medications that may cause bleeding.
5. Rule out pregnancy.
6. Assess surgical risk from anemia and need for treatment.
7. Rule out atypical endometrial hyperplasia or invasive carcinoma of endometrium and cervix.
8. Counsel patient regarding potential need for surgical sterilization.
9. Counsel patient that cyclic menstrual bleeding of reduced volume may occur.
Contraindication: Desire to maintain fertility.

American Medical Association (AMA): In 1991, the AMA published a Diagnostic and Therapeutic Technology Assessment (DATTA) on laser ablation of the endometrium.¹⁰ A total of 93 of 103 DATTA panelists surveyed on the safety and effectiveness of Endometrial Laser Ablation responded for a response rate of 90%. Of those that responded, 34 (36%) offered no opinion on the procedure. Thus, the responses of 59 panelists are included in the evaluation of the safety and effectiveness of the procedure. The panelists, who included 13 gynecologists and 46 obstetrician-gynecologists of whom 56 were certified by the American Board of Obstetrics and Gynecology, were asked whether ELA with prior danazol therapy is a safe and effective treatment alternative for women with menorrhagia that is unresponsive to drug treatment or dilation and curettage. The consensus of the DATTA panelists found ELA to be appropriate with respect to safety and effectiveness for the treatment of menorrhagia although a number of panelists considered the treatment to be investigational. No panelist considered the technique to be doubtful or investigational.¹⁰

The AMA also reviewed the literature on Endometrial Laser Ablation and briefly summarized the data on key findings from seven case series comprised of from 10 to 335 patients.¹⁰ In this review, the percentage of patients who were amenorrheic after ELA ranged from 12% to 60% (median, 33%); from 40% to 88% of patients (median, 67%) continued to menstruate after ELA. The percentage of successful procedures, defined as the induction of amenorrhea or the reduction of the menstrual flow to acceptable levels not requiring additional therapy, ranged from 52% to 100% (median, 91%). From 0% to 48% (median, 8%) of patients required additional treatment by hysterectomy. The authors pointed out that predictors of treatment failure have not been identified but stated that the most common underlying diagnosis for women who required a hysterectomy after ELA was adenomyosis.¹⁰

The reviewers concluded that ELA appears to be safe and effective for the treatment of menorrhagia in approximately 90% of patients. However, they added that due to the uncertainties regarding the long-term effects of ELA, and the difficulties in diagnosing endometrial cancer post-ablation, some physicians reserve Endometrial Laser Ablation for patients who have contraindications to hysterectomy.¹⁰

The AMA has not performed a DATTA for TCRE or TBEA.

American Association of Gynecologic Laparoscopists (AAGL): As per a communication with an AAGL spokesperson on June 17, 1999, the association has not issued a position statement or policy statement on endometrial ablation for menorrhagia.

Health Care Financing Administration (HCFA): As per a phone conversation with a HCFA spokesperson on July 7, 1999, there is no national coverage policy for endometrial ablation; coverage is decided by the individual Medicare carriers.

Agency for Health Care Policy and Research (AHCPR): The AHCPR is sponsoring several ongoing studies to examine hysterectomy and its alternatives. The following studies were initiated in 1996; however, no results are available to date.²

The Surgical Treatment Outcomes Project for Dysfunctional Uterine Bleeding is a 5-year study - To assess the effectiveness of hysterectomy compared with ELA for dysfunctional uterine bleeding in women who have failed drug therapy.
Study on Hysterectomy and Dysfunctional Uterine Bleeding - A multicenter study of 375 patients to compare the effectiveness and relative costs of hysterectomy compared with endometrial ablation versus GnRH agonist and oral contraceptive therapy for women with refractory dysfunctional uterine bleeding.
An unnamed study to compare the efficacy of medical and surgical treatments for abnormal uterine bleeding in premenopausal women with and without uterine fibroids.

Appendix III: Randomized Controlled Trials on the Efficacy of TCRE versus Hysterectomy

Key: AH, abdominal hysterectomy; CI, confidence interval; f/u, follow-up; HYST, hysterectomy; NA, not statistically analyzed; PID, pelvic inflammatory disease; pt(s), patient(s); SF36, short form 36; TCRE, transcervical resection of the endometrium; tx, treatment; US, ultrasound; VAS, visual analog scale; VH, vaginal hysterectomy wk, weeks(s); yr, year(s)

Authors, Study Objective, and Follow-up Time	Patient Inclusion Criteria and Number	Treatment Protocol	Results	Conclusions/Comments
Gannon et al. (1991) Royal Berkshire Hospital, Reading, UK; Regional Technical College, Galway, Republic of Ireland To compare outcome after TCRE or AH. f/u time, mean 12 months; range, 9-16 months.	54 pts awaiting AH for menorrhagia Pts with leiomyomata, endometrial or cervical neoplasia, ovarian disease, PID, or endometriosis were excluded. TCRE (n=26) (median age, 40 years); AH (n=28) (median age, 40 years).	4-6 wk pretreatment with 150 mg medroxyprogesterone followed by TCRE (no surgical details provided). f/u with transvaginal US, symptom diaries, and interviews.	3 pts withdrew after randomization; 1 in TCRE group and 2 in AH group. Hospital stay: median 1 day; range, 1-3 days for TCRE versus median 7 days; range, 5-12 days for AH (P<0.001). Analgesia time: median 0 days; range, 0-2 days for TCRE versus median 7 days; range, 2-56 days for AH (P<0.001). Recovery time: median 16 days; range, 5-62 days for TCRE versus median 58 days; range, 11-125 days for AH (P<0.001). After TCRE, 16/25 (64%) were amenorrheic; 4/25 (16%) had light menstrual periods lasting > 1 day; 1/25 (4%) became pregnant; 4/25 (16%) were tx failures due to persistent bleeding or pain and required repeat TCRE at > 6 months after tx. No TCRE pts required HYST during the f/u.	The length of hospitalization, recovery time, and need for analgesia were significantly lower in the TCRE pts compared with the AH pts; these findings remained significant in intent-to-treat analyses. No pts required HYST during the f/u. TCRE may be a useful alternative to AH for menorrhagia in the absence of pelvic pathology; however, longer term f/u and a larger number of pts are needed to fully evaluate the efficacy of this tx. Study limitations: Unclear how menorrhagia was assessed at baseline; unspecified menopausal status of pts; 24 pts were excluded from study for various reasons; small sample size; relatively short f/u time.
Dwyer et al. (1993) St. Michael's Hospital, Bristol, UK To compare outcome after TCRE or AH. f/u time, 4 months.	200 pts < 52 yr old awaiting AH for menorrhagia unresponsive to medical tx. Pts with uterine size > 12 wk gestation or presence of other pathology necessitating AH were excluded. TCRE (n=100) (mean age, 40.4 yr); AH (n=100) (mean age, 40.6 yr).	TCRE by electrocautery without preoperative endometrial thinning. f/u at 4 months by clinical examination and symptom diaries. Menstrual blood loss assessed by alkaline hematin method (n=50).	4 pts withdrew after randomization; 1 from TCRE group and 3 from AH group; 99 TCRE pts and 97 AH pts available for analysis. TCRE was not performed in 1 pt due to dense uterine adhesions, and 1 TCRE pt required AH for endometrial cancer; 2 AH pts required additional surgery. Menstrual blood loss > 80 ml per cycle: TCRE group (12/25; 51%) and AH group (13/25; 52%). Hospital stay: median 2 days; range, 1-8 for TCRE group versus median 6 days ; range, 5-10 days for AH group (P<0.001). Time off work, return to normal functioning, and return to sexual activity was shorter for TCRE pts versus AH pts (median 2 wk versus 11 wk; median 1 wk versus 4 wk; and median 3 wk versus 6 wk, respectively (P<0.001). Analgesia doses (> 1): 5/99 (5%) for TCRE group versus 96/97 (99%) for AH group (NA). At 4 months, 84/99 (85%) TCRE pts and 89/95 (94%) AH pts were satisfied or very satisfied. TCRE was unsatisfactory tx for 15/99 (15%) TCRE pts versus 6/95 (6.3%) AH pts (95% CI 1.1%-17.5%). 13/99 (13%) TCRE pts were amenorrheic; 76/99 (76%) were hypomenorrheic, and 9 (9%) had unchanged periods. Premenstrual symptoms such as dysmenorrhea improved in a higher percentage of AH pts than TCRE pts. 10/99 (10%) TCRE pts with unsatisfactory results were tx failures due to persistent menorrhagia or pain. 11/99 (11%) pts required repeat TCRE (n=7; 7%) or AH (n=4; 4%) by 4 mo.	Pt satisfaction was significantly higher after AH compared with TCRE; however, TCRE resulted in shorter hospital stay, quicker return to normal activities, and fewer complications. Tx failures after TCRE were related to persistent menorrhagia or dysmenorrhea. Longer term follow-up needed to further evaluate pt satisfaction. Study limitations: 16/216 (7.4%) women excluded from study entry; 4/200 (2%) drop-outs; objective data on blood loss of >80 ml available for very few pts in each group; no data on menopausal status; lack of preoperative endometrial thinning creates difficulties in comparing results with other studies that utilize thinning; inadequate f/u time.
Sculpher et al. (1996) Brunel University, Uxbridge, Middlesex, UK To assess longer term outcome since f/u at 4 months after TCRE or AH in pts who participated in Dwyer et al. (1993) trial. f/u time, mean, 2.8 yr; range, 0.3-3.8 yr.	196 pts from the trial by Dwyer et al. (1993) (see inclusion criteria above) 155/196 (79%) of the pts responded; 82 in TCRE group and 73 in AH group.	See above for tx protocol. SF36 and VAS questionnaires were sent to all pts who were randomized to TCRE or AH as part of the Dwyer et al. (1993) trial.	At f/u, 54/77 (70%) of the TCRE pts were experiencing bleeding. Pain: 46/72 (64%) TCRE group and 13/70 (19%) AH group. Worse premenstrual symptoms compared with preoperative symptoms: 13/71 (18%) TCRE group and 3/68 (4%) AH group Time off work for menstrual problems: 18/80 (23%) TCRE group and 3/71 (4%) AH group. On the SF36 form, the only significant difference between groups was for pain; AH pts reported fewer problems with pain compared with TCRE pts (P=0.01). No differences were observed between the TCRE (mean, 82.3) and AH (mean, 83.8) scores on the VAS (P=0.5). A higher percentage of AH pts (96%) were satisfied with tx compared with TCRE pts (79%) (P=0.002). A total of 10/99 (10%) TCRE pts required repeat TCRE, and 18/99 (18.2%) required AH.	A high percentage of TCRE pts experienced bleeding. Compared with AH pts, more TCRE pts had pain, worsening symptoms, and required time off from work, and fewer pts were satisfied with the tx results. A number of TCRE pts required retreatment during this longer term f/u period. The between-group difference in satisfaction over time (between 4 and 2 yr) increased from 9% to 17% demonstrating the need for long-term f/u after tx. Study limitations: possibility of bias since 17% of TCRE pts and 25% of AH pts were lost to f/u, and pts who respond may differ from pts who do not respond; lack of baseline data for SF36 for comparison.
Crosignani et al. (1997) University of Milan and Institute of Pharmacological Research, Milan, Italy To assess long term outcome after TCRE or VH. f/u time, 2 yr.	92 pts < 50 yr requiring VH for menorrhagia unresponsive to medical tx ; uterine volume < 12 wk gestation (< 380 ml at US); negative cervical cytology within past 12 months; no atypical hyperplasia on endometrial biopsy; no adnexal tumors. Pts with PID, endometriosis, stress incontinence, moderate-to-severe prolapse, clotting disorders, intrauterine device use, use of anticoagulants, unstable general condition, or presence of myomas > 3 cm in diameter or > 50% intramural extension were excluded. TCRE (n=45) (mean age, 46 yr) and VH (n=47) (mean age, 45 yr).	TCRE performed in early proliferative phase with rollerball and wire loop electrodes.	7 pts withdrew after randomization; 4 in TCRE group and 3 in VH group; 41 pts in TCRE group and 44 in VH group were analyzed. Hospital stay: median 1 day (range, 1-1 days) for TCRE group versus median 5 days (range, 5-6 days) for VH group (P<0.01). Return to normal activity: median 8 days (range, 6-8 days) for TCRE group versus median 13 days (range, 13-14.5 days) for VH group (P<0.01). Return to work: median 14 days (range, 11-16 days) for TCRE group versus median 30 days (range, 26-34 days) for VH group (P<0.01). 1/41 (2.4%) TCRE pt required VH for atypical hyperplasia, and 3/41 (7.3%) for recurrent menorrhagia or dysmenorrhea. 3/41 (7.3) TCRE pts and 5/44 (11.4%) VH pts were lost to follow-up by 2 yr. 8/35 (22.8%) of the remaining TCRE pts were amenorrheic, 4 (11.4%) had light spotting, 10 (28.7%) were hypomenorrheic, 12 (34.4%) had normal periods, and 1 (2.7%) remained menorrhagic. 9/38 (23.7%) TCRE pts were very satisfied with results, 24 (63.1%) were satisfied, 3 (7.9%) were uncertain, 2 (5.3%) were very dissatisfied. In comparison, 18/39 (46.1%) VH pts were very satisfied, 19 (48.7%) were satisfied, 2 (5.2%) were uncertain and none were dissatisfied (P=0.26 for difference between groups). Vitality (P=0.01) and social functioning (P=0.04) scores were higher, and anxiety scores (P=0.03) were lower for the VH group compared with TCRE group; however, no significant differences in sexual functioning were observed between the two groups.	Length of hospitalization, and the time to return to normal activities and work were significantly shorter for the TCRE pts compared with the VH pts. The overall tx failure rate was 10%. Pt satisfaction rates for TCRE were slightly higher for VH, and 5.3% of TCRE pts were very dissatisfied with tx results. VH pts did not display more psychological symptoms or sexual problems related to surgery compared with the TCRE pts. Study limitations: Method to assess preoperative menorrhagia unclear; no baseline data on quality of life indices; no data on menopausal status; some pts were excluded from study (26/118; 22%) or lost to follow-up by 2 yr (8/85; 9.4%).
O'Connor et al. (1997) Royal Free Hospital, London, U.K.; London School of Hygiene and Tropical Medicine, London, U.K. To compare outcomes after TCRE and HYST (AH or VH). f/u time, median, 2 yr.	202 pts with symptomatic menorrhagia requiring HYST; aged 30-50 yr; uterine size < 12 wk gestation; regular menstrual cycles lasting < 50% of 21-35 day cycle; normal endometrial histology within past 12 months; normal cervical smear within past 3 yr; completed childbearing. Pts with serious comorbidity; intermenstrual or postcoital bleeding; uterine size > 12 wk gestation; submucous fibroids > 5 cm; PID; endometriosis; prolapse; urinary symptoms; severe premenstrual or menopausal symptoms were excluded. TCRE (n=134) (mean age, 40.1 yr) and AH (n=68) (mean age, 39.4 yr).	Loop or rollerball resection with or without pharmacological endometrial thinning with f/u at 3 months, and 1, 2, and 3 yr.	15 TCRE pts and 11 HYST pts withdrew after randomization before receiving tx. 119 TCRE and 57 HYST pts received allocated tx; however, 3 TCRE pts and 1 HYST pt were lost to f/u, leaving 116 and 56 pts, respectively, available for 3 months f/u. Hospital stay: mean, 1.3 ± 1.2 days for TCRE group and mean, 6.3 ± 1.9 days for HYST group (P<0.001). Return to normal activities: mean, 1.9 ± 2.7 wk for TCRE group and mean, 4.6 ± 2.5 wk for HYST group (P<0.001). Return to work: mean, 2.9 ± 2.9 wk for TCRE group and mean, 7.4 ± 3.2 wk for HYST group (P<0.001). Resumption of sexual activity: mean, 3.9 ± 3.3 wk for TCRE group and mean, 5.9 ± 2.4 wk for HYST group (P<0.001). Amenorrheic: 48/104 (46%) TCRE pts 1 yr and 11/54 (21%) at 3 yr. Pt satisfaction at 1 yr: 90/104 (87%) TCRE pts and 42/46 (91%) HYST pts (P=0.59). Pt satisfaction at 3 yr: 46/54 (85%) TCRE pts and 27/28 (96%) HYST pts (P=0.16). 14/104 (13.5%) TCRE pts required repeat TCRE during f/u and 12/104 (11.5%) required HYST. The relative risk for subsequent surgery was similar between the TCRE and HYST groups (0.46 [95% CI 0.2-1.1]; P=0.053) All psychological and social scores improved to a similar degree after TCRE or HYST (P<0.01).	The hospital stay and length of recovery was significantly shorter in the TCRE group compared with the HYST group. TCRE improved symptoms in high percentage of pts. Pt satisfaction rates were slightly higher in HYST group, but this did not reach statistical significance. Study limitations: the 202 pts selected for study represented only 25% of pts with menorrhagia; drop-out rate of 13.4% in TCRE group and 14.7% in HYST group; loss of pts to f/u resulting in small sample size for later f/u; no distinction between AH and VH in comparisons with TCRE; 75% pts preferred one procedure over the other and it is unclear how this may have affected results.

Appendix IV: Clinical Studies on the Efficacy of ELA

Key: CI, confidence interval; ELA, endometrial laser ablation; f/u, follow-up; GnRH, gonadotropin releasing hormone; Nd:YAG, neodymium yttrium-aluminum-garnet; pt(s), patient(s); RR, relative risk; TCRE, transcervical resection of the endometrium; tx, treatment; US, ultrasound; wk, weeks(s); yr, year(s)

Authors, Study Design and Objective, and Follow-up Time	Patient Inclusion Criteria and Number	Treatment Protocol	Results	Conclusions/Comments
Erian (1994) Farnborough Hospital, Orpington, Kent, U.K. Prospective case series to assess the outcome after ELA. f/u time, up to 1 yr.	1866 pts with severe menorrhagia unresponsive to medical tx who were candidates for hysterectomy or for whom hysterectomy was contraindicated; endometrial neoplasm excluded by curettage or hysteroscopy. Mean age, not given.	Preoperative administration of 600 mg danazol daily for 6 wk followed by ELA with Nd:YAG laser.	At > 1 yr f/u, 1043/1866 (56%) pts were amenorrheic; 353 (19%) were hypomenorrheic, 348 (18%) had normal periods; and 122 (7%) were tx failures with persistent menorrhagia. The percentage of pts with amenorrhea increased with increasing age: 2/35 (6%) of pts aged 20-29 yr compared with 105/112 (97%) of pts aged 50-56 yr. 84/1866 (4.5%) pts had repeat ELA and 33 (1.8%) had a hysterectomy following tx failure; the subsequent tx for 5/1866 (0.3%) was undetermined. Adenomyosis was diagnosed in 27/33 (81.8%) pts who underwent hysterectomy. 34/84 (40.5%) repeat ELA pts were amenorrheic; 17 (20.2%) were hypomenorrheic; 6 (7.1%) had normal periods, and 27 (32%) were tx failures with persistent menorrhagia.	In this pt group, ELA with preoperative endometrial thinning was an effective therapy for severe menorrhagia in the absence of underlying pathology. 22% of ELA failures occurred in pts with adenomyosis, and tx was less successful in younger pts. Repeat ELA was less successful than primary tx. Study limitations: lack of control group and direct comparison with hysterectomy; unclear how menorrhagia was assessed; no data on menopausal status; inadequate f/u time to assess long-term outcome or complications; f/u data only for 1866/2342 (80%) pts participating in trial
Garry et al. (1995) South Cleveland Hospital, Middlesbrough, Cleveland, U.K. Prospective case series to assess the outcome after ELA. f/u time, mean, 15 months; range, 6-42 months.	524 pts with severe menorrhagia whose childbearing was completed and who used reliable contraception; uterine size < 12 wk gestation. Pts with endometrial cancer or hyperplasia, uterine size > 12 wk gestation, presence of leiomyomas > 2 cm in diameter were excluded. Mean age, 43 yr; range, 20-59 yr.	Preoperative administration of 200 mg danazol daily for > 4 wk followed by ELA with Nd:YAG laser.	501/524 (96%) pts completed questionnaires. 75/524 (14.3%) pts had repeat ELA. 34/501 (6.8%) had a hysterectomy (mean, 11 months post-ELA). Among 467 pts who did not undergo hysterectomy, 135 (28.9%) were amenorrheic and 332 (71.1%) continued to menstruate. 309 (66.2%) were hypomenorrheic, 21 (4.5%) unchanged blood loss, 1 (0.2%) had increased blood loss, and 1 (0.2%) had no response. Among 146 pts with dysmenorrhea at baseline, 86 (58.9%) reported an improvement, 48 (32.9%) had no change, and 12 (8.2%) had increased pain. 6/467 (1.3%) developed dysmenorrhea after ELA. 418/501 (83.4%) pts described the tx as successful; and 83 (16.6%) considered it to be a failure. 85/501 (17%) retrospectively preferred hysterectomy to ELA. Pt satisfaction with ELA was associated with increasing age (P=0.001); this relationship remained significant in a multiple regression analysis (P=0.0001). Among the 34 pts who required hysterectomy, 19 (56%) had underlying conditions: 7 had endometriosis, 10 had leiomyomas, and 2 had adenomyosis.	ELA had a high success rate, and a low rate of complications in this pt group; however, 14.3% required repeat ELA, and 6.8% required a subsequent hysterectomy. Over half of the pts who required hysterectomy after ELA had an underlying condition that may have contributed to tx failure. Pts with such conditions may not benefit from ELA. The tx was more successful for older pts. Study limitations: lack of controls and direct comparison with hysterectomy, loss of 23 (4.4%) pts to f/u; no objective measurement of menstrual blood loss; no data on menopausal status.
Garry et al. (1996) South Cleveland Hospital, Middlesbrough, Cleveland, U.K. Randomized controlled trial to compare efficacy of danazol and goserelin for endometrial thinning prior to ELA. f/u time, 6 months.	160 pts > 20 yr with dysfunctional uterine bleeding who required hysterectomy. Pts who were pregnant, breast-feeding, or who had concomitant illness, who had taken oral contraceptives within 1 months, or who had taken danazol or any GnRH analogue within 6 months were excluded. Mean age ranged from 39.8-42.1 yr	Pts were randomized (40 per group) to one injection of goserelin (3.6 mg) with surgery at 28 days (GOS1); two injections of goserelin (3.6 mg each) with surgery at 56 days (GOS2); 800 mg danazol daily for 28 days (DAN1); or 800 mg danazol daily for 56 days (DAN2) followed by ELA. Endometrial thickness was determined by vaginal US and endometrial biopsy at surgery.	6/160 (3.8%) pts withdrew after randomization; 15/154 (9.7%) remaining pts dropped out of the study. At a 6 months f/u, 9/37 (24%) of the GOS1, 9/37 (24%) of the GOS2, 3/34 (9%) of the DAN1, and 1/31 (3%) of the DAN2 group were amenorrheic (P=0.005) for the difference between the goserelin and danazol groups). 19/37 (51%) of the GOS1, 20/37 (54%) of the GOS2, 15/34 (44%) of the DAN1, and 17/31 (55%) of the DAN2 groups were hypomenorrheic. 7/37 (19%) of the GOS1 group, 8/37 (22%) of the GOS2 group, 14/34 (41%) of the DAN1 group, and 12/31 (39%) of the DAN2 group had normal menstrual flow. 2/37 (5%) of the GOS1 group, 0% of the GOS2 group, 2/34 (6%) of the DAN1 group, and 1/31 (5%) of the DAN2 group were not improved. Overall, 57/74 (77%) of the goserelin group, and 36/65 (54%) of the danazol group were amenorrheic or hypomenorrheic (P<0.001) at 6 months. The odds ratio for overall improvement in menorrhagia for the goserelin groups versus the danazol groups was 3.09 (95% CI 1.59-6.02) (P<0.001). 35/37 (95%) of the GOS1 group, 35/37 (95%) of the GOS2 group, 32/34 (94%) of the DAN1 group, and 29/31 (94%) of the DAN2 group were satisfied with the tx results, while 5%, 5%, 6%, and 6%, of these groups, respectively, were dissatisfied.	Menorrhagia was significantly improved in the goserelin groups compared with the danazol groups at 6 months after ELA. Pt satisfaction rates were high in both groups but the tx failure rate ranged from 3% to 5%. Due to the short f/u time, it is unknown if any of the pts eventually required repeat ELA or hysterectomy. Study limitations: loss of pts to f/u (21/160, 13%); inadequate f/u time; no data on menopausal status.
Bhattacharya et al. (1997) Aberdeen Royal Infirmary, University of Aberdeen, and Aberdeen Maternity Hospital, Aberdeen, U.K. Randomized controlled trial to compare efficacy of ELA with TCRE f/u time, 12 months.	372 pts < 50 yr old; menorrhagia due to dysfunctional uterine bleeding; < 100 kg body weight; uterine size < 10 wk gestation; normal endometrial histology. ELA (n=188) (mean age, 40.4 yr) or TCRE (n=184) (mean age, 40.9 yr).	Preoperative endometrial thinning with one injection of 3.6 mg goserelin 5 wk prior to ELA with Ng-YAG laser or TCRE (rollerball coagulation and loop resection).	332/372 (89%) and 321/372 (86%) were available for f/u at 6 months and 12 months, respectively. After randomization, 3/188 (2%) of the ELA group had hysterectomy, and 3/184 (1.6%) of the TCRE group had hysterectomy or no tx. Hospital stay: median, 2 days (range, 1-3 days) for ELA and TCRE groups. By 12 months, 30/185 (16%) ELA pts (11% repeat ELA and 5% hysterectomy), and 36/181 (20%) TCRE pts (6% repeat TCRE and 14% hysterectomy) had been retreated (95% CI -15% to -3%) for difference between groups). At 12 months, 37/160 (23%) ELA pts versus 32/146 (22%) TCRE pts were amenorrheic; 79/160 (49%) ELA pts versus 68/146 (46%) TCRE pts were hypomenorrheic; 8/160 (5%) ELA pts versus 6/146 (4%) TCRE pts had no change, and 1/160 (1%) ELA pts versus 1/146 (1%) TCRE pts had heavier periods. Dysmenorrhea improved in similar numbers of pts in both groups (40/160; 35% for ELA and 50/146;48% for TCRE. In both groups, menstrual symptoms and anxiety and depression levels were improved after surgery (P<0.001). 114/166 (69%) ELA pts and 102 (155) (66%) TCRE pts were very satisfied, and 34 (21%) and 38 (25%), respectively, were moderately satisfied after tx. 12/166 (7%) ELA pts and 9/155 (6%) TCRE pts were dissatisfied, and 5 (3%) and 5 (3%) were very dissatisfied, respectively [(P=0.8) for differences between groups].	Menorrhagia improved to a similar degree in both ELA and TCRE pts with no significant differences between groups. Pt satisfaction rates were high in both groups. Overall, more TCRE pts than ELA pts (20% versus 16%) required repeat tx at 12 months after initial tx but the difference was not statistically significant. Significantly more ELA pts required repeat ablation, while significantly more TCRE pts required hysterectomy. The results of this study indicate that these tx have a similar level of efficacy. Study limitations: loss of pts to f/u, inadequate f/u time; lack of objective measure of blood loss and data on menopausal status.
Phillips et al. (1998) South Cleveland Hospital, Middlesbrough, Cleveland, U.K. Prospective case series to assess the risk of hysterectomy after ELA. f/u time, 26.5 to 76.8 months. Note: Most pts were subjects in the Garry et al. (1995) trial.	873 consecutive pts (1000 ELA) with menorrhagia whose childbearing was completed; uterine size < 12 wk gestation. Pts with endometrial cancer or hyperplasia, uterine size > 12 wk, large, submucous fibroids unsuitable for hysteroscopic surgery were excluded. Mean age, 41.7 yr.	Preoperative endometrial thinning with 600 mg danazol daily for > 4 wk or 3.6 mg goserelin for 1-3 months or both followed by ELA with Nd:YAG laser. Endometrial polyp (n=94) or submucous fibroid (n=85) were resected during ELA. Pts were discharged from f/u at 6 months if tx was satisfactory. Repeat ELA or hysterectomy were offered if tx was unsatisfactory or if referred by general practitioner.	746/873 (85.4%) pts had single ELA; 124 (14.2%) had two tx, and; and 3 (0.4%) had three tx. 87.3% response rate (762/873) to questionnaire. 179/873 (20.5%) had underlying intrauterine lesion. 249/663 (37.6%) pts who did not undergo hysterectomy, and whose menstrual status was known were amenorrheic; 358/414 (86.3%) of those who continued to menstruate had lighter periods. 592/663 (89.3%) were satisfied with tx results. Survival curve analysis demonstrated that the hysterectomy rate was 21% over a f/u of 6.5 yr (95% CI 16%-27%). By multivariate regression analysis, the risk of hysterectomy was significantly increased with repeat ELA (RR=2.93; 95% 1.59-5.40; P=0.002) but decreased for pts with intrauterine pathology (polyps, fibroids, and uterine malformations) (RR=0.26; 95% CI 0.08-0.86; P=0.008). Pt age, dysmenorrhea, length of uterine cavity, operating time, amount of fluid absorbed, endometrial preparation, and premenstrual syndrome did not influence the risk of hysterectomy.	Over three-quarters of the pts did not require a hysterectomy after ELA for menorrhagia during the f/u period. The risk of hysterectomy was significantly increased (by 71%) in pts who required repeated ELA whereas the presence of intrauterine pathology decreased the risk by 74%. ELA may improve menorrhagia when an underlying lesion is causing the bleeding, and may be less effective when the cause of bleeding is unknown. Study limitations: lack of controls and direct comparison to hysterectomy; loss of pts to f/u; no objective measurement of menstrual blood loss; lack of data on menopausal status of pts
Everett (1999) University of Oklahoma Health Science Center, Oklahoma City, OK Prospective case series to assess the outcome after ELA. f/u time, 12-72 months.	86 pts with menorrhagia who were candidates for hysterectomy. Pts with uterine size > 10 cm or > 10 wk gestation were included. Mean age, 41.9 yr.	Preoperative endometrial thinning with 3.75 mg leuprolide acetate on day 21 to 25 of cycle with second dose administered at 3-4 wk. ELA performed 3 wk later with Nd:YAG laser.	Overall, 70/86 (81.3%) pts were amenorrheic; 15 (17.4%) had light spotting; 1 (1.3%) was hypomenorrheic. At 1 yr, 19/21 (91%) of pts were amenorrheic and 2/21 (9%) had light spotting; 0% were hypomenorrheic or tx failures. At 2 yr, 9/12 (75%) were amenorrheic; 3/12 (25%) had light spotting; 0% were hypomenorrheic or tx failures. At 3 yr, 14/15 (93%) were amenorrheic; 1/15 (7%) had light spotting; 0% were hypomenorrheic or tx failures. At 4 yr, 16/22 (73%) were amenorrheic; 5/22 (23%) had light spotting; 1/22 (4%) was hypomenorrheic; 0% were tx failures. At 5 yr, 12/16 (75%) were amenorrheic; 4/16 (25%) had light spotting; 0% were hypomenorrheic; and 0% were tx failures.	ELA was an effective tx for menorrhagia in all pts who were available for f/u. Depending on the f/u time, the percentage of amenorrheic pts ranged from 73% to 93%; the percentage with light spotting ranged from 9% to 25%; and the percentage who were hypomenorrheic ranged from 0% to 4%. There were no tx failures and few complications. Study limitations: lack of controls and direct comparison with standard tx; varying f/u times (only 16 available at 5-yr f/u); no long-term outcome data on retreatment or hysterectomy.

Appendix V: Clinical Studies on the Efficacy of TBEA

Key: f/u, follow-up; months, month(s); PID, pelvic inflammatory disease; pt(s), patient(s); TBEA, thermal balloon endometrial ablation; TCRE, transcervical resection of the endometrium; tx, treatment; wk, weeks(s); yr, year(s)

Authors, Study Design and Objective, and Follow-up Time	Patient Inclusion Criteria and Number	Treatment Protocol	Results	Conclusions/Comments
Singer et al. (1994) Multicenter international study Prospective case series to assess the efficacy and safety of TBEA. f/u time, range, 6-34 months.	18 premenopausal pts with menorrhagia unresponsive to medical tx or curettage who were candidates for hysterectomy or endometrial ablation; negative cervical cytology; uterine cavity < 12 cm in length; no endometrial hyperplasia or cancer; no submucous fibroids.	No preoperative endometrial thinning. TBEA with ThermaChoice under general (n=16) or local (n=2) anesthesia followed by overnight hospitalization. Hysteroscopy (n=7) or laparoscopy (n=4) performed for monitoring.	7/18 (38.9%) pts had light bleeding during f/u; 8/18 (44%) were amenorrheic or had spotting at final f/u. Overall, 15/18 (83%) pts were amenorrheic or hypomenorrheic after TBEA. 2/18 (11.1%) pts subsequently required hysterectomy for dysmenorrhea related to intrauterine scarring at 34 months (n=1) or persistent menorrhagia at 9 months (n=1) after TBEA. 1/18 (5.6%) had an endometrial resection for bleeding at 9 months post-TBEA.	TBEA was an effective tx for menorrhagia and had a low complication rate in this pt group; however, nearly 17% failed tx and subsequently required hysterectomy or endometrial resection. Areas of scarring and normal endometrium were demonstrated on biopsy. Study limitations: lack of controls and direct comparison with hysterectomy; case series design with potential for selection bias; very small sample size; varying and inadequate f/u times for individual pts; unclear how menorrhagia was assessed. Study funded by manufacturer, Gynelab Product Inc., Raleigh, North Carolina.
Vilos et al. (1996) University of Western Ontario, London, Ontario, Canada Prospective case series to assess the efficacy and safety of TBEA f/u times, 1 wk, 3, 6, 12, and 18 months.	30 pts with menorrhagia affecting quality of life, and chronic fatigue from iron-deficiency anemia (n=9) who no longer desired medical tx and who requested endometrial ablation or hysterectomy. Pts with submucous fibroids, polyps, endometrial cancer or precancer, uterine cavity length > 12 cm, or who wished to maintain fertility were excluded. Pts who desired elimination of menses were encouraged to have a hysterectomy. Mean age, 38 yr; range, 27-46 yr.	Preoperative endometrial thinning with danazol 100-200 mg daily for 1-2 months (n=9). Outpatient TBEA with ThermaChoice under general (n=9) or local and neuroleptic anesthesia (n=21) followed by discharge within 4 hours.	At 6 months, 25/30 (83%) pts were improved: 1 (3%) pt was amenorrheic; 4 (13%) had spotting (< 1 pad/day); 13 (43%) were hypomenorrheic (< 3 pads/day) and 7 (23%) had moderate improvement (< 12 pads/day); 5 (17%) pts failed tx. At 12 and 18 months, similar patterns of improvement were observed; however, 2 pts with moderate improvement had recurrent menorrhagia. Thus, the failure rate increased to 7/30 (23%) by this f/u. By 12-18 months, 13/30 (43%) pts with no improvement, moderate improvement, or hypomenorrhea had a hysteroscopic endometrial ablation or hysterectomy [2 (6.7%) hysterectomy and 11 (36.7%) endometrial ablation]. Age, parity, uterine size, or operative technique did not appear to be related to tx failure; however, no clinical details or statistical analysis of these variables was provided.	While a high percentage (83%) of pts were considered improved by 6 months after TBEA, only 3% were amenorrheic, and 23% of pts had only moderate improvement. The tx failure rate increased by the 12-18 months f/u, and a total of 43% of pts required repeat surgery. The complication rate was relatively low. The outcome measures are questionable since pts considered to be moderately improved pts still demonstrated excessive bleeding. Study limitations: feasibility study that lacked controls and direct comparison with standard tx; potential selection bias since pts themselves requested surgery; small sample size.
Fernandez et al. (1997) Antoine Béclère Hospital, Clamart, France Prospective case series to assess the efficacy and safety of TBEA. f/u time, median, 13.9 months; range, 4-24 months.	18 pts with menorrhagia, no uterine pathology, and normal endometrial histology who failed or were unwilling to continue medical tx. Pts with submucous fibroids, polyps, endometrial cancer or precancer; uterine cavity length of > 12 cm; or who wished to maintain fertility were excluded. Menorrhagia assessed by pad count, number of days per cycle, and dysmenorrhea score. Mean age, 47.1 yr; range, 39-55 yr.	Preoperative endometrial thinning with oral progestogens (n=12) and suction curettage (n=14). Outpatient TBEA with ThermaChoice under local anesthesia (paracervical block) with discharge after 6-10 hours.	Mean number of pads per day: Decreased from preoperative 10.2 ± 1.6 to postoperative 3.2 ± 2.7 (P<0.0001). Mean days per cycle: Decreased from preoperative 8.3 ± 3.4 to postoperative 3 ± 2.4 (P<0.0001). Mean number of pads per cycle: Decreased from preoperative 86 ± 40.4 to postoperative 14.3 ± 12.4 (P< 0.0001). At 3 months post-TBEA, 4/18 (22%) pts were amenorrheic; 3 (17%) were hypomenorrheic; 11 (61%) were eumenorrheic; and 0% had persistent menorrhagia. At 6 months, 4/16 (25%) pts were amenorrheic; 2 (13%) were hypomenorrheic; 10 (62%) were eumenorrheic; and 0% had persistent menorrhagia. At > 12 months, 6/11 (55%) pts were amenorrheic; 1 (9%) were hypomenorrheic; 4 (36%) were eumenorrheic; and 0% had persistent menorrhagia. No pts required retreatment during f/u.	Pad use and the number of days per cycle were significantly decreased after TBEA. After TBEA, a high percentage of pts were improved, no pts had persistent menorrhagia at any f/u, and no pts required additional tx; however, the true response rate is difficult to assess since only 61% of pts were available for the final f/u. No major complications were reported. Study limitations: lack of controls and direct comparison with standard tx; very small sample size; loss of 7 (39%) pts to f/u; role of endometrial thinning in outcome is unclear.
Amso et al. (1998) International Collaborative Uterine Thermal Balloon Working Group Multicenter prospective case series to assess the safety and efficacy of TBEA. f/u times, 3, 6, and 12 months.	296 pts with severe menorrhagia who failed or were unwilling to continue medical tx and were candidates for hysterectomy or endometrial ablation; normal endometrial histology. Pts with uterine pathology, atypical endometrial hyperplasia, suspected genital tract infection or cancer, uterine cavity length > 12 cm, previous endometrial ablation, or who wished to maintain fertility were excluded. Mean age, 41 yr; range, 26-54 yr.	Varying preoperative endometrial thinning regimens (drugs, follicular phase timing, curettage, and hormonal manipulation) (n=251; 85%) or no preparation (n=45; 15%). TBEA with ThermaChoice performed under general anesthesia (n=180); intravenous sedation with paracervical block (n=54); intravenous sedation (n=9); paracervical block (n=34). All pts were discharged within 24 hours.	Mean number of pads per day (n=171): Decreased from preoperative 9.2 ± 3.2 to postoperative 3.4 ± 2.7 (P<0.0001). Mean number of days per cycle (n=244): Decreased from preoperative 9.6 ± 5.4 to postoperative 4.1 ± 3.1 (P<0.0001). Mean number of pads per cycle (n=173): Decreased from preoperative 86 ± 51.9 to postoperative 18.6 ± 28.2 (P<0.0001). At 6 months, 40/291 (14%) pts were amenorrheic; 39 (13%) had spotting; 102 (35%) were hypomenorrheic; 84 (29%) were eumenorrheic; and 26 (9%) were tx failures with persistent menorrhagia. At 12 months, 25/163 (15%) pts were amenorrheic; 27 (17%) had spotting; 50 (31%) were hypomenorrheic; 41 (25%) were eumenorrheic; and 20 (12%) had persistent menorrhagia. Dysmenorrhea scores were improved at > 3 months f/u (P<0.0001). 28/296 (10%) pts required additional surgery: 15 (5.1%) had hysterectomies and 13 (4.4%) had endometrial ablations. In logistic regression analyses, several variables were related to tx outcome. At 6 mo, tx failure was related to higher uterine volumes (P=0.02), preoperative curettage (P<0.01), and higher preoperative pad use per cycle (P<0.05). At 12 mo, higher patient age was related to a higher likelihood of success (P=0.01).	In pts for whom data were available, menstrual flow, indicated by the number of pads per day and days per cycle, and dysmenorrhea decreased significantly after TBEA. Although the overall rate of tx success was high at 12 months (88%), data were missing for almost half of the pts, thus, the true response rate is difficult to determine. Tx failure was significantly related to higher preoperative menstrual blood loss. 10% of pts required additional surgery due to tx failure. TBEA had a low complication rate. Study limitations: lack of controls and direct comparison with standard medical or surgical tx; data on preoperative and postoperative menstrual flow patterns missing for some pts; data missing for a number of pts at various f/u [available for only 163 (55%) of pts at 12 months].
Meyer et al. (1998) Multicenter U.S. study funded by Gynecare Inc., Menlo Park, California Randomized controlled trial to compare TBEA with TCRE. f/u time, 12 months.	275 premenopausal pts > 30 yr old with menorrhagia for > 3 months; failed medical tx; score of > 150 derived from menorrhagia diary indicating excessive bleeding; normal cervical cytology and endometrial histology within past 6 months; normal uterine cavity as assessed by hysteroscopy, hysterosalpingography, or ultrasound that sounded between 4 and 10 cm; no desire for future fertility; willing to use contraception for 3 yr post-ablation. Pts with submucous myomas, suspected genital tract infection or cancer, or previous ablation were excluded. Mean age: TBEA, 40.2 yr; range, 30-51 yr; TCRE, 40.9 yr; range, 29-50 yr.	No preoperative endometrial thinning. Random assignment to TBEA with ThermaChoice (n=128) under general anesthesia in 53% of pts or TCRE by rollerball electrosurgery (n=117) under general anesthesia in 84% of pts. All pts were discharged on day of procedure.	20/275 (7.3%) pts withdrew after randomization before tx (n=15) or were not treated under the protocol (n=5; 2 TCRE and 3 TBEA). 245/275 (89.1%) pts were available for f/u. Prior to 12 months f/u, 2/128 (1.6%) TBEA pts and 3/117 (2.6%) TCRE pts had hysterectomy. At 12 months f/u, data were available for 125 pts from TBEA group and 114 pts from TCRE group. Mean menstrual diary scores decreased 85.5% in the TBEA group and 91.7% in the TCRE group. 1 (0.8%) TBEA pt and 1 (0.9%) TCRE pt required hysterectomy for persistent menorrhagia. Menorrhagia scores decreased to normal in 80.2% of TBEA group and 84.3% of TCRE group. 19/125 (15.2%) pts in TBEA group and 31/114 (27.2%) pts in TCRE group were amenorrheic at 12 months (P<0.05). 107/125 (85.6%) pts in TBEA group and 99/114 (86.7%) of pts in the TCRE group were very satisfied with tx results. 88/125 (70.4%) pts in TBEA group and 86/114 (75.4%) pts in TCRE group had improved dysmenorrhea.	Although a significantly higher number of pts achieved amenorrhea after TCRE compared with TBEA, menorrhagia improved in both groups, and pt satisfaction rates were high for both tx. Both groups reported a similar degree of improvement in quality of life measures, as well as a reduction in dysmenorrhea and premenstrual syndrome. TBEA had fewer intraoperative complications. Study limitations: loss of a number of pts to f/u; short f/u period; lack of intent-to-treat analysis.
Lissak et al. (1999) Lady Davis Carmel Medical Center and Technion School of Medicine, Haifa, Israel and Chaim Shiba Medical Center, Tel Aviv University, Ramat-Gan, Israel Randomized controlled trial to compare the efficacy of TBEA with and without endometrial thinning, and to assess its efficacy in pts with myomas or prior cesarean sections. f/u time, 6 months.	30 pts with perimenopausal menorrhagia who were unresponsive to medical tx and were candidates for endometrial ablation or hysterectomy; age 30-55 yr; good health; score of 150 on menstrual diary; anovulatory pts who failed oral contraceptive tx; normal endometrial biopsy and cervical cytology within 6 and 12 months, respectively; uterine cavity > 4 cm and < 12 cm. Pts with active or chronic PID; undiagnosed genital bleeding; atypical endometrial hyperplasia; uterine polyps; septate uterus; or adnexal pathology; drug or alcohol abuse; latex allergy; pregnancy; or desire to become pregnant were excluded. Mean age: No GnRH agonist tx (45.6 ± 1.47 yr); GnRH agonist pretreatment (48.7 ± 1.23 yr).	Randomized to no GnRH agonist pretreatment and immediate tx (n=17) or GnRH pretreatment with single intramuscular injection of 3.75 mg of decapeptyl (n=13) followed by TBEA by ThermaChoice within 4 to 6 wk. Submucosal myomas were resected before ablation (n=2); pts with prior cesarean sections (n=6) were screened for the presence of fenestrations through surgical scar.	At the 6 months f/u, no significant differences in outcome were found between the two groups. 4/17 (23.5%) of the no GnRH group and 3/13 (23%) of the GnRH group were amenorrheic. 12 (70.5%) of the no GnRH group and 8 (61.5%) of the GnRH group were hypomenorrheic. 1 (5.8%) of the no GnRH group and 2 (15.3%) of the GnRH group were eumenorrheic. Mean days of flow per cycle at 6 months: no GnRH group, 2.1 ± 0.75 versus GnRH group, 1.8 ± 0.42. Pt satisfaction: no GnRH group, 15 (88%) and GnRH group, 11 (92%). Subsequent hysterectomy: 2/17 (11.8%) of the no GnRH group, and 1/13 (7.7%) of the GnRH group.	No significant differences in tx outcome were observed between pts who received GnRH agonist pretreatment and those who did not. 90% of pts initially responded to TBEA with reduced menstrual flow or amenorrhea, and the satisfaction rate ranged from 88% to 92%; however, 10% of pts subsequently required hysterectomy. No intraoperative complications occurred in any pt, those who underwent prior surgery for myomas or who had cesarean sections. Study limitations: small sample size and lack of statistical power; inadequate f/u time; no data on when hysterectomies were performed.

Appendix VI: Randomized Controlled Trials on the Efficacy of EA versus Standard Treatment

Key: ELA, endometrial laser ablation; ENDO, endometrial ablation or resection; f/u, follow-up; HYST, hysterectomy; Nd:YAG, neodymium yttrium-aluminum-garnet; pt(s), patient(s); TCRE, transcervical resection of the endometrium; tx, treatment; wk, weeks(s); yr, year(s)

Authors, Study Design and Objective, and Follow-up Time	Patient Inclusion Criteria and Number	Treatment Protocol	Results and Complications	Conclusions/Comments
Pinion et al. (1994) Aberdeen Royal Infirmary and University of Aberdeen, Aberdeen, U.K. Randomized controlled trial to compare the efficacy and safety of ENDO (ELA and TCRE) with hysterectomy. f/u time, 12 months. Note: See study by Grant et al. (1999) for longer term f/u of these pts.	204 pts < 50 yr old who weighed < 100 kg, with dysfunctional uterine bleeding; uterine size < 10 wk gestation; normal endometrial histology; candidate for hysterectomy. Mean age: HYST, 40.3 ± 5.2 yr; ENDO, 44.8 ± 13.7 yr.	Preoperative endometrial thinning with goserelin for 5 wk followed by ELA with Nd:YAG laser (n=53) or TCRE by rollerball coagulation and wire loop resection (n=52). Hysterectomy (n=99) (87 abdominal and 12 vaginal).	6/204 pts withdrew or refused allocated tx after randomization; 1 pt was withdrawn after HYST due to suspected cancer. There were 97 pts in HYST group and 105 in ENDO group. Mean hospital stay: HYST, 7.3 nights; ENDO, 2.5 nights (P<0.001). Postoperative pain: HYST, 1-2 wk; ENDO, < 1 wk (P<0.001). Time to recovery: HYST, 2-3 months; ENDO, 2-4 wk (P<0.001). Time to return to work: HYST, 2-3 months; ENDO, 2-4 wk (P<0.001). 5/99 (5.2%) pts in HYST had major complications compared with 1/105 (1%) in the ENDO group. Postoperative infection was lower in the ENDO group (difference, -32%; 95% CI -20% to -40%) (P<0.001). At 12 months, 21/96 (22%) of the ENDO pts were amenorrheic; 59 (62%) were hypomenorrheic; 3 (3%) were unchanged; and 13 (14%) had a hysterectomy, 11 (11.5%) for persistent menorrhagia (3 had fibroids, 4 had adenomyosis, and 4 had no evidence of disease). 11 (11.5%) had repeat ENDO for persistent menorrhagia between 6 and 12 months. At 12 months, menstrual bleeding scores had decreased from 26.8 to 4.5 (95% CI 3.2 to 5.7) and pain scores had decreased from 10.0 to 3.6 (95% CI 2.4 to 4.7). At 12 months, 79/89 (89%) in the HYST group and 75/96 (78%) in the ENDO group were very satisfied with the tx results (P<0.05) and 85.89 (95%) in the HYST group and 86/96 (90%) in the ENDO group believed that tx improved symptoms.	The rate of intraoperative complications was lower, and the length of hospitalization, and time to recover and return to work were significantly shorter in the ENDO group compared with the HYST group. While pt satisfaction was significantly higher in the HYST group, the surgery is associated with a higher rate of complications and a longer recovery period. Study limitations: no objective measure of menstrual blood loss; loss of pts to f/u; inadequate f/u time.
Grant et al. (1999) Aberdeen Endometrial Ablation Trials Group Aberdeen, U.K. Randomized controlled trial to compare the efficacy and safety of ENDO (ELA and TCRE) with hysterectomy. f/u time, 4 yr. Note: f/u of Pinion et al. (1994) study.	See above.	See above.	151/204 (74%) pts responded to questionnaire 4 yr after tx (73 HYST and 78 ENDO). Additional surgery for menstrual problems among 4-yr responders: 26/76 (34%) of ENDO group; 14/76 (18%) had hysterectomy, 2 (3%) had ablation and hysterectomy, 9 (12%) had ablation, and 1 (1%) had two ablations. No (0%) pts in the HYST group had additional surgery. Additional surgery for menstrual problems among all randomized pts: 39/102 (38%) of ENDO group and 1/95 (1%) of HYST group; 20 (20%) of ENDO group had hysterectomy, 4 (4%) had endometrial ablation and hysterectomy, 13 (13%) had endometrial ablation, and 2 (2%) had two endometrial ablations. Retreatment was required for bleeding (30%), pain (3%), or bleeding and pain (5%). At 4 yr, 33/73 (45%) of the ENDO pts were amenorrheic; 29 (40%) were hypomenorrheic; 9 (12%) had spotting; and 2 (3%) had no change. At 4 yr, 14/77 (18%) of the ENDO pts and 10/68 (15%) of the HYST pts had pelvic pain. At 4 yr, significantly fewer of the HYST pts experienced the premenstrual symptoms of breast tenderness (39% versus 62%); bloatedness (30% versus 45%); and irritability (25% versus 45%) compared with the ENDO group. 80% of the ENDO pts and 89% of the HYST pts expressed satisfaction with the results (P=0.10); however, only 69% of pts who required retreatment reported being satisfied with tx results. A higher percentage of HYST pts reported that their general health was much better (67%) compared with the ENDO group (38%) (P<0.001).	While a retreatment rate of 38% after endometrial ablation was found in this group of pts, 76% of these pts avoided a hysterectomy during the 4-yr f/u. In addition, the majority of pts in both tx groups reported being satisfied with the tx results; satisfaction rates were lower among pts requiring retreatment. The amenorrhea rate (45%) was higher than that reported at 12 months (22%) [(See Pinion et al. (1944)]. While premenstrual symptoms were significantly less common among the HYST group, similar numbers of pts had pelvic pain despite tx. Study limitations: loss of pts to f/u and possible bias; although responder analysis showed no differences in prognostic variables between responders and non-responders, fewer responders required retreatment compared with non-responders.
Cooper et al. (1997) Aberdeen Royal Infirmary, University of Aberdeen, Aberdeen, U.K. Randomized controlled trial to compare the safety and efficacy of TCRE and medical tx f/u time, 4 months	187 pts with menorrhagia, uterus < 10 wk gestation, normal endometrial histology; completed childbearing Mean age: TCRE, 41.7 ± 5.2 yr; Medical tx, 41.4 ± 5.2 yr	Preoperative endometrial thinning with 3.6 mg goserelin followed 5 wk later by TCRE by rollerball coagulation and wire loop resection (n=93) > 3 cycles of medical tx that had not been used before including progestogens, combined oral contraceptives, tranexamic acid, danazol, or hormone replacement therapy (n=94)	1/94 (1.2%) pt in medical tx group had TCRE at 2 months for persistent menorrhagia. In TCRE group, 1/93 (0.1%) pt had hysterectomy for submucous fibroids. At 4 months, 34/93 (37%) TCRE pts were amenorrheic compared with 3 (3%) of the medical tx group (P<0.001), and the mean bleeding score was lower in the TCRE group [5.1 ± 6.97 versus 17.8 ± 9.15] (P<0.001); however, the mean bleeding score decreased from the pretreatment value in the medical tx group (P<0.001). More pts in the medical tx group had unchanged or heavier bleeding compared with the TCRE pts [48 (52%) versus 7 (8%)] (P<0.001). Premenstrual symptoms also significantly improved in a higher percentage of TCRE pts compared with the medical tx group. 27/60 pts (45%) in the medical tx group found that the side effects were unacceptable. Pt satisfaction rates: 70/93 (76%) TCRE group and 25/93 (27%) medical tx group (P<0.001)	A significantly higher percentage of TCRE pts were amenorrheic and had improved menstrual symptoms and dysmenorrhea compared with pts in the medical tx group. A high number (45%) of pts were dissatisfied with medical tx due to adverse side effects. Significantly more of the pts in the TCRE group found the tx to be satisfactory. In this pt group, TCRE was more efficacious than medical tx for menorrhagia. Study limitations: 31% of 273 eligible pts did not consent to randomization and were not included in study; short f/u time; unclear if optimal medical tx used for individual pts

Appendix VII: Public Comment

From:	Tania Hughes <HUGHET1@mdh-hpsc2.health.state.mn.us>
Organization:	Minnesota Department of Health
To:	nancy.cusick@health.state.mn.us, brenda.holden@health.state.mn.us
Date sent:	Wed, 31 May 2000 10:46:58 -0500
Subject:	(Fwd) HTAC report
Priority:	normal

-----Forwarded message follows ----
From:	Jan Strathy@aol.com
Date sent:	Wed, 26 April 2000 21:28-08 EDT
Subject:	HTAC report
To:	Tania.hughes@health.state.mn.us

I thought I would share with you that I thought the report on surgical alternatives to hysterectomy was quite good and would be useful to those not familiar with the field. Jan Strathy MD -----
End of forwarded message ----

References

1.: Agency for Health Care Policy and Research (AHCPR): Common Uterine Conditions: Options for Treatment. Rockville, MD; U.S. Department of Health and Human Services, Public Health Service, Agency for Health Care Policy and Research. AHCPR Publication No. 98-0003. December 1997.
2.: American College of Obstetricians and Gynecologists (ACOG): Abnormal Uterine Bleeding. Washington, DC; The American College of Obstetricians and Gynecologists. ACOG Publication No. AP095. June 1996a.
3.: American College of Obstetricians and Gynecologists (ACOG): ACOG Criteria Set. Endometrial Ablation. Number 1 November 1994. Committee on Quality Assessment. Int J Gynecol Obstet. 1996b;52:99–100. [PubMed: 8620999]
4.: Brooks PG, Clouse J, Morris LS Hysterectomy vs. resectoscopic endometrial ablation for the control of abnormal uterine bleeding. A cost-comparative study. J Reprod Med. 1994;39:755–760. [PubMed: 7837119]
5.: Brown E. Diagnostic and Therapeutic Technology Assessment (DATTA). Laser Ablation of the Endometrium. July 1991
6.: Carlson KJ, Nichols DH, I Schiff Current concepts: indications for hysterectomy. N Engl J Med. 1993;328:856–860. [PubMed: 8357364]
7.: Carlson KJ, Schiff I Alternatives to hysterectomy for menorrhagia. N Engl J Med. 1996;335:198–199. [PubMed: 8657219]
8.: Chen BH, Giudice LC Dysfunctional uterine bleeding. West J Med. 1998;169:280–284. [PMC free article: PMC1305317] [PubMed: 9830356]
9.: Chuong CJ, Brenner PF Management of abnormal uterine bleeding. Am J Obstet Gynecol. 1996;175:787–792. [PubMed: 8828563]
10.: Cooper JM, Brady RM Gynecologic operative endoscopy. Hysteroscopy in the management of abnormal uterine bleeding. Obstet Gynecol Clin. 1999;26:217–236. [PubMed: 10083940]
11.: Ke RW Endometrial ablation: an alternative to hysterectomy. Clin Obstet Gynecol. 1997;40:914–927. [PubMed: 9429805]
12.: Luerti M, Garuti G Endometrial ablation with the Uterine Balloon Therapy J System: a new therapeutic option for menorrhagia. Int J Gynecol Obstet. 1996;4:153–156.
13.: Meyer WR, Walsh BW, Grainger DA, et al Thermal balloon and rollerball ablation to treat menorrhagia: a multicenter comparison. Obstet Gynecol. 1998;92:98–103. [PubMed: 9649102]
14.: Singer A, Almanza R, Gutierrez A, et al Preliminary clinical experience with a thermal balloon endometrial ablation method to treat menorrhagia. Obstet Gynecol. 1994;83:732–734. [PubMed: 8164933]
15.: R, Christian B Patterns and perception of menstruation - a World Health Organization international study. London: Croom Helm 1983. Cited in: Luerti M, Garuti G Endometrial ablation with the Uterine Balloon Therapy J System: a new therapeutic option for menorrhagia. Int J Obstet Gynecol. 1996;4:153–156.
16.: Sculpher MJ, Dwyer N, Byford, Stirrat GM Randomised trial comparing hysterectomy and transcervical endometrial resection: effect on health related quality of life and costs two years after surgery. Br J Obstet Gynecol. 1996;103:142–149. [PubMed: 8616131]
17.: Hysteroscopy. Washington, DC; The American College of Obstetricians and Gynecologists. ACOG Publication No. AP084. June 1992.
18.: Garry R Good practice with endometrial ablation. Obstet Gynecol. 1995;85:144–151. [PubMed: 7784012]
19.: Clarke A, Black N, Rowe P, et al Indications for and outcome of total abdominal hysterectomy for benign disease: a prospective cohort study. Br J Obstet Gynecol. 1995;102:611–620. [PubMed: 7654638]
20.: Vilos GA, Vilos EC, Pendley L. Endometrial ablation with a thermal balloon for the treatment of menorrhagia. J Am Assoc Gynecol Laparosc. 1996;3:383–387. [PubMed: 9050660]
21.: Garry R Endometrial ablation and resection: validation of a new surgical concept. Br J Obstet Gynecol. 1997;104:1329–133. [PubMed: 9422007]
22.: Fernandez H, Capella S, Audibert F Uterine thermal balloon therapy under local anaesthesia for the treatment of menorrhagia: a pilot study. Human Reproduction. 1997;12:2511–2514. [PubMed: 9436696]
23.: O'Connor H, Broadbent JA, Magos AL, McPherson K. Medical Research Council randomised trial of endometrial resection versus hysterectomy in management of menorrhagia. Lancet. 1997;349:897–901. [PubMed: 9093249]
24.: Amso NN, Stabinsky SA, McFaul P, et al Uterine thermal balloon therapy for the treatment of menorrhagia: the first 300 patients from a multi-center study. Br J Obstet Gynecol. 1998;105:517–523. [PubMed: 9637121]
25.: Hidlebaugh DA, Orr RK. Long-term economic evaluation of resectoscopic endometrial ablation versus hysterectomy for the treatment of menorrhagia. J Am Assoc Gynecol Laparosc. 1998;5:351–356. [PubMed: 9782137]
26.: Grant AM, Bhattacharya S, Mollison J, et al A randomised trial of endometrial ablation versus hysterectomy for the treatment of dysfunctional uterine bleeding: outcome at four years. Aberdeen Endometrial Ablation Trials Group. Br J Obstet Gynecol. 1999;106:360–366. [PubMed: 10426244]
27.: Cooper KG, Parkin DE, Garratt AM, Grant AM A randomised comparison of medical and hysteroscopic management in women consulting a gynecologist for treatment of heavy menstrual loss. Br J Obstet Gynecol. 1997;104:1360–1366. [PubMed: 9422013]
28.: Minnesota Hysterectomy Figures. HTAC staff. 11-9-1999.Ref Type: Personal Communication.
29.: Dwyer N, Hutton J, Stirrat GM Randomised controlled trial comparing endometrial resection with abdominal hysterectomy for the surgical treatment of menorrhagia. Br J Obstet Gynecol. 1993;100:237–243. [PubMed: 8476829]
30.: Lepine LA, Hillis SD, Marchbanks PA, et al Hysterectomy surveillance--United States. 1980-1993 MMWR 199746(SS-4):1–15. [PubMed: 9259214]
31.: Pokras R, Hufnagel VG Hysterectomy in the United States, 1965-84. Vital and Health Statistics 13, No. 92. PHS 87-1753. Public Health Service. Washington D.C. Cited in: Clarke A, Black N, Rowe P, et al Indications for and outcome of total abdominal hysterectomy for benign disease: a prospective cohort study. Br J Obstet Gynecol. 1995;102:611–620. [PubMed: 7654638]
32.: Wilcox LS, Koonin LM, Pokras R, et al Hysterectomy in the United States, 1988-1990. Obstet Gynecol. 1994;83:549–555. [PubMed: 8134065]
33.: Amirikia H, Evans T Ten year review of hysterectomies: trends, indications and risks. Cited in: Clarke A, Black N, Rowe P, et al Indications for and outcome of total abdominal hysterectomy for benign disease: a prospective cohort study. Am J Obstet Gynecol. Br J Obstet Gynecol. 1979;1995;134102:431–437. 611–620. [PubMed: 453279]
34.: Dicker RC, Greenspan JR, Strauss LT, et al Complications of abdominal and vaginal hysterectomy among women of reproductive age in the United States. Cited in: Clarke A, Black N, Rowe P, et al Indications for and outcome of total abdominal hysterectomy for benign disease: a prospective cohort study. Am J Obstet Gynecol. Br J Obstet Gynecol. 1982;1995;144102:841–848. 611–620. [PubMed: 7148906]
35.: Crosignani PG, Vercellini P, Apolone G, et al Endometrial resection versus vaginal hysterectomy for menorrhagia: long-term clinical and quality-of-life outcomes. Am J Obstet Gynecol. 1997;177:95–101. [PubMed: 9240590]
36.: Erian J Endometrial ablation in the treatment of menorrhagia. Br J Obstet Gynecol. 1994;101:19–22. [PubMed: 8043557]
37.: Phillips G, Chien PFW, Garry R Risk of hysterectomy after 1000 consecutive endometrial laser ablations. Br J Obstet Gynecol. 1998;105:897–903. [PubMed: 9746384]
38.: Wingo PA, Huenzo CM, Rubin GL, et al The mortality risk associated with hysterectomy. Cited in: Meyer WR, Walsh BW, Grainger DA, et al. Obstet Gynecol. 1998;92:98-103. Am J Obstet Gynecol. 1995;152:803–808. [PubMed: 4025434]
39.: HCUPnet, Healthcare Cost and Utilization Project: HCUP Nationwide Impatient Sample, 1996. Agency for Health Care Policy and Research, Rockville, MD. Available at http://www.ahcpr.gov/data/hcup/hcupnet.htm. Accessed on January 25, 2000.
40.: Everett RB Five-year review of endometrial ablation with the SideFire Laser Fiber. J Am Assoc Gynecol Laparosc. 1999;6:65–69. [PubMed: 9971854]
41.: Garry R, Khair A, Mooney P, Stuart M A comparison of goserelin and danazol as endometrial thinning agents prior to endometrial laser ablation. Br J Obstet Gynecol. 1996;103:339–344. [PubMed: 8605131]
42.: Parkin DE Prognostic factors for success of endometrial ablation and resection. Lancet. 1998;351:1147–1148. [PubMed: 9643682]
43.: Bhattacharya S, Cameron IM, Parkin DE, et al A pragmatic randomised comparison of transcervical resection of the endometrium with endometrial laser ablation for the treatment of menorrhagia. Br J Obstet Gynecol. 1997;104:601–607. [PubMed: 9166205]
44.: Gannon MJ, Holt EM, Fairbank J, et al A randomised trial comparing endometrial resection and abdominal hysterectomy for the treatment of menorrhagia. BMJ. 1991;303:1362–1364. [PMC free article: PMC1671634] [PubMed: 1760601]
45.: Garry R, Shelley-Jones D, Mooney P, Phillips G, et al Six hundred endometrial laser ablations. Obstet Gynecol. 1995;85:24–29. [PubMed: 7800318]
46.: Lissak A, Fruchter O, Mashiach S, et al Immediate versus delayed treatment of perimenopausal bleeding due to benign causes by balloon thermal ablation. J Am Assoc Gynecol Laparosc. 1999;6:145–150. [PubMed: 10226122]
47.: HHS NEWS. FDA approves new device to treat excessive menstrual bleeding. December 12, 1997. Available at: http://www.fda.gov/bbs/topics/NEWS/NEW00607.html. Accessed June 9, 1999.
48.: Pinion SB, Parkin DE, Abramovich DR, et al Randomised trial of hysterectomy, endometrial laser ablation, and transcervical endometrial resection for dysfunctional uterine bleeding. BMJ. 1994;309:979–983. [PMC free article: PMC2541307] [PubMed: 7772106]
49.: Gannon MJ, Day P, Hammadich N, et al A new method of measuring menstrual blood loss and its use in screening women before endometrial ablation. Cited in: Parkin DE Prognostic factors for success of endometrial ablation and resection. Br J Obstet Gynecol. Lancet. 1996;1998;103351:1029–1033. 1147–1148. [PubMed: 8863704]
50.: The Gray Sheet: J&J Ethicon launch of ThermaChoice uterine balloon backed by direct sales force of roughly 60 reps. 1997. pp. I&W–1.

: June 2000 Physician Report

Bookshelf ID: NBK52515

Contents

< Prev Next >

PubReader
Print View
Cite this Page
Minnesota Health Technology Advisory Committee. Minnesota Health Technology Assessments [Internet]. St. Paul (MN): Minnesota Department of Health; 1995-2001. Surgical Alternatives to Hysterectomy for Abnormal Uterine Bleeding. 2000 Jun.

Other titles in this collection

Health Services/Technology Assessment Text (HSTAT)

Recent Activity

Clear Turn Off Turn On

Surgical Alternatives to Hysterectomy for Abnormal Uterine Bleeding - Minnesota ...
Surgical Alternatives to Hysterectomy for Abnormal Uterine Bleeding - Minnesota Health Technology Assessments

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

Bookshelf

Minnesota Health Technology Assessments [Internet].

Surgical Alternatives to Hysterectomy for Abnormal Uterine Bleeding

Executive Summary

Conclusions

Recommendations

Medical Background

Surgical Treatments

Hysterectomy

Endometrial Ablation

Patient Selection Criteria

Patient Evaluation

Pre-operative Preparation

Post-operative Follow-up

Endometrial Ablation Techniques

Transcervical Resection of the Endometrium (TCRE)

Procedure

Findings

Complications

Endometrial Laser Ablation (ELA)

Procedure

Findings

Complications

Thermal Balloon Ablation of the Endometrium (TBEA)

Procedure

Findings

Complications

Endometrial Ablation Compared to Hysterectomy

Endometrial Ablation Compared to Drug Treatment

Safety

Distention-Media Complications

Mechanical Complications

Thermal Complications

Endometrial Cancer

Pregnancy

Issues of Controversy

Assessment of Menorrhagia and Menstrual Blood Loss

Over-use of Endometrial Ablation

Outcome Measures

Patient Satisfaction and Menstrual Blood Loss

Costs And Cost-effectiveness

Table

Table

Conclusions

Recommendations

Appendix I: Methodology

Appendix II: Government Agencies And Professional Organizations Guidelines

Appendix III: Randomized Controlled Trials on the Efficacy of TCRE versus Hysterectomy

Appendix IV: Clinical Studies on the Efficacy of ELA

Appendix V: Clinical Studies on the Efficacy of TBEA

Appendix VI: Randomized Controlled Trials on the Efficacy of EA versus Standard Treatment

Appendix VII: Public Comment

References

Views

In this Page

Other titles in this collection

Recent Activity