Treatment of Common Non-Cancerous Uterine Conditions

Issues for Research

Conference Summary

---

Following are highlights of a May 1994 AHCPR conference on research issues in the effectiveness of hysterectomy and alternative therapies for common non-cancerous uterine conditions. The full report of the conference is available from the AHCPR Publications Clearinghouse. Call toll free 800-358-9295. Order AHCPR Pub. No. 95-0067 (July 1995).

---

Overview

Hysterectomy is the most common nonpregnancy-related major surgery performed on women in the United States. In 1995, approximately 590,000 women in this country will undergo the procedure. Surgical removal of the uterus, and frequently the ovaries, is widely accepted both by medical professionals and the public as appropriate treatment for uterine cancer, and for various common non-cancerous uterine conditions that can produce often disabling levels of pain, discomfort, uterine bleeding, emotional distress, and related symptoms. Yet, while hysterectomy can alleviate uterine problems, less invasive treatments are available.

Most women who undergo hysterectomy are between the ages of 35 and 54, with the highest age-specific rate for women 35 to 44 years of age. Overall, fibroids account for approximately one-third of all hysterectomies performed in the United States. Endometriosis is the second most common condition leading to hysterectomy, accounting for 18 percent. Hysterectomy rates also are correlated with a number of non-clinical characteristics of patients, such as socioeconomic status, and with provider variables, such as physician training.

Health services research findings since the 1970s have highlighted wide, unexplained variations in rates of hysterectomy in different parts of the United States, and much higher rates in the United States compared with other Western countries. There is no way, however, to determine from these studies which rate is right.

Thus, AHCPR initiated work to identify specific research opportunities related to the outcomes of hysterectomy and its alternatives, and to encourage such research. The conference had a dual purpose: to assess the state of the science, and to identify the most important areas for effectiveness research.

Conclusions

The current scientific literature is weak and incomplete. Studies containing original data typically are small, observational studies; the few which compare treatments focus on one type of hysterectomy versus another type (e.g., abdominal versus vaginal surgery). Outcomes addressed in these studies are limited almost exclusively to traditional endpoints, such as mortality, complications of surgery, and other physician assessments. These studies confirm that the risk of mortality is low; however, complications are common occurrences.

Very few studies provide information about the effects of hysterectomy on the symptoms that led women to seek treatment in the first place or on the long-term outcomes that contribute to the patient's quality of life. Reports often lack enough data about study design, sample size, patient characteristics, reasons for treatment, and other information critical to interpreting and weighing the results.

Even the best studies beg the critical question: For non-cancerous uterine conditions, what treatment is most effective? Only a few, preliminary studies have compared the outcomes of hysterectomy with other treatment alternatives and considered outcomes from the patient's perspective.

Alternatives to hysterectomy fall into three general categories: conservative surgical management; pharmacologic therapies (hormonal and nonhormonal); and other strategies, including psychosocial support and therapy, and watchful waiting. There has been little research on how physicians or their patients choose among available treatments. Potential applications of these treatments are summarized in the following table:

Alternatives to Hysterectomy for Common Non-Cancerous Uterine Conditions

Condition	Conservative Surgery	Pharmacologic Therapies		Other Strategies
		Hormonal	Nonhormonal
Fibroids	Myomectomy Endometrial ablation	GnRH(a) agonists with add-back therapy Oral contraceptives Androgens RU-486(b) Gestrinone(b)	NSAIDS(c)	Watchful waiting
Endometriosis	Adhesiolysis Excision of endometrial ablation Resection of cul-de-sac obliteration Nerve blocks Uterosacral nerve ablation	GnRH(a) agonists with add-back therapy Danazol Progestins Oral contraceptives Tamoxifen(b) RU-486b	NSAIDS(c) Analgesics Anxiolytics	Watchful waiting Biofeedback Acupuncture Hypnosis Lifestyle change(nutrition, exercise)
Prolapse	Anterior or posterior colporrhaphy Laparoscopic or vaginal suspension techniques	Estrogen		Watchful waiting Kegel exercises Pessaries Electrical stimulation Urethral beads Periurethral injections of GAX(b), collagen, fat, silicon, etc.
Dysfunctional bleeding	Dilation and curettage Endometrial ablation	Progestins Estrogen Oral contraceptives Danazol Prostaglandin inhibitors GnRH(a) agonists Antifibrinolytic agents Luteinizing hormone agonists		Watchful waiting Antidepressants
Chronic pelvic pain	Adhesiolysis Nerve blocks Denervation procedure Uterosacral nerve ablation	Danazol GnRH(a) agonists with add-back therapy Oral contraceptives Medroxyprogesterone acetate	NSAIDs(c) Analgesics Nerve blocks Narcotics	Watchful waiting Counseling Biofeedback Relaxation techniques Trigger point injections Acupuncture Psychotropics Antidepressants Physical therapy

Notes:
(a) Gonadotropin-releasing hormone.
(b) Experimental treatment.
(c) Nonsteroidal anti-inflammatory drugs.

The research needs identified by AHCPR's conferees address the scarcity of attempts to prove the effectiveness of hysterectomy, the methodological weaknesses in much of the clinical research that has been done, and the limited attention to outcomes important to patients.

AHCPR is particularly interested in supporting randomized clinical trials designed to answer important questions about effectiveness and relative effectiveness. Studies addressing related issues, such as methodological and epidemiological topics, also are encouraged as appropriate submissions for AHCPR's program of research on medical effectiveness.

Research Recommendations

All Non-Cancerous Uterine Conditions:

• Evidence regarding the effectiveness and relative effectiveness of hysterectomy and alternative treatments is seriously lacking. Prospective randomized studies are needed comparing hysterectomy with watchful waiting, and comparing different treatment strategies (surgical, pharmacologic, psychological, and combinations thereof).
• A broad range of patient outcomes need assessment. Typically, symptoms are what drive women to treatment for non-cancerous uterine conditions. Thus, in addition to traditional clinical endpoints, it is critical to understand the effects of treatment (or time) on the presenting symptoms and development of new symptoms, and to measure the value of particular outcomes to individual women.
• Assessment of multiple outcomes require basic methodological work to validate existing measures or develop new measures. Measures need to be validated and standardized so that the findings of different studies can be compared and, possibly, to permit aggregation of results from small studies.
• Epidemiologic studies, especially large, prospective, community-based cohort studies, are needed to determine the incidence and prevalence of uterine problems, their natural history, and the factors that place some women at high risk.
• Important variables influencing patients' and providers' perceptions and expectations of different treatments are poorly understood. Research is needed to explain how physician-patient interaction affects treatment decisions.

Uterine Fibroids:

• Management of asymptomatic fibroids—when, if, or how to treat.
• Importance of fibroid size. Published criteria generally recommend surgical removal of fibroids if and when they reach the size of the uterus at 12-weeks gestation, and this has become usual practice. Studies are needed to determine how changes in fibroid size influence patients' functional status and quality of life. The scientific literature does not provide adequate evidence to support the recommendation/practice of hysterectomy at 12-weeks gestational size.
• Formation of leiomyosarcoma. With the possibility of malignant change in fibroids, rapid growth in the size of the uterus or fibroids is often used as justification for hysterectomy. The scientific basis for this practice is inadequate.
• Mechanisms of hormone ablation, add-back (norethindrone) therapy, and fibroid growth in the absence of estrogen. Investigations in hormonal therapies and management need to address receptor content.
• Etiology of fibroids, the mechanisms that influence their growth, and reasons for apparently higher rates in black women relative to white women.

Endometriosis:

• Determine what attributes of patient, provider, and treatment predict relief of pain, cost, health status, functional status, and health-related quality of life.
• Develop techniques and methods that are less invasive than laparoscopy for diagnosing endometriosis, e.g., imaging techniques and/or blood-serum markers.
• Investigate the relationship between endometriosis and dysmenorrhea, and between endometriosis and non-cyclic chronic pelvic pain.
• Investigate the biologic etiology of endometriosis and how various growth factors as well as the immune system affect the initiation and progress of the disease.

Pelvic Prolapse/Urinary Dysfunction:

• No scientific evidence was found favoring hysterectomy as the best alternative for managing pelvic prolapse. In fact, the studies containing original data were all conducted in women who experienced prolapse subsequent to hysterectomy.
• For uterine prolapse, determine the effectiveness, relative to hysterectomy, of non-surgical treatments including the use of pessaries, estrogen, and exercises to strengthen the pelvic floor.
• For urinary stress incontinence, determine the relative effectiveness of available treatments. Panelists recommended a prospective clinical trial, with minimum 5-year followup, to determine the effectiveness of surgical and nonsurgical treatments for urinary stress incontinence.

Dysfunctional Uterine Bleeding and Chronic Pelvic Pain:

• For dysfunctional bleeding, compare the outcomes of alternative treatments, including medical treatments, surgical alternatives (endometrial ablation and hysterectomy), and watchful waiting.
• For the management of chronic pelvic pain, evaluate the effectiveness of hysterectomy, surgical procedures other than hysterectomy (lysis of pelvic adhesions, etc.), medical therapies (trigger point injections, etc.), and non-traditional therapies, such as biofeedback.
• Investigate the effectiveness of imaging methods commonly used to diagnose the cause of chronic pelvic pain.