The prevalence of endometriosis may be as high as 45 percent in women of reproductive age. Inpatient treatment costs of endometriosis alone were estimated at $579 million in 1992. In addition, endometriosis is believed to account for a significant proportion of infertility and 97 percent of cases of chronic pelvic pain, which is estimated to have direct medical costs of $2.8 billion annually plus indirect costs of at least $600 million. Winkel provides an overview of the evaluation and management of women with endometriosis.

Endometriosis should be suspected in women with symptoms of pelvic pain, dysmenorrhea, dyspareunia, or infertility, but it may be asymptomatic. Although most patients have normal pelvic examinations, findings can include enhanced tenderness on bi-manual examination, nodularity, especially in the posterior cul-de-sac or along the uterosacral ligament, decreased uterine mobility or retroversion, and adnexal masses. No laboratory tests are helpful in making the diagnosis, although serum CA 125 levels may be elevated in severe cases. Imaging studies such as computed tomography, magnetic resonance imaging, or ultrasonography are useful only if masses are present. Laparoscopy can be misleading if lesions are confused with a variety of pathologic lesions. Histologic confirmation is required. The reliability of a clinical diagnosis is similar to that of one based on laparoscopy with biopsy.

Because pregnancy and menopause are associated with resolution of endometriosis symptoms, most treatments are based on hormonal manipulation. Oral contraceptives used continuously are recommended as first-line therapy in women who have no contraindications to these drugs and who do not wish to become pregnant. Because progesterone causes regression of endometriosis, medroxyprogesterone acetate in a dosage of 20 to 30 mg daily is prescribed frequently. However, daily doses of 50 mg were no more effective than placebo in a 12-week trial and were associated with significant adverse effects. Depot progesterone also has been associated with a high incidence of side effects in 15 to 65 percent of patients. The androgen derivative danazol was reported to induce clinical improvement in 55 to 93 percent of patients treated for six months. However, it also had a high incidence (85 percent) of adverse effects. Gonadotropin-releasing hormone (GnRH) analogs can induce 85 to 100 percent rates of improvement that last for at least six to 12 months after cessation of therapy. Most of the side effects of GnRH analogs are caused by hypoestrogenism; these effects can be ameliorated by “adding back” low doses of estrogen, progestogen, or both. The add-back regimen should be individualized for each patient. If treatment is continued for longer than 12 months, bone mineral density should be checked at least every 24 months.

Surgery has been advocated as first-line treatment for women with severe pain who wish to become pregnant. Laparoscopic approaches are usually preferred and are as effective as laparotomies. Results of surgical treatment remain controversial. Studies have suffered from many methodologic problems, including multiple different surgical techniques and study designs, and a considerable placebo effect. Besides uncertainty over whether lesions should be excised or ablated by a variety of techniques, the contribution of adjunctive presacral neurectomy or uterosacral nerve ablation-transection is unclear. Radical surgery, including hysterectomy and bilateral oophorectomy plus removal or ablation of lesions, is associated with recurrence in up to 10 percent of women.

The author concludes that endometriosis remains an enigmatic condition for which the outcomes of medical and surgical treatment appear equivalent. He stresses the need to individualize treatment for each patient.