Postmenopausal endometriosis: a retrospective analysis of 69 patients during a 20-year period

Endometriosis, characterized by endometrial-like tissue outside the uterus, is a frequently occurred benign gynecologic disease that afflicts mainly women of reproductive age, with a prevalence of 6% to 10%. Endometriosis can be subdivided into ovarian, peritoneal and deep infiltrating types, with the deep infiltrating endometriotic lesions closely related with pain symptoms.1 Though a rare clinical occurrence, endometriosis does affect 2%-5% women in their postmenopausal years, with many of the cases reported having been under hormone replacement therapy (HRT). With the ovaries being its most common sites, postmenopausal endometriosis poses a 1% risk of malignant transformation, as well as an increased risk of ovarian cancer.2 Usually with abdominal or pelvic mass the only clinical finding before surgery, almost all cases reported were diagnosed retrospectively by pathologic examinations. Postmenopausal endometriosis appears in most of the literature as case reports. The prevalence of postmenopausal endometriosis is obscure, so are the clinical significance, best management strategy and prognosis. In this study, to define the clinical importance, the management and the prognosis of postmenopausal endometriosis, we retrospectively analyzed the clinical characteristics of 69 patients, surgically treated and pathologically confirmed in Peking Union Medical College Hospital, China during a 20-year span. METHODS A computerized database searching was conducted for the medical records and pathology reports of all the patients who underwent surgery with histopathology diagnosis of endometriosis at our hospital from April, 1993 to April, 2013. Among the 8376 patients retrieved from our database, 69 were found to be postmenopausal, defined as having been menopaused for more than 2 years naturally or surgically before index surgery. Their medical records were further reviewed to collect the following variables: age, age at menopause, previous history of pelvic pain, infertility and endometriosis, body mass index (BMI), previous use of HRT, chief complaints when admitted, ultrasonography reports, CA-125 values and complete pathology diagnosis. Descriptive statistics were used in the following analysis. Normally distributed data (i.e. age) are expressed as mean and standard deviation (SD) while skewed data (i.e. duration of follow-up) are represented as median and range. Their clinical characteristics were summarized and analyzed as numbers counted and percentages. RESULTS The age of the 69 patients ranged from 47 to 80 ((56.5±6.4) years). They had been postmenopausal for 2 to 29 years, with an average of (5.9±5.5) years. The average BMI of these patients was 27.3±4.4, ranging from 18.8 to 41.7. And they were distributed according to the WHO BMI classifications for Chinese people as the following: underweight (less than 18.5): 0; normal (18.5–24.9): 29; overweight (25.0–27.5): 7; obese (more than 27.5): 33. The percentage of overweight or obesity was 60.0%. Among them, 2 had had hysterectomy and the rest had come into menopause naturally. Four of them were primary infertile, the rest had been pregnant at least once, ranging from 1 to 8 times, averaging 2.9 times. Among the 69, 19 (27.5%) had history of pelvic pain (cyclic or non-cyclic); 8 (11.6%) had been diagnosed of endometriosis. Only 2 of the 69 patients had been under HRT (Tibolone, 2.5mg per day for 6 and 10 months, 18 and 33 months before admission, respectively). Forty-five of the 69 patients (65.2%) were referred to our hospital with no obvious symptoms but pelvic or abdominal masses detected by ultrasonography. Among the 24 patients (34.8%) came with self-sensed symptoms; most of them were for vaginal bleeding (11, 45.9%) and abdominal pain (8, 33.3%). 41 of the 69 patients (59.4%) had had their CA125 values tested before surgery, measuring in a range of 1.8 to 218.4 U/ml, with an average value of (24.5±9.7) U/ml. Abdominal/pelvic masses were found by ultrasonography in all the 41 patients, among whom 11 (26.7%) had CA125 values higher than normal. Other than endometriosis, myoma (36, 52.2%) and adenomyosis (21, 30.1%) are the top two commonest coexiting lesions confirmed pathologically. Among the 69 patients, ovarian endometriomas were found in 62 patients (89.9%). Superficial peritoneal endometrioses were found in 7 patients (10.1%) and retrovaginal septum endometrioses in 6 (8.7%) (one patient could have more than one type of endometriosis). Apart from the 10 patients with pelvic malignant tumor, 46 had total hysterectomy and bilateral salpingo-oophorectomy, 13 had bilateral or unilateral salpingo-oophorectomy without hysterectomy, 21 had adhesiolysis, 9 had endometriosis lesion excision or ablation (one patient could have more than one type of surgery). The 10 patients with coexisting endometrial, ovarian and cervical cancer received corresponding surgery plus chemotherapy or radiotherapy. In the follow-up visits during a period of 2–78 months ((14.8±4.7) months), none of the patients showed signs of recurrence of endometriosis on ultrasonography or pelvic examination. DISCUSSION We retrospectively reviewed 69 cases of postmenopausal endometriosis in our study, aiming to portray the clinical profiles, explore its clinical significance, best management strategy and prognosis. In our study, the patients showed two major characteristics. One is that only 2 (2.9%) of them had been under HRT before diagnosis of postmenopausal endometriosis. The other is that most of them were obese or overweight. Only 2 out of the 69 patients had used HRT before admitted to our hospital. This may seem to be a very low rate compared to that of peri- and postmenopausal women from Western countries. But this usage rate of HRT is quite common in China, and it's partly because of a low prevalence of menopausal symptoms and partly because of poor education on menopause and little knowledge of HRT among Chinese women. The fact that 21 of the 69 patients (30.4%) had history of infertility or pelvic pain (8 of them had been previously diagnosed as having endometriosis clinically or surgically) suggests there is a good chance that some of the patients had had premenopausal endometriosis which persisted into postmenopausal period. As elucidated by Bendon and Becker,3 the most plausible mechanism of postmenopausal endometriosis pathogenesis is that they are preceded by premenopausal ones. Endometriosis has long been considered an estrogendependent disease. We do not have the estrogen levels of most of the patients before surgery due to the limitation of a retrospective study. But the high ratio of coexisting estrogen-dependent diseases, such as adenomyosis (30.1%) and myoma (52.2%), supports a higher-than-general population estrogen level. After menopause, estrogens are derived either from exogenous administration such as HRT or endogenous extra-ovarian production. Most of the reported cases of postmenopausal endometriosis developed in women with previous history of HRT. In our study, only 2 (2.9%) of the 69 patients in this study had been under HRT. Thus the estrogens were mainly from endogenous extra-ovarian production, including the adrenal glands, endometrial stroma, adipose tissue and the endometriosis lesion itself. Among them, the adipose tissue provides the largest part of estrogen. In keeping with their findings, 60% of the patients from our study were obese or overweight, which is much higher than the prevalence rate of 42.6% for general female Chinese population aged 60 years or above. We speculate that obesity is a possible risk factor for postmenopausal endometriosis as it is for endometrial carcinoma. It is now well established that endometriosis lesion can produced estrogen itself by aromatase and propel its growth in a positive feedback style. For the patients in our study, it is possible that one source of their higher-thangeneral population estrogen is the endometriosis lesion. Compared with endometrioses of reproductive women, postmenopausal ones lack characteristic symptoms and CA125 levels are usually within normal range. History of pelvic pain, infertility and HRT may provide a hint of postmenopausal endometriosis (especially true when the patient is obese), but surgical excision and pathologic confirmation are needed to make a definite diagnosis. Although medical and surgical treatments are both first-line treatments for endometriosis, since postmenopausal women have little ovarian function left, the medical treatments aiming at suppressing ovarian function seem implausible. And because of the increased malignant potential of abdominal and pelvic masses and possibility of malignant transformation of endometrioses in postmenopausal women, surgery is the recommended treatment. These years, aromatase inhibitors aiming to suppress all the estrogen produced by the adipose tissue and skin as well as by the endometriosis lesion itself have shown prospects for patients who fail to respond to other therapies or not suitable for surgeries, some of whom are postmenopausal ones.4 This is a possible option for medical treatment of postmenopausal endometriosis, though further researches are needed to validate wide clinical use.