Comment on the "Recommendations for the Diagnosis and Treatment of Endometrial Cancer-Update 2013" by G. Emons and P. Mallmann for the Uterus Commission of AGO.

Abstract

Prior to starting work on the planned S3 guideline on the “Diagnosis and Treatment of Endometrial Cancer”, G. Emons and P. Mallmann, acting for the Uterus Commission of AGO, provided a summary of the current state of affairs 1. The summary, which consisted of a brief outline of the main points, was discussed in detail by the Uterine Malignancies Project Group of the Munich Tumor Center, and two points in particular prompted the following comments: 1. “Every bleeding in the postmenopausal period should be investigated histologically irrespective of ultrasound assessment of endometrial thickness.” This recommendation should be qualified. About 5 % of all gynecological consultations are for postmenopausal bleeding, and 4–11 % of all women experience some form of such bleeding. The causes of postmenopausal bleeding are listed below and include: Endometrial atrophy: 59 % Endometrial polyps: 12 % Hormonal causes: 7 % Serometra, pyometra, haematometra: 2 % Endometrial hyperplasia: 10 % Endometrial cancer: 10 % If we were to follow the recommendation cited above, around 80–90 % of all women with postmenopausal bleeding would require hysteroscopy and fractionated curettage with all the potential associated complications to investigate the cause of bleeding to diagnose a benign (in some cases such as atrophy, physiological) change. We therefore consider that the procedure recommended by the American College of Obstetricians and Gynecologists (ACOG) is more suitable, as this is based on an assessment of endometrial thickness which can usually be measured extremely well with vaginal ultrasound. The ACOG recommends only carrying out histological investigations if endometrial thickness is > 4 mm (double endometrial thickness, measured in the sagittal plane as the distance between the anterior stratum basalis to the posterior stratum basalis layer), and which can be done either by endometrial biopsy or hysteroscopy with curettage 3. This approach has a negative predictive value of 99.4–100 % and is sufficiently reliable to exclude a diagnosis of malignancy as the cause of postmenopausal bleeding; this would spare the majority of women from having to undergo an operative intervention. We therefore propose the following wording: “Every first-time bleeding in the postmenopausal years should be investigated histologically if an endometrial thickness of > 4 mm is detected with vaginal ultrasound. Repeated bleeding in the postmenopausal years should be investigated histologically irrespective of ultrasound findings.” 2. “Atypical bleeding in perimenopausal women should be investigated histologically.” But what does “atypical bleeding in perimenopausal women” mean? During perimenopause, the intervals between bleeding may be quite long without the bleeding being atypical. In most cases, constant bleeding due to hormonal imbalance during perimenopause can be remedied with hormone therapy and histological investigation is only necessary if bleeding is refractory to hormone therapy. No further treatment is required as either the menstruation cycle will spontaneously become regular again or postmenopausal amenorrhea will follow. We therefore suggest changing this recommendation as follows: “Therapy-resistent or persistent perimenopausal bleeding disorders should be investigated histologically”. If histological investigation is necessary because endometrial thickness cannot be measured or endometrial thickness exceeds the cut-off of 4 mm described above in a patient with postmenopausal bleeding or with persistent or therapy-resistant perimenopausal bleeding disorder, endometrial biopsy should be carried out (e.g. using the Pipelle®), as this could avoid the need for operative procedures. The Pipelle® had a sensitivity of 99.6 % in postmenopausal woman and 91 % in pre-/perimenopausal women for the diagnosis of endometrial cancer; its sensitivity for the detection of atypical endometrial hyperplasias was 81 % 2.