British Journal of Obstetrics and Gynaecology May 1992, Vol. 99, pp. 399401

GYNAECOLOGY

An objective and subjective assessment of uterine blood loss in postmenopausal women on hormone replacement therapy TOMMY SPORRONG G O R A N RYBO L A R S AKE MATTSSON Department of Obstetrics and G ynaecology East Hospital University of Goteborg Sweden GUDJON VILBERGSON NILS CRONA Department of Obstetrics and Gynaecology Sahlgrenska Hospital University of Goteborg Sweden

ABSTRACT Objective To determine the volume of withdrawal blood loss in women taking hormone replacement therapy (HRT). Design Prospective randomized double blind cross-over trial. Subjects Twenty-nine postmenopausal women, with climacteric symptoms. Interventions The women were treated with cycles of 2 mg oestradiol valerate for 21 days combined with either 2.50 p g levonorgestrel or 10 mg medroxyprogesterone acetate, for the last ten days. Each woman was randomly allocated to one of the preparations and switched to the other after four months. Main outcome measures Uterine blood loss was estimated by the alkaline haematin method. Bleeding patterns were recorded specifying the number of days with bleeding and subjective rating of the amount of blood lost. Results and conclusions The mean uterine blood loss was 35,2 ml and was equivalent to the normal menstrual blood loss. There was no significant difference between therapies but the women themselves considered that they bled more heavily when taking the levonorgestrel combination. The intraindividual variation in blood loss was small, independent of preparation taken, but the variation between individuals was large. Withdrawal bleeding started 2-3 days earlier when the women used the medroxyprogesterone acetate compared to the levonorgestrel formulation.

Many women suffering from climacteric symptoms are advised to take hormone replacement therapy (HRT). Previous studies have shown that treatment with oestrogen alone increases the risk of hyperplasia and carcinoma of the endometrium (Smith et al. 1975). This risk may be eliminated by adding a progestogen to the oestrogen treatment (Persson et al. 1989). The type and the duration of progestogen therapy probably affects women’s feeling of well-being, the appearance of the endometrium and bleeding patterns. Most women (90%) taking sequential or cyclic oestrogen-progestogen therapy will have regular withdrawal bleeds (Whitehead et al. 1978). Some women using HRT report heavy bleeding, but the correlation between subjective and measured blood loss, at least in younger women, is known to be poor (Hallberg et al. 1966). It was reported that 9-1470 of healthy menstruating women have menorrhagia (Eijkeren et al. 1989), defined as a menstrual blood loss >SO ml, but to our knowledge, no previous trial has investigated withdrawal blood loss during HRT in postmenopausal women. This study was designed to measure blood loss objectively during HRT and to study the influence of two different progestogens. Correspondence: Tommy Sporrong, Department of Obstetrics and Gynecology, East Hospital, S-416 85 Goteborg, Sweden.

Subjects and methods The study included 29 women who attended our outpatient clinics mainly because of vasomotor symptoms, such as hot flushes and sweats. Their mean age was 52.5 (SD 5) years and, on average, they had been without regular periods for 5.2 (SD 5) years. At least 6 months had elapsed since the last menstrual period in each woman. Twenty-three women had taken HRT previously and they were subjected to a wash-out period of 4-6 weeks before entering the study. Three women discontinued after I , 8 and 10 weeks of treatment because of minor discomforts such as bleeding disturbances or lack of symptom alleviation. All the women gave their written informed consent. The study was approved by the ethics commitee of the Faculty of Medicine, University of Goteborg.

Hormone therapy Two oestrogen-progestogen combination regimens were used. Each consisted of 2 mg oestradiol valerate (E2V) per tablet taken daily for 3 weeks out of 4. The progestogen component was added during the last 10 days and consisted of either 250 pg levonorgestrel (LNG), or 10 mg medroxyprogesterone acetate (MPA), daily. The study was designed in a doubleblind, cross-over fashion. Each woman was allocated, using random numbers, to one of the formulations and switched to

399

400

T . S P O R R O N G ET A L .

the other after 4 months. The physicians and the women were not aware of the treatment sequence until all data had been analysed. The tablets and the packages containing the tablets looked the same.

Protocol All the women had a physical examination, including an endometrial biopsy, before starting therapy and on days 19-21 after 4 and 8 months of treatment. Endometrial biopsies were taken using either a Mi-Mark helix or a Vabra currette and the histological sections were examined by light microscopy. Bleeding patterns were recorded continuously by all the women, specifying the number of days with bleeding and each woman’s own daily rating of the amount of blood lost. The woman attributed a score of 3 if she considered her blood loss to be as during a menstruation, 2 if the bleeding was small, 1 for spotting and 0 if there was no bleeding at all. The blood loss during withdrawal bleeding was measured in treatment months 2 and 3 for each formulation, as described by Hallberg & Nilsson (1964) and modified by Newton et al. (1977). The vaginal tampons and sanitary towels were collected in plastic bags. The determination of blood loss was based upon the formation of alkaline haematin after the blood had been extracted from vaginal tampons and sanitary towels by NaOH in an automatic Stomacher Lab-Blender 3500.

Conventional methods were used for calculating the mean and standard deviation. The Mann-Whitney U-test and Wilcoxon signed rank test were used to compare differences during treatment. The Spearman rank correlation coefficient was used to measure correlation between blood losses and bleeding scores. A P-value 80

15

4

3 1

1

4

1 1 I

3 1 3

3

>80 No. of women

Endometrial histology Endometrial hyperplasia or atypia was not found in any of the women in this study. There was no difference in the histopathological appearance of the endometrium in the two groups and no significant correlation between the histopathological appearance of the endometrium during treatment month 4 and the amount of uterine bleeding during treatment months 2 and 3 could be demonstrated.

Discussion

Statistics

Blood loss with MPA (ml)

There were wide inter-individual variations in blood loss (mean variation 87 ml), but the intra-individual variation was small (mean variation 26 ml) (Table 1). The mean blood loss during the two oestrogen-progestogen treatments are shown in Table 2. No significant difference in the measured blood loss could be detected between the two formulations (Wilcoxon signed rank test). However, the subjective ratings of the blood loss were higher when the levonorgestrel containing preparation was given (Table 3). The mean duration of bleeding each month was 3.8 (SD 2.1) and 3.6 (SD 2.5) days for the LNG and MPA preparations, respectively (P>O.OS). The onset of withdrawal bleeding after the last tablet occurred after 3.1 (SD 2.2) days after administration of the LNG and after 0.4 (SD 3.2) days with the MPA formulation, P

An objective and subjective assessment of uterine blood loss in postmenopausal women on hormone replacement therapy.

To determine the volume of withdrawal blood loss in women taking hormone replacement therapy (HRT)...
308KB Sizes 0 Downloads 0 Views