This pooled analysis shows that E2V/DNG rapidly reduces MBL in women with heavy and/or prolonged menstrual bleeding. This effect is achieved at the first withdrawal bleed after treatment initiation and is sustained over the duration of treatment. By pooling data from the two Phase III trials conducted in Europe and Australia and North America, the decrease in MBL achieved with E2V/DNG was shown to be consistent across a larger and more diverse population of women. Moreover, the reduction in MBL was associated with a decrease in the number of sanitary protection items used and an improvement in iron metabolism parameters.
In the present study, the percentage reduction in MBL with E2V/DNG appears to be independent of baseline MBL. Interestingly, small reductions in MBL were also observed in the placebo group, with higher percentage reductions observed among those with highest baseline MBL. It may be postulated that the willingness to continue to adhere to study procedures (such as the highly demanding process of collecting used sanitary items) may have been compromised if those receiving treatment perceived a lack of clinical efficacy and, therefore, became less motivated to participate fully. This lack of motivation to fully participate would likely disproportionately affect the placebo group, especially those with highest baseline MBL, and, as such, would have affected the total count of sanitary items used, leading to an underestimation of blood loss determination by using the alkaline haematin method.
Although not directly comparable, the median decrease in MBL achieved by treatment cycle 7 with E2
V/DNG treatment (88%) appears to approach that achieved with the LNG-IUS (median 95% and 96% reduction) over six cycles in two studies that also used the alkaline haematin method to objectively assess blood loss in women with heavy menstrual bleeding15,16
. The LNG-IUS (n
= 82 and n
= 25, respectively) produced significantly greater reductions in median monthly MBL than the active comparators used: short-course oral medroxyprogesterone (n
= 83) 10 mg daily for ten days starting on the 16th day of the menstrual cycle (21%)15
or mefenamic acid (n
= 26) 500 mg three times daily for the first four days of the menstrual cycle (17%)16
. By comparison, a median 24% reduction in MBL by treatment cycle 7 occurred in the placebo group in our study. Placebo-related reductions in measured blood loss have been recorded in heavy menstrual bleeding in a previous study over six menstrual cycles17
Oral tranexamic acid (1.3 g three times daily for five days from the start of menstruation) was shown to reduce mean MBL by 40.4% (vs. 8.2% reduction with placebo) in a recent study that also used the alkaline haematin method to objectively assess blood loss17
.These results with oral tranexamic acid appear rather modest in comparison with the reductions in blood loss achieved with E2
V/DNG in our pooled analysis. Additionally, the added advantage of using E2
V/DNG is that it provides effective contraceptive protection and use of back-up contraception is not required. Agents such as tranexamic acid do not offer contraceptive protection and should not be used in women using COCs18
. Similarly, cyclical progestogens do not provide adequate contraceptive protection and additional non-hormonal backup contraception would be also required.
V/DNG represents the first COC approved for the treatment of heavy and/or prolonged menstrual bleeding in many countries worldwide. Other oral contraceptives have not been formally approved for the treatment of heavy menstrual bleeding but, have nonetheless, been frequently used off-label with some success.The available studies assessing the effectiveness of other COCs in reducing MBL in women with heavy menstrual bleeding are limited to three randomised comparative studies with the LNG-IUS19–21
, one of which was in women with fibroid-related heavy menstrual bleeding20
, and a randomised crossover study that included mefenamic acid, naproxen and danazol7
In the first comparative study21
, which used the semi-quantitative pictorial blood loss assessment chart (PBAC) to assess MBL, treatment with a COC containing norethindrone acetate (NETA) 1 mg and ethmylestradiol (EE) 20 µg (21/7 regimen; n
= 19) resulted in a 68% reduction in median MBL scores at 12 months (vs. 83% with the LNG-IUS; between-group differences p
= 0.002). In the second comparative study19
, treatment with a COC containing LNG 150 µg/EE 30 µg (21/7 regimen; n
= 56) resulted in a mean reduction in MBL at 12 months of 35%, as assessed using the alkaline haematin method (vs. 87% with the LNG-IUS [n
= 56]; p
= 0.013).PBAC scores were also used in this trial and suggested a 2.5% reduction in scores with the COC (vs. 87% with the LNG-IUS; p
< 0.001). The discrepancy between MBL measured by the alkaline haematin method and the PBAC score may be due, in part, to the inherent wider variability between PBAC scores and measured MBL with increased volume of blood loss. The study by Higham et al.
also shows the limitations of the PBAC method for the determination of MBL and supports the validity of the methodology used in the present study22
. In the study of women with fibroid-related heavy menstrual bleeding20
, the mean reduction at 12 months in MBL with the COC (n
= 29), also assessed using the alkaline haematin method, was 13% (vs. 91% with the LNG-IUS; p
< 0.001). Pictorial blood loss assessment chart scores corresponded to a 54% reduction in scores (vs. 88% with the LNG-IUS; p
In the randomised cross-over study that included mefenamic acid, naproxen and danazol7
, treatment with a COC containing LNG 150 Hg/EE 30 µg (21/7 regimen; n
= 12) for two cycles resulted in a 43% decrease in mean MBL compared with control cycles (assessed using the alkaline haematin method). In comparison, mefenamic acid, naproxen and danazol reduced mean MBL by 20–39%, 12% and 49%, respectively. There was no significant difference between the COC, mefenamic acid, low-dose danazol or naproxen treated groups.
The reduction in MBL achieved with E2
V/DNG in the present study was accompanied by significant rises in haemoglobin, haematocrit and ferritin levels from baseline, which were significantly greater than with placebo. Although some improvements were observed in the placebo group, these were not significant and may, in part, have been due to the use of iron supplements in a minority of patients during the study. Similarly, treatment with the COC containing NETA 1 mg/EE 20 µg (21/7 regimen) or the LNG-IUS increased mean haemoglobin levels significantly from baseline to 12 months (p
<0.001) in women with idiopathic heavy bleeding21
. Interestingly, the raised haemoglobin levels plateaued after six months of treatment with the LNG-IUS, while the increase observed with the COC occurred at a much slower rate, progressing linearly with time over the 12 months. In contrast, in the two studies comparing the LNG-IUS with a COC containing LNG 150 µg/EE 30 µg (21/7 regimen) over 12 months, haemoglobin and ferritin levels significantly increased with the LNG-IUS, but not with the COC19,20
. No improvements in haemoglobin and ferritin levels were reported in a placebo-controlled study assessing tranexamic acid (n
= 123) over six cycles of treatment despite significant reductions in MBL17
E2V/DNG represents a valid alternative to current treatments for heavy menstrual bleeding. The consideration of contraindications and precautions for use of any individual treatment option, including E2V/DNG, is warranted to ensure a positive benefit-risk balance. The side effect profile of COCs is well established and includes an increased risk for thrombosis which must be carefully balanced against the benefits achieved. Hysterectomy and endometrial ablation are effective treatments, but are associated with surgical risks and are not suitable for women who wish to retain their fertility. Other approved medical therapies, apart from the LNG-IUS, do not provide effective contraceptive protection and use of back-up contraception would be required. However, active comparator studies with E2V/DNG in the treatment of heavy and/or prolonged menstrual bleeding are currently lacking.
No specific safety concerns were noted with E2V/DNG treatment across the two studies pooled. The type and frequency of adverse events reported with E2V/DNG were typical of hormonal contraceptive use.
In conclusion, E2V/DNG rapidly reduces MBL in women with heavy and/or prolonged menstrual bleeding. This effect is achieved at the first withdrawal bleed after treatment initiation and is sustained for the duration of treatment, irrespective of initial MBL.