71 INTRODUCTION
A wide variety of drugs are used to treat abnormal uterine bleeding. Medical therapy is indicated when there is no obvious pelvic abnormality and the woman wishes to retain her fertility. Whereas the number of hysterectomies for menorrhagia has been estimated to have fallen by 36%
between 1989 and 2002/3, and the number of endome- trial ablations is increasing, some women do not wish to have surgery. 1 , 2 Medical treatment avoids surgery, but may have side effects and must be taken long term. 3 Thus, the drug regimen chosen must be effective, have few or mild side effects, and must be acceptable to the patient.
The aims of therapy are to reduce blood loss, reduce the risk of anemia, and improve the quality of life. Menor- rhagia is the commonest cause of iron deficiency anemia in Western women, and thus iron therapy is often indi- cated as well as the options discussed below. It could be argued that menstrual blood loss (MBL) should be reduced to be within the normal range (i.e. less than 80 ml per period). However, women who are keen to avoid surgery may accept a higher loss if they can cope with the flow and any anemia is controlled with iron supplements.
It is important to assess drug therapies in terms of reduction of measured MBL, since, as already discussed in Chapter 5, there is poor correlation between objective and subjective assessment. Well-designed randomized controlled trials provide the best evidence of the efficacy of any intervention, as any differences between groups can be more confidently attributed to differences in treatment.
Medical treatments for abnormal uterine bleeding can be divided into two main classes: non-hormonal and hormonal.
NON-HORMONAL TREATMENTS
Non-hormonal treatments are detailed in Table 11.1.
Non-steroidal anti-inflammatory drugs
Non-steroidal anti-inflammatory drugs (NSAIDs) can be chemically classified into five main groups: salicylates (aspirin), indoleacetic acid analogues (indomethacin), aryl- proponic acid derivatives (naproxen, ibuprofen), fena- mates (mefenamic acid, flufenamic acid, meclofenamic acid), and coxibs (celecoxib, rofecoxib). The first four groups inhibit cyclooxygenase-1 (COX-1) and the last cyclooxygenase-2 (COX-2). Interest in the use of NSAIDs started in the early 1980s when the role of uterine prosta- glandins in the genesis of abnormal uterine bleeding was first examined. Excessive levels of prostaglandins were found in menstrual blood and endometrium and prosta- glandin E (a vasodilator) receptor concentrations were significantly higher in women with menorrhagia. 4– 6
NSAIDs have been evaluated in numerous studies, which to date have been limited to COX-1 inhibitors. It is unlikely that COX-2 inhibitors will be studied in view of concerns of cardiovascular risk. 7 Patients subjective perception of improvement and decrease of blood loss has been shown in many placebo-controlled trials. A meta- analysis including 16 randomized studies has provided no evidence that one NSAID is superior to another. 8 Effectiveness in reducing blood loss
Of the NSAIDs, the fenamates (such as mefenamic acid) have been the most extensively studied. The fenamates have the unique property that they inhibit prostaglandin synthesis and also bind to prostaglandin receptors, which are significantly increased in women with menorrhagia. 6 In addition, fenamates are thought to improve endo- metrial hemostasis. 9 Reductions in menstrual blood flow range from 20% to 50%. 10 , 11 Furthermore, treatment continues to be effective long term. A follow-up of 12–15 months after commencing treatment showed that mefenamic acid continued to be effective. 12
Other NSAIDs such as naproxen, ibuprofen, sodium diclofenac, and flurbiprofen also reduce blood loss.
Medical management of abnormal
The percentage reduction in blood loss varies from 25%
to 47%, depending on the agent and dosage used. 8 , 13 , 14 With regard to dosing regimens, most studies have used treatments starting with the first day of menstruation and continuing for 5 days or until the cessation of bleeding.
Since NSAIDs are prescribed during menstruation, they are suitable for women trying to conceive. Mefenamic acid and naproxen are typically prescribed in a dose of 250–
500 mg two to four times daily; ibuprofen has been studied in doses ranging from 600 mg/day to 1200 mg/day. 8 Side effects
Gastrointestinal symptoms are a common side effect of NSAIDs and these drugs are therefore contraindicated in women with peptic ulceration. Gastrointestinal effects are less likely with mefenamic acid than naproxen. 8 Overall, since NSAID use for menorrhagia is intermittent rather than chronic, as for treatments of conditions such as osteoarthritis, the risk of adverse effects in otherwise healthy women is low.
Use in women with inert or copper-bearing intrauterine contraceptive devices
Inert intrauterine contraceptive devices (IUDs) have been shown to at least double the MBL, whereas copper devices increase MBL by an average of 40–50% over 6–12 months compared with preinsertion values. 15 NSAIDs are effec- tive in reducing blood loss in women with a copper or non-hormonal IUD. Mefenamic acid, 500 mg three times a day from the start of bleeding, reduced MBL by 34% in IUD users with a pretreatment MBL >80 ml and by 23%
in women with a pretreatment MBL <80 ml. 16
Ibuprofen, 400 mg four times a day, from day 1 gave an average reduction of 32% in MBL, although a greater reduc- tion was seen in the inert as opposed to copper coil users. 17 Naproxen also reduces blood loss by a third. 18 Diclofenac reduced MBL by 20% compared with placebo. 19 NSAIDs combined with other agents
There is no published evidence to support simultaneous use of NSAIDs and other agents such as tranexamic acid or progestogens. They are, however, commonly prescribed together in clinical practice. This pragmatic approach may be of benefit, since NSAIDs also reduce menstrual pain. 20
Comparison of NSAIDs and other agents
Randomized studies comparing NSAIDs with other agents for abnormal uterine bleeding suggest both dana- zol and tranexamic acid to be superior with respect of decreasing blood loss 21 (Table 11.2). However, danazol and tranexamic acid are more likely to cause adverse events when compared with NSAIDs.
Table 11.1 Non-hormonal treatments for abnormal uterine bleeding
Non-steroidal anti-inflammatory drugs Mefenamic acid
Meclofenamic acid Naproxen
Ibuprofen Flurbiprofen Diclofenac Antifibrinolytics Tranexamic acid Other
Ethamsylate
Table 11.2 Comparison of medical treatments for abnormal uterine bleeding
Treatment Reduction of menstrual
blood loss
Non-steroidal anti-inflammatory 20–50%
drugs
Tranexamic acid 47–54%
Etamsylate Possibly 13% – not clinically significant Oral progestogens for 21 days 30–90%
Oral progestogens, luteal Ineffective adminstration
Long-acting progestogen 50–66% of women with menorrhagia experience amenorrhea between 1 and 2 years of use Levonorgestrel intrauterine 74–97%
system
Combined oral contraceptive pill 43%
Danazol 50–80%
Gonadotropin-releasing > 90%
hormone agonists Adapted from Reference 20.
In summary, NSAIDs can be an effective first-line treat- ment in abnormal uterine bleeding. The degree of reduc- tion of menstrual blood loss is modest compared to other agents, but NSAIDs have a low profile of adverse effects in otherwise healthy women. An additional beneficial effect is that these drugs also alleviate menstrual pain.
Antifibrinolytics
The endometrium possesses an active fibrinolytic system.
An increase in the levels of plasminogen activators, a group of enzymes that cause fibrinolysis, has been found in the endometrium of women with heavy menstrual bleeding compared to those with normal menstrual loss. 22 Plasminogen activator inhibitors (antifibrinolytic agents) have therefore been promoted as a treatment for menorrhagia.
Effectiveness in reducing blood loss
Tranexamic acid, a synthetic derivate of the amino acid lysine, exerts its antifibrinolytic effect through the revers- ible blockade of plasminogen. 23 , 24 Tranexamic acid reduces blood loss by 40–50%. 25–27 Since it is taken only during menstruation, tranexamic acid is therefore suitable for women trying to conceive.
Side effects
Side effects of antifibrinolytic drugs, mainly gastrointesti- nal and dose-dependent effects, are reported by about one-third of patients. 28 Adverse effects can be reduced by limiting the number of days on which the drug is taken.
As about 90% of menstrual blood is lost during the first 3 days, medication could be limited to that time.
The main reported adverse effect of tranexamic acid is thromboembolism. Concerns were fuelled by single case reports about cerebral sinus thrombosis and central venous stasis retinopathy. 29 , 30 However, long-term studies in Scandinavia have shown that the incidence of thrombosis in women treated with tranexamic acid is comparable to the spontaneous frequency of the condition in the female population. 31
Use in women with inert or copper-bearing intrauterine contraceptive devices
Tranexamic acid is also effective in women with a copper or an inert intrauterine contraceptive device. 19 ,32 The effect is dose dependent, with a 54% reduction in MBL when 3 g daily was used compared to 71% when initially 6 g was given for the first 5 days of bleeding followed, if necessary, by 3 g.
Comparison of antifibrinolytics and other agents Comparative studies have shown tranexamic acid to be superior to NSAIDs, oral progestogens, and etham- sylate in reducing menstrual blood loss. 11 , 27 However antifibrinolytics do not alleviate menstrual pain.
Thus, antifibrinolytics should be considered as a first- line treatment for menorrhagia. However, it is probably prudent not to use an antifibrinolytic with the combined oral contraceptive pill (see below).
Ethamsylate
Ethamsylate (ε−aminocaproic acid) is believed to act by reducing capillary fragility, although the precise mecha- nisms are unknown. However, studies have produced conflicting results, with some showing an increase in blood loss. 11 ,33 Its use is not recommended in the UK. 20 It has been studied in women with IUDs, but the reduc- tion of blood loss was small (<20%). 34 , 35
HORMONAL TREATMENTS
Hormonal treatments are detailed in Table 11.3.
Table 11.3 Hormonal treatments for abnormal uterine bleeding
Oral progestogens Norethisterone
Medroxyprogesterone acetate Dydrogesterone
Intrauterine progestogens Levonorgestrel IUD Progestasert IUD
Combined estrogen/progestogens Oral contraceptives
Hormone replacement therapy Other
Danazol Gestrinone GnRH analogues Mifepristone
becomes common and hemoglobin levels increase. 46 After 1 year of use, 45–50% of women are likely to be amenorrheic. 44,46 There are concerns about the use of DMPA increasing the risk of osteoporosis. 47 The best- available evidence suggests that the amenorrhea induced by DMPA contraception is associated with a 5–10% loss of bone, which, however, is not progressive. 48 , 49 Much depends on the patient's risk factors for osteoporosis, such as family history and smoking. In any case, DMPA prob- ably should be discontinued at age 40 years old to allow resumption of ovarian cycling for the remaining 10 years or so up to the natural menopause. The long-term skeletal effects of DMPA in teenagers who have yet to achieve peak bone mass is uncertain. Any fall in bone mineral density (BMD) on starting treatment, however, seems to reverse rapidly on stopping DMPA.
Subdermal implants (such as Implanon) that release etonogestrel over several years are another way of admin- istering long-acting progestogens for contraceptive pur- poses. Like DPMA, bleeding disturbances such as irregular bleeding and periods of prolonged bleeding are common. 50 Intrauterine progestogens
Progesterone or progestogen-releasing intrauterine sys- tems (PPRIUS) were initially introduced in an effort to reduce IUD expulsion, by the addition of ‘uterine-relax- ing hormones’. 51 Subsequently, prolonged use of both the Progestasert (the first hormonally impregnated device releasing 65 µg of progesterone per day) and Mirena (a device releasing 20 µg of levonorgestrel per day) intrauter- ine systems was associated with a profound reduction in MBL in women using the IUD for contraception. Twenty percent of the women using the levonorgestrel-releasing intrauterine system were amenorrheic after 1 year’s use while still continuing to ovulate. 51 The levonorgestrel-releasing intrauterine system (Mirena) is discussed in detail in Chapter 13.
Progestasert requires reinsertion approximately annu- ally. The levonorgestrel-releasing intrauterine system has to be replaced every 5 years.
Estrogen/progestogen combinations
Combined oral contraceptive pills consisting of an estro- gen/progestogen combination are often used to reduce MBL. When taken in a cyclical fashion, they induce regu- lar shedding of a thinner endometrium and inhibit ovula- tion. The exact endometrial mode of action of contraceptive pills is unclear, and probably involves induction of endometrial atrophy.
From clinical experience, the beneficial effect of com- bined oral contraceptive preparations on menorrhagia is Progestogens
The use of progestogens is based on the erroneous con- cept that women with menorrhagia principally have ano- vulatory cycles. 36 However, many studies have shown that most women with regular excessive menstrual bleeding have normal ovulatory cycles. 37 , 38
A variety of routes of administration and dose sched- ules have been used, ranging from intermittent luteal phase oral administration, through intramuscular injec- tion, to continuous local administration by an IUD, each of which has a different efficacy in distinct clinical situa- tions; the last mentioned also provides contraception.
Owing to the lack of objective randomized controlled trials the use of synthetic progestogens has been previously largely empirical.
Oral progestogens
Traditionally, administration was in the luteal phase and studies have mainly examined norethisterone. Studies with measured menstrual loss with luteal administration for 7 days of norethisterone, 5 mg twice daily, show either a slight decrease or even an increase in flow. 39 This regi- men is not recommended in the UK. 20 However, nore- thisterone, 5 mg three times daily from day 5–26, is effective. 40 Of note, there have been no studies comparing progestogens to placebo. 39 Side effects include weight gain, headache, and bloatedness. There are also concerns that progestogens used in doses greater than those needed in progestogen-only contraception increase the risk of venous thromboembolism. 41 Limited data suggest that medroxyprogesterone acetate reduces MBL, but there are no publications to date regarding dydrogesterone. 42 , 43
Oral progestogens at very large doses, such as 30 mg/
day, can be successfully applied to control torrential bleeding. This is usually effective within 24–48 hours, when the dose can be reduced and then finally stopped over the next few days, when another, usually lighter bleed, will occur.
Depot or implant progestogens
Continued use of long-acting progestogens renders most women amenorrheic and therefore could be considered for use in menorrhagia. However, there are no published data with MBL measurements.
Medroxyprogesterone acetate is available as a depot injection (DMPA) for contraception at a dose of 150 mg administered every 3 months. It may cause unpredictable, irregular spotting and bleeding in the first few months of use and can cause heavy bleeding in 1–2% of women. 44– 46 However, with repeated administration, amenorrhea
Given these side effects, danazol is probably best restricted to women awaiting surgery. It can also be used as a short- term endometrial thinning agent prior to endometrial ablation.
Gestrinone
Gestrinone is a 19-nortestosterone derivative which has antiprogestogenic, antiestrogenic, and androgenic activity.
Gestrinone was shown in a placebo-controlled study to successfully diminish MBL and to cause amenorrhea in patients suffering from menorrhagia. 63 However, most women treated with gestrinone experience androgenic side effects such as acne, hirsutism, and weight gain. 57,63 These side effects preclude long-term therapy. Again, effec- tive contraception needs to be used to prevent potential fetal virilization.
Gonadotropin-releasing hormone agonists
Gonadotropin-releasing hormone (GnRH) agonists administered continuously create a reversible hypogonad- otropic state by down-regulation of GnRH receptors, which blocks ovarian function and causes a hypoestro- genic state. 57 GnRH agonists reduce uterine volume by 40–60%, and many investigators have evaluated their effectiveness in treating bleeding associated with leiomyo- mas. GnRH agonists frequently induce amenorrhea. 57
However, there are significant disadvantages to the use of GnRH analogues. The duration of treatment of these agents is limited to 6 months, owing to rapid bone demin- eralization associated with estrogen withdrawal. Further- more, the occurrence of side effects associated with the hypoestrogenic state, particularly hot flushes and vaginal dryness, reduce the woman’s quality of life. 57
Concerns about bone loss and poor tolerability of GnRH agonists have led to a number of studies on estro- gen and progestogen ‘addback therapy’ during GnRH agonist treatment. However, there are still concerns about bone loss. 64
The side effects of GnRH analogues usually preclude their long-term use. However, they are highly effective in short-term therapies such as for preoperative endometrial thinning or fibroid shrinkage.
Antiprogestogens
Mifepristone (RU-486) is a synthetic 19-norsteroid with antiprogestogen activity that has been shown to inhibit ovulation and to disrupt endometrial integrity. 65 Mife- pristone was recently shown to act as a contraceptive agent and to induce amenorrhea at a dose of 5 mg/day in the majority of women studied. 66 However, there are very well known. However, placebo-controlled random-
ized controlled trials with adequate patient numbers, duration of at least three to six cycles and adequate follow-up have not been conducted. 52
One small study of women taking an oral contracep- tive containing 30 µg ethinyl estradiol showed a 43%
reduction in MBL compared to baseline. No significant difference was found between groups treated with an estrogen/progestogen combination, mefenamic acid, low- dose danazol, or naproxen. 14 Other studies found users of oral contraceptives less likely to experience heavy men- strual bleeding or anemia. 53
However, some trials employed a higher dose than the 30–35 µg ethinyl estradiol of preparations currently in use. It is, therefore, less clear whether low-dose prepara- tions are as effective in reducing MBL as the higher-dose combinations, and whether particular progestogens make any difference. Although there is some indication that the frequency of menstrual disturbances, such as spotting and breakthrough bleeding, is less when the oral contraceptive contains a relatively high dose of progestogen, no evi- dence is available to suggest that changing the progesto- gen influences total MBL.
The combined oral contraceptive is probably best used when contraception is also required and it will also reduce menstrual pain. 54 This is partly caused by a fear of throm- boembolic events, particularly in the older age groups. 55 Although withdrawal bleeds induced by sequential estrogen/
progestogen hormone replacement therapy regimens are not excessive, there have been no trials in menorrhagia. 56 Androgens
Danazol
Danazol is an isoxazol derivative of 17α-ethinyl testoster- one, which acts on the hypothalamic–pituitary–ovarian axis, suppressing ovulation, and directly on the endo- metrium, leading to atrophy. 57
Danazol was evaluated for the treatment of menor- rhagia in various randomized trials and compared with several other medical treatments. Danazol reduces MBL by up to 80% in a dose-dependent manner. 58–60 It usually leads to amenorrhea at doses >400 mg/day. 61 Danazol reduces MBL more effectively than NSAIDs or oral progestogens. 60 ,62
Although effective, the clinical use of danazol is lim- ited by its androgenic side effects, which are experienced by up to three-quarters of the patients. Side effects include weight gain, oily skin, acne, and deepening of the voice. 57 In addition, women must be advised to use barrier meth- ods of contraception because of potential virilization of a fetus if pregnancy occurs while on danazol treatment.
concerns that it induces endometrial hyperplasia, and further research is required. 67
DECISION AIDS AND EDUCATION PACKAGES
Increasing patient participation in treatment decision making is important in view of the wide choices available.
Decision aids come in a variety of formats, including leaf- lets, audiotapes, decision boards, computer programs, videos, websites, and structured interviews. These were examined in a randomized controlled trial of 894 women with 2 years of follow-up. 68 Women were randomized to the control group, information-alone group (booklet and video), or information ⫹ interview group (interview).
Hysterectomy rates were lower for women in the inter- view group (38%) (adjusted odds ratio [OR], 0.60; 95%
confidence interval [CI], 0.38–0.96), than in the control group (48%) and women who received the information alone (48%) (adjusted OR, 0.52; 95% CI, 0.33–0.82).
The interview group had lower mean costs (US$1566) than the control group (US$2751) (mean difference US$1184; 95% confidence interval [CI], US$684–2110) and the information group US$2026 (mean difference US$461; 95% CI US$236–696. Thus, providing women with information alone did not affect treatment choices;
however, the addition of an interview to clarify values and elicit preferences had a significant effect on women’s management and resulted in reduced costs.
Education packages in primary care have also been evaluated. For example in East Anglia in the UK a
randomized trial of 100 general practices was undertak- en. 69 The interventions were an educational package based on principles of ‘academic detailing’ with independent academics and were given in small practice-based inter- active groups with a visual presentation, a printed evidence-based summary, a graphic management flow chart, and a follow-up meeting at 6 months. There were significantly fewer referrals (20% vs 29%; OR ⫽ 0.64;
95% CI 0.41–0.99) and a significantly higher use of tranexamic acid (OR ⫽ 2.38; 95% CI 1.61–3.49).
CONCLUSION
In January 2007 the National Institute for Health and Clinical Excellence published guidelines on heavy men- strual bleeding. 70 The guideline makes recommendations on a range of medical and surgical treatment treatment options. They recommended that treatments should be considered in the following order:
(1) levonorgestrel-releasing intrauterine system provided long-term (at least 12 months) use is anticipated;
(2) tranexamic acid or non-steriodal anti-inflammatory drugs or combined oral contraceptives;
(3) norethisterone (15 mg) daily from days 5 to 26 of the menstrual cycle, or injected long-acting progestogens.
What would suit an individual patient would of course depend on her contraceptive needs and fertility goals.
References
1. Reid PC , Mukri F . Trends in number of hysterectomies per- formed in England for menorrhagia: examination of health epi- sode statistics, 1989 to 2002-3. BMJ 2005 ; 330 : 938 – 9 . 2. Reid PC . Endometrial ablation in England–coming of age? An
examination of hospital episode statistics 1989/1990 to 2004/2005.
Eur J Obstet Gynecol Reprod Biol 2006 Oct 11 ; [Epub ahead of print] .
3. Oehler MK , Rees MC . Menorrhagia: an update. Acta Obstet Gynecol Scand 2003 ; 82 : 405 – 22 .
4. Rees MCP , Anderson ABM , Demers LM , Turnbull AC . Pros- taglandins in menstrual fluid in menorrhagia and dysmenor- rhoea. Br J Obstet Gynaecol 1984 ; 91 : 673 – 80 .
5. Rees MCP , Anderson ABM , Demers LM , Turnbull AC . Endo- metrial and myometrial prostaglandin release during menstrual cycle in relation to menstrual blood loss. J Clin Endocrinol Metab 1984 ; 58 : 813 – 18 .
6. Adelantado JM , Rees MCP , Lopez-Bernal A , Turnbull AC . Increased uterine prostaglandin E receptors in menorrhagic women. Br J Obstet Gynaecol 1988 ; 95 : 162 – 5 .
7. McGettigan P , Henry D . Cardiovascular risk and inhibition of cyclooxygenase: a systematic review of the observational studies of selective and nonselective inhibitors of cyclooxygenase 2.
JAMA 2006 ; 296 : 1633 – 44 .
8. Lethaby A , Augood C , Duckitt K . Nonsteroidal anti-inflamma- tory drugs for heavy menstrual bleeding. Cochrane Database Syst Rev 2000 ; 2 : CD000400 .
9. van Eijkeren MA , Christiaens GC , Geuze HJ , Haspels AA , Sixma JJ . Effects of mefenamic acid on menstrual hemostasis in essential menorrhagia. Am J Obstet Gynecol 1992 ; 166 : 1419 – 28 . 10. Cameron IT , Haining R , Lumsden MA , Thomas VR , Smith
SK . The effects of mefenamic acid and norethisterone on mea- sured menstrual blood loss. Obstet Gynecol 1990 ; 76 : 85 – 8 . 11. Bonnar J , Sheppard BL . Treatment of menorrhagia during men-
struation: randomised controlled trial of ethamsylate, mefenamic acid, and tranexamic acid. BMJ 1996 ; 313 : 579 – 82 .
12. Fraser IS , McCarron G , Markham R , Robinson M , Smyth E . Long-term treatment of menorrhagia with mefenamic acid.
Obstet Gynecol 1983 ; 61 : 109 – 12 .