Do nonsteroidal anti-inflammatory drugs (NSAIDs) effectively reduce heavy menstrual bleeding in premenopausal patients?
NSAIDs are effective for reducing heavy menstrual bleeding in premenopausal patients with menorrhagia when compared with placebo. However, NSAIDs are less effective than tranexamic acid (Cyklokapron) and the levonorgestrel-releasing intrauterine system (Mirena) for reducing heavy menstrual bleeding. Adverse effects of NSAIDs, particularly gastrointestinal effects, are variable in frequency, although typically not severe.1 (Strength of Recommendation: B, based on inconsistent or limited-quality patient-oriented evidence.)
Menorrhagia, or heavy menstrual bleeding, is defined as 80 mL or more of menstrual blood loss per period, although in practice the diagnosis is based on patient report of an amount or frequency of blood loss that interferes with physical or psychosocial well-being. This common gynecologic problem affects 30% of patients of reproductive age. It is an important cause of health care resource utilization and a common condition seen in the primary care setting.2,3 The authors of this Cochrane review sought to determine the safety and effectiveness of NSAID therapy for heavy menstrual bleeding in comparison with placebo and other common treatments.
This Cochrane review included 19 randomized controlled trials involving 759 premenopausal patients with heavy menstrual bleeding without a pathologic or iatrogenic cause.1 One trial took place in Canada, 13 in Europe, two in Australia, and three in Asia, and they involved patients 12 to 55 years of age. The primary outcomes included subjective and objective menstrual blood loss. Because of differing study designs, the meta-analysis included nine trials with 419 patients; the other 10 trials were described individually.
Compared with placebo, mefenamic acid (Ponstel; 500 mg three times daily) taken from start to finish of menses resulted in fewer reports of heavy menstrual bleeding (absolute risk reduction = 56%; 95% CI, 38% to 69%; number needed to treat = 2; 95% CI, 1 to 3; n = 80). In another trial, mefenamic acid (500 mg three times daily) taken from five days before menses to cessation of bleeding reduced measured menstrual blood loss when compared with placebo (mean difference = −124 mL per cycle; 95% CI, −186 to −62; n = 11).
The remaining six of eight trials comparing NSAIDs with placebo were crossover trials; although data could not be pooled, five of six trials showed results favoring NSAIDs. Compared with placebo, mefenamic acid (500 mg three times daily) taken at onset of menses decreased blood loss by 16 to 23 mL in two studies (n = 38) but showed no difference in another (n = 15). Naproxen (250 mg two times daily and 500 mg two times daily) taken at the onset of menses reduced menstrual blood loss by 37 to 54 mL when compared with placebo (two studies; n = 18). Ibuprofen (400 mg three times daily) taken throughout the menstrual cycle decreased menstrual blood loss by 36 mL compared with placebo. Ibuprofen (600 mg daily) taken throughout the cycle did not decrease menstrual blood loss vs. placebo (one study; n = 13). The dosages of mefenamic acid in these studies were higher than the typical recommended dosing in the United States, which is a single dose of 500 mg, followed by 250 mg four times daily for up to three days. NSAIDs are contraindicated in patients with both heavy menstrual bleeding and bleeding disorders caused by platelet aggregation inhibition, and they carry an increased risk of serious cardiovascular thrombotic events and adverse gastrointestinal events, including bleeding.4
Tranexamic acid (1 g four times daily) taken for five days decreased menstrual blood loss compared with mefenamic acid (500 mg three times daily; mean difference = 73 mL per cycle) without a significant difference in subjective report of menstrual blood loss. The levonorgestrel-releasing intrauterine system significantly reduced median menstrual blood loss compared with NSAIDs (82 mL per cycle at three months; one study; n = 51). There was no significant difference in menstrual blood loss between NSAIDs (mefenamic acid, 500 mg three times daily) and ethamsylate (not available in the United States; 500 mg four times daily; two studies; n = 82), oral progesterone (norethindrone, 5 mg two times daily for eight to 10 days of the menstrual cycle; two studies; n = 48), and an oral contraceptive (ethinyl estradiol/levonorgestrel, 30 mcg/150 mcg daily for 21 of 28 days; one study; n = 26). These studies were mostly underpowered. The addition of NSAIDs to other medical treatments has not been well studied. Treatment with danazol is no longer recommended because of the risk of hepatotoxicity, thromboembolism, and intracranial hypertension.
The results of this review, although based on small and mostly underpowered studies without data amenable to pooling, are consistent with the recommendations from the National Institute for Health and Care Excellence, suggesting that the levonorgestrel-releasing intrauterine system be the initial treatment for heavy menstrual bleeding. Nonhormonal medications (tranexamic acid or NSAIDs) and other hormonal therapies (luteal phase progestins or combined oral contraceptives) are considered secondary options with similar effectiveness.3 Desire for contraception, presence of dysmenorrhea, and comorbidities should be considered when discussing treatment options with individual patients.
The practice recommendations in this activity are available at http://www.cochrane.org/CD000400.