Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

COXIBs for dysmenorrhoea

Coxib trials
Results
Comment

Bandolier 120 looked at a systematic review [1] of NSAIDs for dysmenorrhoea, and the question arose about effectiveness of newer coxibs. A search found four randomised trials and the results appear below, but these coxib trials differed from the NSAID trials in the outcomes used, so a brief digression into trial designs is needed before we get to the results.

Pain with dysmenorrhoea usually lasts for about three days, though with considerable individual variation. Trials of analgesics can have various forms. The simplest might be to give the same analgesic for the whole of the painful cycle, and ask a global question concerning efficacy at the end. Women might then be crossed over to a different treatment at the next cycle. This was the design used for NSAID comparisons with placebo, with the outcome of at least moderate pain relief over the cycle.

A variation would be to use the same basic structure, but make more detailed evaluations of pain or pain relief over a limited time during the first day. This was the approach used in the coxib trials.

Coxib trials

Bandolier searched PubMed for randomised trials comparing any coxib (celecoxib, etoricoxib, lumaricoxib, rofecoxib, valdecoxib) with placebo or analgesic in women with painful dysmenorrhoea. Four published trials [2-5] were found (Table 1). All investigated the analgesic efficacy of a first dose in women with moderate or severe menstrual pain. All included both placebo and naproxen sodium 550 mg. Three were double blind, and one [3] was an open study that also had a global outcome over the cycle, but compared with once-daily naproxen sodium 550 mg, rather than the usual dose of twice daily naproxen.



Table 1: Individual randomised trials comparing a coxib with placebo and/or NSAID in dysmenorrhoea



Reference Design Treatments
(women)
Outcomes Results
Morrison et al, 1999 Randomised, double blind, comparison of rofecoxib 25 and 50 mg, naproxen sodium 550 mg and placebo. First dose efficacy, crossover between menses in 127 women Placebo (118)
Rofecoxib 25 mg (115)
Rofecoxib 50 mg (118)
Naproxen sodium 550 mg (122)
First dose, then every 12 hours as needed (once daily rofecoxib 25 mg, twice daily naproxen)
Pain relief and intensity over first 8 hours
Peak relief
Additional medication in first 12 hours
Additional analgesia in 12-72 hours
All active better than placebo, except additional analgesics taken 12-72 hours
No difference between analgesics
No adverse event difference, and no serious adverse events
Sahin et al, 2003 Randomised, open comparison of rofecoxib 25 and 50 mg, naproxen sodium 550 mg and placebo. First dose efficacy, crossover between menses in 55 women Placebo (55)
Rofecoxib 25 mg (55)
Rofecoxib 50 mg (55)
Naproxen sodium 550 mg (55)
First dose, then once daily as needed
Pain intensity over first 24 hours
Overall analgesic efficacy score over the cycle
Pill count
All active better than placebo for pain intensity
"Perfect" global efficacy in 90% with rofecoxib (both doses), and 33% with naproxen
Total pills taken lower for rofecoxib than placebo
64% gastrointestinal adverse events for naproxen
Daniels et al, 2002 Randomised, double blind, comparison of valdecoxib 20 and 40 mg, naproxen sodium 550 mg and placebo. First dose efficacy, crossover between menses in 30 women Placebo (29)
Valdecoxib 20 mg (30)
Valdecoxib 40 mg (30)
Naproxen sodium 550 mg (29)
First dose, then twice daily as needed
Pain relief and intensity over first 8 and 12 hours
Global assessment at 12 hours
Additional medication in first 12 hours
All active better than placebo, except additional analgesics taken over 12 hours, where both valdecoxib doses but not naproxen had fewer than placebo
No difference between analgesics
No adverse event difference
Malmstrom et al, 2003 Randomised, double blind, comparison of etoricoxib 120 mg, naproxen sodium 550 mg and placebo. First dose efficacy, crossover between menses in 73 women Placebo (73)
Etoricoxib 120 mg (73)
Naproxen sodium 550 mg (73)
First dose only
Pain relief and intensity over 24 hours
Analgesia onset
Peak relief
Additional medication in first 12 hours
Both active treatments same, and better than placebo for relief over first 8 hours, and other outcomes. No adverse event difference, and no serious adverse events


Results

Over the first 8-24 hours, all analgesics were better than placebo, but there was no difference between coxibs and naproxen sodium 550 mg for any analgesic measurement, or use of additional analgesic therapy. Only one trial [4] provided a patient outcome of good or excellent response over 12 hours, which gave numbers needed to treat compared with placebo of about 3 for naproxen sodium 550 mg and valdecoxib 40 mg, and 4 for valdecoxib 20 mg. Over a cycle in the open trial [3] both 25 mg and 50 mg daily rofecoxib was better than naproxen sodium 550 mg.

Adverse events did not differ between groups, though a very high rate of gastrointestinal adverse events (64%) was reported for naproxen sodium in the open trial [3]. Discontinuations were infrequent.

Comment

There were 285 women in these four studies, almost all of whom had both placebo and naproxen sodium 550 mg. This is more than the 180 or so women in trials of naproxen and placebo in the systematic review of NSAIDs [1], so these four trials constitute an important addition to knowledge, even if one of them was open. They all come up with the same answer, namely that coxibs at standard doses are equivalent to naproxen sodium 550 mg over eight to 12 hours after the first dose at the onset of menstrual pain.

These outcomes are, of course, important for regulatory authorities, but are less helpful for every day consideration, though once-daily dosing for some coxibs is a theoretical benefit. What was missing was more woman-centred outcomes, and, also, information on adverse events over a cycle, in sufficient numbers of women to spot a difference. Were these outcomes actually captured in the trials, but not reported because they are of little interest to regulatory agencies? It would be fascinating to know. So while we move on a bit, we have still to make the great leap forward.

References:

  1. J Marjoribanks J et al. Nonsteroidal anti-inflammatory drugs for primary dysmenorrhoea (Cochrane Review). In: the Cochrane Library, Issue 4, 2003.
  2. BW Morrison et al. Rofecoxib, a specific cyclooxygenase-2 inhibitor, in primary dysmenorrhoea: a randomized controlled trial. Obstetrics & Gynecology 1999 94:504-508.
  3. I Sahin et al. Dysmenorrhea treatment with a single daily dose of rofecoxib. International Journal of Gynecology & Obstetrics 2003 83: 285-291.
  4. SE Daniels et al. Valdecoxib, a cyxlooxygenase-2-specific inhibitor, is effective in treating primary dysmenorrhoea. Obstetrics & Gynecology 2002 100:350-358. 5 K Malmstrom et al. Analgesic efficacy of etoricoxib in primary dysmenorrhea: results of a randomized, controlled trial. Gynecologic and Obstetric Investigation 2003 56: 65-69.

previous or next story