Daniel Carlat, MDDr. Carlat has disclosed that he has no significant relationships with or financial interests in any commercial companies pertaining to this educational activity.
If you really want to understand Premenstrual Dysphoric Disorder (PMDD), you'd better have a pretty good understanding of the normal menstrual cycle.
Its purpose is to prepare the uterus for pregnancy. It lasts 28 days on average. The first day of the cycle is defined as the first day of menstrual blood flow (the "period"), which lasts 4 to 5 days. Then, the "follicular (proliferative) phase" gets into gear, during which rising estrogen levels cause endometrial thickening. Day 14 brings the "LH surge" and ovulation, and with it the onset of the "luteal (secretory) phase."
From the standpoint of psychiatry, the real action occurs during the latter part of the luteal phase, roughly days 22-28 of the cycle. During this week, the lining of the uterus is thick and vascularized, essentially in a state of high readiness for the implantation of a fertilized egg. However, if no such embryo-to-be arrives, hormonal events occur that lead to a dramatic decrease in levels of estrogen and progesterone, which trigger the shedding of the uterine lining, or menstrual bleeding. During this period of a week or so, the symptoms commonly termed "premenstrual syndrome" (PMS) affect about 75% of premenopausal women. These symptoms include abdominal cramping (from uterine contraction); water retention and bloating; and irritability and moodiness. If these symptoms are joined by neurovegetative symptoms of depression and are severe enough to significantly impair functioning, DSM-IV provides the diagnosis Premenstrual Dysphoric Disorder (PMDD), which strikes about 5% of premenopausal women.
From this description, it may seem obvious that PMS is caused by "fluctuations in hormone levels," a phrase that most of us have used when educating our patients about PMS. While hormones undoubtedly have something to do with PMS, the best research minds of our generation have had the devil of a time elucidating the exact mechanism. In one study, for example, researchers artificially truncated the luteal phase by given mifepristone, a progesterone antagonist, to women with PMS. This procedure had no effect on PMS symptoms at all, either in the timing or severity (NEJM 1991; 324:1174-1179.)
We're a ways from understanding the causes of PMDD, but what about diagnosis and treatment? Much is made in the literature about elaborate screening tests and daily diaries as diagnostic aids, but any of us in the trenches know that PMDD is easy to diagnose the old fashioned way--asking the right questions and listening carefully to answers. Good screening questions include, "Do you have really bad PMS?" and "Do you get much more irritable, tense, or depressed for a week before your period?"
Three SSRIs are now FDA approved for PMDD: Prozac (fluoxetine, marketed for PMDD as "Sarafem"), Zoloft (sertraline), and Paxil CR (paroxetine). All three of these drugs were evaluated using standard placebo-controlled double-blind research methods. All three of these SSRIs work well for PMDD. All have been shown effective when dosed continuously and when dosed "intermittently," which in research trials has been defined as beginning on day 14 of the cycle and ending on the first day of menstrual bleeding.
Here are some specific notes on each agent.
Sarafem has been evaluated in three fixed-dose trials. For continuous dosing, 20 mg QD works just as well as 60 mg QD, and 20 mg is better tolerated (NEJM 1995; 332:1529-1534). For intermittent dosing, 20 mg has been compared with 10 mg and placebo; only 20 mg beat placebo in that trial (Obstet Gynecol 2002; 100:435-444). The creation of "Sarafem," by the way, was a particularly ingenious way of extending Prozac's patent life, which had expired in 2001. Eli Lilly simply changed Prozac's color from green and white to pink and lavendar, provided the FDA with PMDD efficacy data, and won a "method-of-use" patent, allowing it marketing exclusivity until sometime in 2004.
Zoloft has only been tested in studies allowing clinicians to vary the dose over a given range ("flexible dose studies"), which deprives us of the ability to compare the efficacy of different doses. Nonetheless, Zoloft works well for PMDD, with an average effective dose ranging from 75 mg QD to 100 mg QD. One study made a direct comparison of continuous with intermittent (i.e., two weeks a month) dosing. Both dosing strategies beat placebo equally, but counterintuitively, during the last month of the 3-month study, those women who received Zoloft intermittently actually reported more side effects than the continuously- dosed group. One possible explanation is that continuous dosing gives the body a chance to build up a tolerance to the side effects (Am J Psychiatry 2004; 161:343-351). So think twice before trying to sell your patients on intermittent dosing based on a "fewer side effects" argument!
Paxil CR has been evaluated using the fixed-dose trials that TCR loves so much, with the conclusion that 12.5 mg QD works as well as 25 mg QD, whether it is dosed continuously or intermittently.
Do the other serotonergic antidepressants work for PMDD?
Undoubtedly. Celexa (citalopram) was shown to be superior to placebo, but oddly enough, intermittent dosing outperformed continuous dosing (J Clin Psychopharmacol 1998; 18:390-398). Effexor (venlafaxine) has one positive controlled trial under its belt (Obstet Gynecol 2001 98:737-44). Luvox (fluvoxamine) and Serzone (nefazodone) are also likely effective, but the data are thus far not placebo-controlled. Incidentally, Bristol-Myers Squibb recently announced it would stop shipping Serzone due to insufficient sales (presumably caused by hepatotoxicity concerns), but the FDA has not withdrawn it, meaning that the generic nefazodone is still available to your patients.
What about non-prescription treatments? Several work, but none have been compared head-to-head with SSRIs. Placebo-controlled trials have endorsed the following strategies: Vitamin B6, up to 100 mg QD (BMJ 1999; 318: 1375-1381); calcium carbonate, 1200 mg QD (Am J Obstet Gynecol 1998; 179:444-452); and the homeopathic remedy agnus castus fruit extract (BMJ 2001; 322:134-138).