Menstrual Cycle–Influenced Disorders

The acronyms PMS for premenstrual syndrome and PMDD for premenstrual dysphoric disorder refer to the same pathologic process at opposite ends of the symptom spectrum (Figure 36-1). In both PMS and PMDD, patients experience adverse physical, psychological, and behavioral symptoms during the luteal phase of the menstrual cycle. There is a crescendo of symptom intensity up to the time that menses begins, with quick resolution thereafter. Some patients have a brief surge of symptoms at the time of ovulation in midcycle.

FIGURE 36-1 Spectrum of premenstrual syndromes. PMDD, premenstrual dysphoric disorder; PMS, premenstrual syndrome.

As many as 80% of regularly ovulating women experience some degree of physical and psychological premenstrual symptomatology. Those who have mild to moderate symptoms are said to have PMS. In 5% or less of women, these symptoms are so severe that they seriously interfere with usual daily functioning or personal relationships. These women are characterized as having PMDD.

Common symptoms reported by patients include depressed mood, anxiety, affective lability and irritability, decreased interest in regular activity, difficulty concentrating, fatigue, change of appetite, sleep disturbance, and feelings of being overwhelmed. Physical symptoms include breast swelling and tenderness, bloating (a sense of abdominal swelling), weight gain, edema, and headache. The diagnosis of these disorders is confirmed by the predominant occurrence of symptoms in the luteal phase as documented on a menstrual calendar of two consecutive cycles.

A formal set of diagnostic criteria has been proposed in the fourth text revision of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) of the American Psychiatric Association for PMDD (Table 36-1). Although the DSM-IV definition of PMDD specifies that this is not just an exacerbation of another disorder, the dividing line between PMDD and other neuropsychiatric disorders is not so clear cut. For example, 46% of PMDD patients have a history of a prior major depressive episode. Moreover, patients with PMDD and clinical depression share similar sleep electroencephalogram alterations, and they are both responsive to the selective serotonin reuptake inhibitor (SSRI) antidepressants.

TABLE 36-1 CRITERIA FOR PREMENSTRUAL DYSPHORIC DISORDER

Although PMS and PMDD patients and controls do not differ in their average cyclic levels of sex steroids, gonadotropins, prolactin, or cortisol, there exists a strong basis to believe that these disorders have a hormonal rather than purely psychologic basis. First, abolition of the menstrual cycle with gonadotropin-releasing hormone (GnRH) agonists, pregnancy, menopause, or spontaneous anovulation provides symptomatic relief, whereas sequential ovarian hormone therapy in hypogonadal patients can induce PMS and PMDD symptoms. Second, cycles with higher luteal phase levels of estradiol are associated with more severe symptoms.

The physiologic mechanism that results in the occurrence of PMS and PMDD is not well understood. Evidence exists that the phenomenon arises, in part, from atypical metabolism of progesterone that results in lower levels of the steroid allopregnenolone within the central nervous system. Allopregnenolone is a neuroactive progesterone metabolite that modulates central gamma-aminobutyric acid (GABA) receptors that modify behavior, emotion, subjective perceptions, and response to stress. In addition, the GABA and serotoninergic neurons may be inherently dysfunctional in PMS and PMDD patients, especially in those with severe depressive symptoms—hence the overlap between PMDD and clinical depression. Major depressive disorder (MDD) persists, however, on a daily basis for weeks without a relationship to the menstrual cycle. MDD, which usually does not have premenstrual physical symptoms, may however be exacerbated during the luteal phase of the menstrual cycle. In such cases, both PMDD and MDD need to be treated (Table 36-2).

Research performed to determine the best therapy for this disorder is problematic, because of the subjective nature of the condition as well as the wide variation in the severity of the symptoms from one cycle to the next. In addition, external influences at work and at home will affect the severity of the symptoms. Finally, placebo interventions produce significant initial benefits in most PMS and PMDD studies. All these considerations necessitate prolonged studies, which are expensive and infrequently performed.

Most women who could be characterized as having PMS should be treated individually and conservatively, with reassurance and mild diuretics for symptoms such as bloating. The mild anxiety that frequently occurs with PMS may be treated with an antianxiety agent such as buspirone. At present, the most effective therapy studied for women with PMDD is the SSRI class of antidepressants. Fluoxetine taken at dosages of 20 to 60 mg per day during the luteal phase of the cycle provides significant symptom improvement in 50% to 60% of patients. Sertraline at 50 to 150 mg per day is equally effective. Side effects of the SSRIs are usually self-limited and include insomnia and sexual dysfunction.

Other preparations have been effective in at least one randomized controlled trial. They include calcium carbonate at 1200 mg per day, for control of mood and behavioral symptoms; spironolactone at 100 mg per day, for mood and bloating; and buspirone at 25 to 60 mg per day, for premenstrual anxiety. Danocrine and bromocriptine are effective for the treatment of cyclic mastalgia. Pyridoxine (vitamin B₆) at 50 to 100 mg a day has demonstrated mixed results in clinical trials.

GnRH agonists used with estrogen and progestin “add back” to minimize hot flashes are effective in eliminating PMS and PMDD symptoms. However, this is an expensive therapeutic approach.

Treatments that have been demonstrated to be ineffective in randomized controlled trials include oral or vaginal progesterone and conventional use of combined oral contraceptives. With the latter, patients have PMS-like symptoms during the placebo week.

Recent studies have shown benefit from the continuous use, or 24- out of 28-day use, of an oral contraceptive containing the progestin drospirenone.

Migraine headaches, which are believed to result from sequential intracranial vasoconstriction and vasodilation, are known to be influenced by menstrual cycling. They are 2 to 3 times more common in women than in men. They improve in about 80% of patients during pregnancy but recur postpartum. Usually, migraines resolve after the onset of the menopause. Sixty percent of women who suffer migraine link the occurrence of their attacks to the menstrual cycle, and 7% exclusively have migraines on the 2 days before or after the onset of menstruation. Menstrual migraines usually occur without a preceding aura and are more long-lasting and resistant to treatment than migraines occurring at other times in the menstrual cycle.

The link between migraine headaches and the hormonal changes of the menstrual cycle is believed to be the phenomenon of estrogen withdrawal. Evidence for this derives from several observations: first, a small proportion of women with menstrual migraine have an upsurge in headache frequency following the preovulatory estradiol surge; second, exogenous estrogen reduces the incidence of migraines; and third, exogenous progesterone may delay the onset of menstruation without preventing the migraine attacks.

Several mechanisms have been proposed to explain why estrogen withdrawal produces migraine headaches. They include abnormal platelet aggregation, central nervous system endogenous opioid dysregulation, and stimulation of increased synthesis of prostaglandin in the central nervous system.

Standard treatment of migraine headaches includes triptans, nonsteroidal antiinflammatory drugs, and ergotamines. Drugs used for the short-term prophylaxis of menstrual migraines can be taken 3 to 5 days before and after the onset of menses. They include nonsteroidal antiinflammatory drugs and triptans. Monitoring of the menstrual cycle by basal body temperature charting or the use of a luteinizing hormone surge detector kit permits the initiation of these agents for more effective headache prevention.

Several hormonal protocols may also be effective in preventing menstrual migraines. For short-term prophylaxis, 100-μg transdermal estrogen patches begun 48 hours before anticipated menses and continued for 3 to 6 days have been shown to be effective. Continuous oral contraceptive pills for intervals of 2 to 4 months (with symptomatic treatment during withdrawal between intervals) provide long-term prevention of migraines.

Seventy percent of female epileptic patients report an increased incidence of seizures premenstrually. Fourteen percent of female epileptic patients have catamenial epilepsy in which seizures only occur in the perimenstrual phase of the cycle. This includes all varieties of epilepsy. In these women, the onset of epilepsy is usually at the time of, or shortly after, menarche. Eight-four percent of catamenial epileptic patients have significant premenstrual syndrome symptoms, in contrast to a 22% incidence of PMS in epileptic patients whose seizures do not correlate with the menstrual cycle.

Two mechanisms are thought to underlie the phenomenon of catamenial epilepsy. The first is a direct effect on the neurons of the brain of the reduced progesterone-to-estradiol ratio. In vitro, estradiol lowers the seizure threshold of many varieties of neurons whereas progesterone raises the threshold, making a seizure more likely. Thus, catamenial epilepsy reflects the effect of a reduced progesterone (and allopregnenolone) concentration or progesterone-to-estradiol ratio during the late luteal phase of the menstrual cycle. This correlates well with several clinical observations: first, some patients with catamenial epilepsy also suffer exacerbations during the preovulatory estradiol surge; second, seizure activity is prone to increase in anovulatory cycles, which may be managed with the use of clomiphene; and third, seizure activity may decrease in incidence after menopause.

A second mechanism explaining this disorder is a reduction in serum levels of anticonvulsants during the late luteal phase. This is believed to be mediated by increased hepatic mono-oxygenase activity resulting directly from reduced sex steroid levels (primarily progesterone). This is the rationale for the treatment option of closely tracking the menstrual cycle and determining anticonvulsant concentrations in the late luteal phase, so that drug dosage may be altered when necessary.

The anticonvulsant effect of progesterone is the basis of using Depo-Provera, progestin-only oral contraceptive pills, or premenstrual progesterone suppositories (50 to 400 mg twice daily) to reduce seizure activity. Ganaxolone, an allopregnenolone analogue taken cycle days 21 through 3 of the next cycle, has also been found to be an effective seizure prophylactic. GnRH agonist therapy has been helpful for intractable cases. Combined oral contraceptives have inconsistent effects, with some patients suffering exacerbations during the placebo week. Moreover, pill efficacy is reduced by anticonvulsants, resulting in a higher contraceptive failure rate with low-dose preparations.

In some individuals, the following conditions may be influenced by the menstrual cycle: rheumatoid arthritis, irritable bowel syndrome, hereditary angioedema, aphthous ulcers, Behçet’s syndrome, acute intermittent porphyria, paroxysmal supraventricular tachycardia, multiple sclerosis, glaucoma, urticaria, erythema multiforme, and myasthenia gravis. Interestingly, in the case of myasthenia gravis, 25% to 50% of female patients improve premenstrually.