Epilepsy poses unique problems for women, but not all of them are appreciated by neurologists. The risks associated with pregnancy are generally understood. But there appears to be a lack of awareness about other gender-specific problems for women with epilepsy, including bone loss, aspects of sexual dysfunction, and features of polycystic ovary syndrome that include metabolic disorders and weight gain, epilepsy experts told Neurology Today.
Now there is enough information to aid in understanding the basis for these health risks and what the concerns might be, said Martha Morrell, MD, Clinical Professor of Neurology at Stanford University and Chief Medical Officer of NeuroPace, Inc., a company developing neurostimulators for epilepsy treatment. Dr. Morrell was formerly Director of the Columbia Comprehensive Epilepsy Center at New York Presbyterian Hospital and Caitlin Tynan Doyle Professor of Clinical Neurology.
Antiepileptic drugs (AEDs) can have negative effects on fertility, gynecological health, contraceptive efficacy, pregnancy outcome, and bone health, she said. Speaking last summer at a symposium at the American Psychiatric Association (APA) Annual Meeting in New York City, Dr. Morrell outlined the risks associated with AEDs for women and explained that AEDs have different side-effect profiles.
“In most cases, you have a choice of AEDs. The challenge is to understand not only the disorder, but the tolerability and effects of AEDs over time,” she said.
SEX STEROID INTERACTION
Some AEDs induce cytochrome (CYP) P450 – an enzyme responsible for metabolizing toxic hydrocarbons – which in turn increases the metabolism of sex steroid hormones and interferes with the effectiveness of oral contraceptives. According to a 20-year old survey, use of CYP-450 enzyme-inducing AEDs leads to unplanned pregnancies in 6 to 15 percent of patients per year.
“The data on unintended pregnancies are from the mid-1980s,” she commented. “Now with lower-dose oral contraceptives, the rate may be even higher.” When asked about new data on this, Allison M. Pack, MD, Assistant Professor of Clinical Neurology at Columbia University, noted that no studies have been done since then and none are planned.
Among newer AEDs, carbamazepine, oxcarbazepine, and topiramate are CYP-450 enzyme-inducers, Dr. Morrell explained. The CYP-450 enzyme is induced at higher dosages of oxcarbazepine, greater than 1200 mg per day, and for topiramate, greater than 200 mg per day, Dr. Morrell said. Lamotrigine, gabapentin, levetiracetam, tiagabine, and valproate do not induce CYP 450 and, therefore, do not interfere with the efficacy of oral contraceptives.
If a woman is taking an enzyme-inducing drug, Dr. Morrell suggested using barrier methods of contraception, or oral contraceptives containing more than 35 mcg of estrogen (Neurology 1998;51:944–948).
Surveys suggest that 65 to 85 percent of primary care doctors are unaware of the adverse interaction between AEDs and oral contraceptives. Dr. Morrell suggested consulting with patients' obstetrician-gynecologists about the effects of AEDs on contraceptive choices and pregnancy.
EFFECTS ON SEXUALITY
CYP-450 enzyme-inducing AEDs also have effects on sexuality. At the 2004 AAN Annual Meeting, Dr. Morrell and colleagues reported that women on monotherapy with an enzyme-inducing AED (phenytoin or carbamazepine) had lower levels of estrogens and androgens compared with women on monotherapy with valproate, gabapentin, or lamotrigine. Lower levels of these hormones were associated with reduced sexual desire and inadequate arousal response. The association was statistically significant. The abstract for the research, which was presented at last year's American Epilepsy Society Annual Meeting, has been submitted for publication.
“It is not surprising that reductions in these hormones might translate into sexual dysfunction,” she said. Other studies have shown that men and women with epilepsy have markedly diminished arousal compared with non-epileptic controls. Dr. Morrell's study appears to be the first to explore the mechanism for effects on sexuality.
Bone loss (osteopenia) leading to fracture is a less recognized effect of anticonvulsant therapy, Dr. Morrell continued. The possible mechanisms include the induction of CYP 450, which leads to increased catabolism of vitamin D; impaired absorption of calcium; inhibition of response to parathyroid hormone in the setting of low levels of vitamin D and calcium; calcitonin deficiency; and a direct effect on osteoclasts (bone resorption) and osteoblasts (bone formation).
The severity of bone loss appears to be correlated with the duration of anticonvulsant exposure, number of anticonvulsant drugs used, and use of CYP-450 enzyme-inducing drugs, she explained. Studies show that fracture rates are four times higher in persons on anticonvulsants than normal controls (Acta Neurol Scand 1999;42:269–275).
She recommended that patients taking AEDs have bone density tests with dual-energy X-ray absorptiometry (DEXA) of the spine or the hip to detect bone loss.
The relationship between AEDs and bone loss in women with epilepsy is not well studied. To explore this relationship, Dr. Morrell and colleagues conducted a prospective study to compare the effects of AED monotherapy in 100 women with epilepsy who were otherwise healthy (Neurology 2003;60(5):S54.004). They compared carbamazepine and phenytoin, enzyme inducers; valproate, a CYP-450 inhibitor; and lamotrigine, which has no effect on CYP 450. Patients were treated for six months or more on monotherapy with one of these agents and followed for one year.
The data showed that calcium and vitamin D levels were low for all the drugs except lamotrigine. Those who took phenytoin exhibited an accelerated rate of markers of bone resorption (osteoclasts) and bone formation (osteoblasts). The other three drugs were neutral for osteoclast formation, and had lower rates of osteoblast formation compared with phenytoin. An update on this study will be presented at this year's American Epilepsy Society Meeting, Dr. Pack noted.
According to one survey, neurologists in general are not aware of the potential for AEDs to produce bone loss. Only one third of adult and pediatric neurologists surveyed evaluated their AED-treated patients for bone disease with occasional DEXA scans (Arch Neurol 2001;58:1369–1374).
“In this survey, few neurologists said that they counseled patients about bone loss, and even fewer referred their patients taking AEDs to an endocrinologist. Treatment of bone disease is more effective if the patient is referred to an endocrinologist,” Dr. Morrell emphasized.
Less than 10 percent of the neurologists surveyed prescribed prophylactic calcium and vitamin D supplementation to patients on AEDs. Although there are no data on calcium and vitamin D supplementation in epilepsy, extrapolation from the general population suggests that “this should be part of routine clinical practice,” she said. In addition, neurologists should discuss the potential for bone loss related to AEDs with their patients and recommend gravity-resistance exercise in all patients on AEDs. Those with evidence of severe osteopenia (with a T-score less than −2.0 on DEXA) should be referred to an endocrinologist, she said.
EFFECTS ON WEIGHT
Anticonvulsant agents have differential effects on metabolism. Carbamazepine, gabapentin, and valproate cause weight gain, while topiramate causes dose-related weight loss. Levetiracetam and lamotrigine do not appear to have an effect on weight (CNS Drugs 2003;17(11):781–791). Carbamazepine and valproate increase LDL cholesterol, while topiramate decreases total cholesterol with or without decreases in LDL cholesterol.
Dr. Pack noted that several studies have suggested that valproate lowers LDL cholesterol (Epilepsia 2003;44(3):457–460). Leptin, a hormone that mediates weight homeostasis, is increased in valproate users and reduced in topiramate users (Epilepsia 2002;43(5):514–517). Insulin may be increased in patients taking valproate and slightly reduced in patients on topiramate (Obes Res 2003;11:556–562).
Weight loss observed in association with topiramate is most pronounced in patients with baseline body weight greater than 100 kg. Dr. Morrell said that this drug should not be used for weight loss because it also has cognitive side effects that “are of concern.” Cognitive side effects include difficulties with word finding, memory, and concentration, said Dr. Pack.
In summing up, Dr. Morrell said that women taking AEDs should be counseled about the effects of these drugs on oral contraceptives. Further, physicians should be aware of the possibility of for bone loss with enzyme-inducing agents; low calcium levels with carbamazepine, phenytoin, and valproate; and unfavorable changes in lipids and carbohydrate metabolism. Weight should be monitored and lipid profiles obtained at baseline and at three-month intervals. DEXA scanning should be done at baseline if the woman is at risk for osteoporosis and should be considered one year after initiating anticonvulsant therapy.
Polycystic ovarian syndrome (PCOS) can also be a reproductive-endocrine consequence of treatment with AEDs, in particular, with valproate. Studies have shown that women with epilepsy have increased rates of infertility, which seem to be related to ovulatory failure and metabolic changes (weight gain and insulin resistance) that are features of PCOS, explained Dr. Morrell.
A cardinal marker of emergent PCOS is alteration of the normal menstrual cycle, said Hadine Joffe, MD, who has studied the effects of AEDs in women with bipolar disorder. Dr. Joffe, of the Massachusetts General Hospital at Harvard Medical School in Boston, also spoke at the APA Annual Meeting.
PCOS occurs in 4 to 7 percent of premenopausal women and is characterized by a cluster of symptoms, including infrequent ovulation and hyperandrogenism. It is also one of the most common causes of infrequent menses.
PCOS can be diagnosed clinically without blood tests or ultrasound, Dr. Joffe said. In fact, 25 percent of females with the classic sign of this syndrome on ultrasound (“string of pearl” multiple follicles on the ovary) do not have PCOS. Symptoms include evidence of hirsutism, acne, and male-pattern baldness.
“Most women who have hirsutism take care of it cosmetically, so your patient may not appear with facial hair,” she noted.
In addition to the cosmetic problems, polycystic ovarian syndrome poses health risks, including diabetes mellitus and infertility, and it may increase the risk of cardiovascular disease and endometrial carcinoma.
“One of the great debates is whether valproate increases the risk of polycystic ovarian syndrome,” Dr. Joffe said. Evidence is contradictory, but some studies suggest that the risk is increased in patients with epilepsy taking valproate. These studies also show that the risk is reversible when valproate is stopped.
PCOS AND BIPOLAR DISORDER
Last year, Dr. Joffe and colleagues published a study on emergent polycystic ovarian syndrome in women with bipolar disorder taking AEDs (Harvard Rev Psychiatry 2003;11(2):99–108). The study evaluated 229 women; 86 were taking valproate and 143 were not, but were on lithium, lamotrigine, topiramate, gabapentin, carbamazepine, or oxcarbazepine. Treatment-emergent polycystic ovarian syndrome was found in 10.5 percent of valproate users versus 1.4 percent of non-users (p < .0001). In cases where valproate was deemed contributory, polycystic ovarian syndrome developed within one year of treatment initiation.
Dr. Joffe recommended that women on valproate should be educated about the risks of polycystic ovarian syndrome and should be monitored for menstrual cycle irregularity, hirsutism, and weight gain. Patients already on the drug for less than 12 months should be closely monitored for signs of the syndrome; the risk is reduced if patients have been on the drug for longer than 12 months.
Dr. Morrell stressed that physicians who use valproate to treat women with epilepsy or bipolar disorder should be alert to changes in menstrual cycle and weight gain. “These changes are not benign. They are undesirable physiological disturbances,” she said.
PREGNANCY AND EPILEPSY
Pregnancy in an epileptic woman can be a high risk if epilepsy is poorly controlled and seizures recur and also because of the side effects of AEDs. Perhaps the single most important question is whether a woman with epilepsy can have a successful pregnancy and give birth to healthy children. “While most women with epilepsy can and will have healthy children, they do face more risks,” explained Mark. S. Yerby, MD, Associate Professor of Neurology, Public Health, and Obstetrics/Gynecology at Oregon Health Sciences University in Portland. Dr. Yerby spoke on this topic at the 2004 AAN Annual Meeting in San Francisco, CA.
Pregnant women with epilepsy cannot safely stop taking AEDs, he said, because this would place them at increased risk of seizures. Seizures are associated with premature labor, miscarriage, suppression of fetal heart rate, epilepsy in the child, and developmental delay.
But, he noted, AEDs carry increased risk of congenital malformations and anomalies as well as neonatal hemorrhage. Further, safety data are incomplete on AEDs introduced prior to 1993 and there are scant data on newer AEDs, he said.
During pregnancy, the concentration of AED declines, potentially reducing efficacy of the agent and increasing the risk of seizures. In the first trimester, use of AEDs is associated with risk of malformations. Further, AEDs are competitive inhibitors of vitamin K, which can cause bleeding. In women who choose to breastfeed, the baby can be exposed to the AED via breast milk.
Adverse pregnancy outcomes reported in women with epilepsy include decreased viability of the fetus leading to increased miscarriage and infant mortality. Babies born to women with epilepsy have lower birth weight and less robust growth. Neonatal complications can include hemorrhage, and risk of malformations and development delay is increased. Polytherapy and high plasma concentrations of AEDs increase the risk of malformation in babies born to women with epilepsy. Congenital malformations have been reported with all AEDs and are 4 to 6 percent more common than in the general population. Most common defects include oro-facial clefts, midline heart defects, and skeletal defects.
The prevalence of developmental delay in offspring of women with epilepsy is 1.2 percent to 6.2 percent. Risk factors for developmental delay include low maternal IQ, polytherapy, and poor control of maternal seizures, he explained.
Breastfeeding is considered generally safe for women with epilepsy, but some experts recommend stopping by three months to reduce the transmission of AEDs to the baby. Babies have been exposed to AEDs during nine months of pregnancy, and “by the time they are born they can metabolize the mother's medication present in breast milk,” Dr. Yerby said. Studies are needed to determine the safety of newer AEDs in pregnancy and breastfeeding.
Dr. Yerby pointed out that no drug is without risks. Strategies for reducing risks associated with AEDs during pregnancy include adjusting the dose of medication, switching to a new medication and monitoring its effectiveness, or withdrawing medication. The optimal situation is to get expectant mothers on monotherapy that is effective; this will minimize side effects and risks.
“The best outcomes occur with monotherapy and good seizure control in the mother,” he stated. “Choose the best AED to control the seizures,” he emphasized.
Since 1993, eight new AEDs have been introduced. Although these agents are potentially safer than older drugs, there are few empirical data on the safety of these drugs during pregnancy, but reports of malformations with monotherapy are rare, Dr. Yerby noted.
The risks of in utero exposure to four commonly used AEDs are currently under study. Preliminary data from one study reported at this year's AAN Annual Meeting suggest that monotherapy with carbamazepine, lamotrigine, phenytoin, and valproate have differential effects on the unborn child. Valproate is associated with the highest number of adverse events and lamotrigine is associated with the lowest.
Lead investigator Kimford J. Meador, MD, Professor and Chair of the Department of Neurology at Georgetown University Medical School in Washington, DC, said that the study involved 28 centers in the UK and US. Dr. Meador presented results on 186 mother-child pairs comprising the US cohort. These mothers were enrolled during pregnancy.
Sixty women took carbamazepine, 60 women took lamotrigine, 41 women took phenytoin, and 25 women took valproate. Fifty-three serious adverse events were reported in the offspring: 15 in the carbamazepine group, 7 in the lamotrigine group, 15 in the phenytoin group, and 16 in the valproate group. The most serious adverse events were congenital malformations and disabilities. And there were deaths in the offspring of women: 10 percent from mothers taking carbamazepine, 2 percent with lamotrigine, 7 percent with phenytoin, and 28 percent with valproate.
In a telephone interview, Dr. Meador said that he presented data on the UK cohort of the study, bringing the number of mother-child pairs up to a total of 322. He said that results were similar, with congenital malformations, developmental delay, or death occurring in 25 percent of children exposed to valproate, 10 percent for carbamazepine and phenytoin each, and 2.5 percent for lamotrigine.
“My study by itself doesn't prove anything. However, at least four other studies have sent the same signal [about valproate]. These studies raise concerns, but there are no final answers. Valproate is a commonly used and very effective drug for primary generalized epilepsy,” he said.
As the study progresses, more data will be reported on monotherapy exposure, including six-year follow-up data on neurobehavioral profiles of exposed children.
ARTICLE IN BRIEF
✓ This Special Report highlights the effects of antiepileptic drugs on fertility, gynecological health, contraceptive efficacy, pregnancy outcome, and bone health for women, and offers guidance on treatment.