1. Discuss the advantages and disadvantages of monotherapy and polypharmacy in bipolar disorder.

2. Interpret the data and conclusions of clinical studies in bipolar disorder.

3. Compare the relative risks of illness and treatment in pregnant patients with bipolar disorder.

4. Counsel parturients about the risks psychoactive medications present to their offspring during pregnancy and breast- feeding.

5. Review the relative reproductive safety information for medication treatment options during pregnancy and breast- feeding.

New Treatment Options for Bipolar Disorder
Terence A. Ketter, MD, Professor of Psychiatry and Behavioral Sciences, and Chief, Bipolar Disorders Clinic, Stanford University School of Medicine, Palo Alto, CA

Monotherapy: in registration studies, in both mania and depression, antipsychotics provide 20% benefit (on average) over placebo; average number of medications in patients with bipolar disorder is 5

Divalproex: monotherapy—studies found that “loading divalproex could be effective,” and loading strategy used in registration studies for divalproex extended release (ER); starting dosage 25 mg/kg per day, rounded up to nearest 500, then increased by another 500 mg on day 3; combination therapy—in American approach, foundational therapy is mood stabilizer (lithium or divalproex), with atypical antipsychotic added; in European approach, foundational therapy is antipsychotic, with mood stabilizer added; in European trial, valproate added to antipsychotic achieved extra benefit of 20% to 25% over antipsychotic monotherapy

Carbamazepine: monotherapy—benefit ≈20% over placebo; starting dosage 400 mg/day in divided doses produced “fair bit of toxicity”; speaker suggests that starting with 200 mg/day and increasing by 200 mg every other day decreases toxicity; carbamazepine lowers blood levels of “almost everything” except maybe ziprasidone; combination therapy—in trial of acute mania where olanzapine or placebo added to carbamazepine, no additional benefit seen; in study where risperidone added to carbamazepine, 40% decrease in risperidone blood levels, and patients on carbamazepine had to be censored to see add-on risperidone separate from placebo and other mood stabilizers; carbamazepine should have no pharmacokinetic interaction with lithium, excreted renally, “so maybe it’s (lithium) got a chance”; even olanzapine, “which seems to be able to make everything else work better,” failed to show additional benefit when added to carbamazepine

Bipolar disorder: patients symptomatic about half the time, and depression more common than mood elevation; Systematic Treatment Enhancement Program (STEP) showed in depressed patients that adding antidepressant to mood stabilizer does not add much benefit over mood stabilizer alone, and mood stabilizer alone achieved only ≈25% response; speaker suggests starting with mood stabilizer as monotherapy in depressed patients, perhaps adding antidepressant later; if antidepressant does not work, try atypical antipsychotic

Pramipexole (Miramex): 2 trials show it may work as add-on, but only 22 patients total, and “there were some mood switches”; speaker starts with 0.125 mg at bedtime and increases every 4 days to 1.75 mg; “look out for nausea”

Other add-ons: 1 trial showed 22% response rate to modafinil (Provigil), with 4.9% switch rate; gabapentin shown not to be effective as primary treatment for mania or for treatment-resistant rapid-cycling bipolar disorder, but may be useful in comorbid conditions such as anxiety or pain

Lamotrigine: in trials, lamotrigine worked well in terms of response rate, but placebo worked well also; not indicated for acute bipolar depression or acute mania; may be effective in preventing recurrent depression; combination of lamotrigine and lithium may prevent depressive and manic episodes; in study, lamotrigine effective in treatment-resistant rapid cycling

Topiramate: ineffective in acute mania, but patients lost weight; may be effective in treating comorbid disorders such as eating disorders, alcohol dependence, and migraine; can cause metabolic acidosis

Zonisamide: produced weight loss in obese people with no psychiatric disorder and in obese people with euthymic medicated bipolar disorder; long half life; cannot be used by patients with sulfonamide allergy; weight loss and tolerability adequate in one third of patients, intolerance in one-third, and it “just doesn’t work” one-third; metabolic acidosis can occur

Oxcarbazepine: “data not too encouraging”; consider oxcarbazepine if patient has failed carbamazepine or is too unreliable to comply with blood testing associated with carbamazepine

Atypical antipsychotics: all seem to work in mania; clozapine is only atypical antipsychotic not approved by Food and Drug Administration (FDA) for use in bipolar mania; clozapine helpful if patient unable to sleep, but requires blood testing because of risk for agranulocytosis

Risperidone: response ≈20% compared to placebo; if added to lithium or divalproex, additional benefit seen

Olanzapine: response rate 20% to 25%; doses in trials 10 mg/day to 20 mg/day; in head-to-head trial with divalproex, olanzapine slightly better in efficacy, but divalproex slightly better in tolerability; in head-to-head trial with lithium, olanzapine slightly better in relapsing mania, but associated with more weight gain than lithium

Olanzapine plus fluoxetine combination: response “so-so” with 7.5 mg of olanzapine and 40 mg of fluoxetine per day; “if you absolutely have to use an antidepressant in bipolar depression and you want to do evidence-based medicine, this is what you would do”; in head-to-head trial with lamotrigine, combination had 9% response rate vs 7% weight gain

Quetiapine: response rate ≈17%; must be titrated to avoid hypotension

Other atypical antipsychotics: ziprasidone and aripiprazole, “we know [they] work in acute mania”; aripiprazole has maintenance indication; asenapine—separated from placebo on primary-outcome measure but not on response rate

Emerging uses for atypical antipsychotics in bipolar disorder: primary therapies—olanzapine for mania, maintenance, and, when combined with fluoxetine, depression; risperidone, ziprasidone, and asenapine for mania; quetiapine for mania and depression; aripiprazole for mania and maintenance; adjunctive therapies—olanzapine and risperidone for mania; quetiapine for mania and maintenance; clozapine for treatment resistance

Conclusions: many new agents with diverse mechanistic efficacy and side-effect profiles in development; new anticonvulsants as class not effective in acute mania and have variable efficacy in bipolar disorders and comorbid conditions; newer antipsychotics as class effective in acute mania and show emerging efficacy in acute depression and maintenance

Depression and Bipolar Illness in Pregnancy
Zachary N. Stowe, MD, Director, Women’s Mental Health Program, and Professor, Department of Psychiatry and Behavioral Sciences, Emory University School of Medicine, Atlanta, GA

Introduction: most mental illnesses have their onset long before family planning begins

Key point: between 9 and 49 yr of age, consider every woman pregnant until proven otherwise

Treatment considerations: stopping or changing medications abruptly when patient reports her pregnancy risks exacerbation of her mental illness; changing medications exposes fetus to 2 medications instead of one; in study, if antidepressants discontinued, 62% of parturients became ill before delivery (however, 25% became ill if medications continued); in bipolar disorder, if mood stabilizer stopped, 83% had episode before delivery, majority of which were depression; if medications tapered, patients do much better, and fetal exposure not increased much by tapering over 7 to 10 days

Depression: depression during pregnancy “huge” risk factor for postpartum depression; risk for woman’s entering psychiatric hospital greatest during postpartum period (1 in 12 of all psychiatric admissions for women occur before child’s first birthday); bipolar disorder—risk for relapse present whether or not woman on treatment; several small retrospective studies suggest lithium might be better choice for prophylaxis in postpartum period, but only 1 prospective study to date, and in it, valproic acid failed to prevent postpartum illness; studies do not distinguish between unipolar and bipolar depression, but show that depressed women do not comply with prenatal care (eg, they smoke more and use more alcohol) and are at higher risk for preterm labor; in addition, their babies at 2 times greater risk for admission to neonatal intensive care unit (NICU); 18-yr prospective study showed that maternal depression in pregnancy and postpartum period directly predicted conduct disorder and violent behavior in offspring at 7 to 12 yr of age; maternal mental illness causes significant adverse impact on biology or behavior of children in adulthood

Course of pregnancy: shown that adverse outcomes can occur when fetus exposed in any trimester; alcohol probably has more effect when consumed in second or third trimester; anticonvulsants do not have teratogenic risk in second and third trimesters, but still carry same neurodegenerative risk

Use of psychotropic medications in pregnancy and lactation: from first visit, treat all women as if they were pregnant; “new and improved” means medication has no data in pregnancy; psychiatrist is prescribing clinician of record, so prudent to ask about and document method of contraception at every visit; pregnancy cannot be screened effectively and accurately in outpatient setting (eg, only 28% of women can accurately remember when they had their last menstrual period; negative pregnancy test simply means pregnancy is <3 wk gestation)

Events during pregnancy: study showed that by 6 mo postpartum, about half of women in speaker’s practice did not remember having been depressed during pregnancy; many also could not accurately remember use of medications during pregnancy (eg, 13% who said they took their medication during pregnancy had undetectable blood levels throughout pregnancy); many did not accurately report use of tobacco, alcohol, or drugs during pregnancy; women who reported inaccurately tended to be less educated, younger, and nonwhite

FDA pregnancy categories for drugs: category A—controlled studies show no risk; adequate, well-controlled studies in pregnant women failed to demonstrate risk to fetus; category B—no evidence of risk in humans; either animal findings show risk but human findings do not, or, if no adequate human studies have been done, animal findings negative; category C—risk cannot be ruled out; human studies lacking, and animal studies are either positive for fetal risk or lacking as well; however, potential benefits may justify potential risk; category D—positive evidence of risk; investigational or postmarketing data show risk to fetus; nevertheless, potential benefits may outweigh risks; category X—contraindicated in pregnancy; studies in animals or humans; investigational, or postmarketing reports have shown fetal risk that clearly outweighs any possible benefit to mother

Psychoactive medications: antidepressants—most were in category B but were moved to category C in 1994 when, although study showed no increased risk of spontaneous abortion and no increased risk of birth defects, “it cued the FDA that we were giving antidepressants to pregnant women; there were no data to support this [move to category C]”; benzodiazepines—benzodiazepine sedative hypnotics only category X medicines in psychiatry, “despite the fact there’s no confirmed birth defect on a sleeping pill”

Bottom line: “when manufacturers and official agencies warn against drug treatment during pregnancy, their warnings serve to protect themselves and are of little use to us [clinically responsible physicians]”

Pharmacokinetics: P-glycoprotein and maybe breast-cancer–resistance protein (BCRP) in placenta control medications that get to fetus; in speaker’s study, all antidepressants studied found in amniotic fluid, and concentration in fetal blood matched that in mother’s; shown that 8% to 15% of children exposed to antidepressant in utero have tremulousness and rapid respirations for several hours after delivery

Specific medications: lithium—follow-up studies failed to find association between lithium and Ebstein’s anomaly; labor and delivery are dehydrating, and stopping lithium at onset of labor or 48 hr before caesarean delivery prevents lithium toxicity in fetus; lithium restarted immediately after delivery with no increase in psychiatric symptoms in mother; valproate—should not be used as first-line treatment in women of reproductive years; in women who receive >1000 mg/day or whose blood level >70 µg/mL, overall rate of birth defects 9.6%; unclear whether folic acid reduces risk; lamotrigine—“cleanest anticonvulsant we’ve seen with respect to birth defects,” although recent study showed higher rate of oral clefts; olanzapine—no increased risk to fetus, but women taking olanzapine during pregnancy had elevated blood glucose levels; even in women who passed glucose tolerance test, babies tended to be large (>10.5 lb)

Lactation: concentrations of psychotropics in breast milk lower than concentrations in mother’s blood; switching medications exposes infant to second medication; postpartum period highest risk period for mother to get psychiatric illness; speaker suggests benefit from treating mother greater than risk to infant

Lactation-safety classification schemes

American Academy of Pediatrics: usually compatible with breast-feeding; unknown but of concern; associated with significant side effects and should be used with caution; requires cessation of breast-feeding

Medications and Mothers’ Milk: L1—safest; L2—safer; L3—moderately safe; L4—possibly hazardous; L5— contraindicated; although this classification scheme is most frequently used by pediatricians and pharmacists, speaker opines that it has “nothing at all to do with data”

Summary: from first visit, treat all women of reproductive years as if they were pregnant; all medications cross placenta and enter human breast milk; large database on psychotropic medications exists, but limitations preclude definitive conclusions, except for antiepileptic drugs; postnatal environment extremely important and may serve to modulate medication effects; medication use during pregnancy results in significant fetal exposure; lactational exposure needs further characterization, but central nervous system exposure may occur in absence of detectable infant plasma concentrations; strong evidence exists that untreated maternal depression during pregnancy and postpartum period has adverse impact on offspring; genetic variation may influence medication and illness exposures