Cross-Module Connection

This module builds directly on the pharmacotherapy foundations established in Modules 3 and 4 (mood stabilizers, atypical antipsychotics, and the bipolar depression treatment algorithm) and the diagnostic framework from Module 2 (differential diagnosis and screening). The pediatric bipolar section connects to the ADHD series (Module 6A: Bipolar as ADHD comorbidity) and the Depression series (Module 6: DMDD). The suicide content cross-references lithium's anti-suicidal properties from Module 3 and the pregnancy section extends the valproate reproductive safety data introduced there. The disparities content connects to the future Schizophrenia series (the diagnostic bias problem linking both diagnoses).

| **In plain language for families: **

Bipolar medications for young people require regular blood tests, not because the medications are dangerous, but because a growing body processes medication differently than an adult body does. A dose that works well at age 12 might become too high or too low by age 14, simply because of how quickly your child is growing. The blood tests let us stay ahead of those changes. Weight and blood sugar checks are especially important for the antipsychotic medications, because these drugs can affect metabolism in ways that matter more the earlier they start.

| **In plain language for families: **

When someone has bipolar disorder and also uses alcohol, marijuana, or other substances, the two problems feed each other in a vicious cycle. The substance use makes the mood swings worse, and the mood swings make the person more likely to use substances. The most important thing families should understand is that these problems need to be treated together, at the same time, by the same team. The old approach of "get sober first, then treat the mood disorder" does not work and actually makes both conditions worse. The goal of mood stabilization is not just to feel better emotionally; when the mood swings are controlled, the drive to self-medicate often decreases on its own.

| **In plain language for families: **

Anxiety and bipolar disorder commonly go together, and when they do, the anxiety makes every part of bipolar disorder harder to manage. The treatment approach follows a clear priority: stabilize the mood first, because trying to treat anxiety without a stable mood foundation tends not to work well and can actually make things worse. Therapy, specifically cognitive-behavioral therapy, is one of the safest and most effective tools available because it works on anxiety without any risk of triggering a mood episode. If anxiety medications are needed, the choice of which one and when to introduce it requires careful consideration by a provider who understands the bipolar context.

| **In plain language for families: **

ADHD and bipolar disorder can genuinely occur together, and when they do, both need treatment. But the order matters enormously. Think of it like building a house: you have to pour the foundation (mood stabilization) before you can frame the walls (ADHD treatment). If you try to treat the ADHD first with stimulant medications before the mood is stable, the stimulant can push the mood in a dangerous direction. Once the mood is solidly stabilized, stimulants can usually be added safely with close monitoring. The key warning sign to watch for is a sudden change in sleep: if your child or family member starts sleeping much less but seems full of energy rather than tired, that is a signal to contact the prescriber immediately, as it may indicate the medication is pushing toward a manic episode.

| **In plain language for families: **

The days and weeks right after someone leaves the hospital are actually the most dangerous period, not the safest. The hospital provided constant structure and monitoring. Coming home means returning to the stresses that may have contributed to the crisis, often while adjusting to new medications. The most important thing a family can do during this transition is ensure the very first outpatient appointment happens within a few days of discharge, not weeks. If the person was started on lithium, making sure they continue taking it exactly as prescribed is critical, because stopping lithium suddenly can trigger a dangerous rebound in suicidal risk. During this transition period, removing firearms from the home, locking up any stockpiled or non-prescribed medications and potentially dangerous household items (while keeping the person's own prescribed psychiatric medications accessible for daily use, with a trusted family member monitoring adherence), and maintaining close daily contact are concrete, evidence-based actions that save lives.

Section 6: Bipolar Disorder in Pregnancy and the Postpartum Period

Bipolar women face the highest-stakes medication decisions in reproductive psychiatry. The treatment dilemma is genuine: every available option carries risk. Continuing medication exposes the developing fetus to pharmacological effects. Discontinuing medication exposes the mother to relapse rates of 70% to 80%, with cascading consequences for both her and the pregnancy. The clinical task is not to eliminate risk, which is impossible, but to make individualized decisions based on the most accurate data available, much of which has been meaningfully updated in recent years.

6.1 The Risk of Untreated Bipolar Disorder During Pregnancy

The assumption that pregnancy protects against psychiatric illness has been systematically invalidated. The foundational data comes from a landmark prospective study by Viguera and colleagues (2007, American Journal of Psychiatry), which tracked 89 pregnant women with DSM-IV bipolar disorder who were clinically euthymic at conception.

The findings demonstrated a substantial disparity based on treatment decisions:

• Overall relapse risk: 71% across the entire cohort experienced at least one major recurrence during pregnancy.

• Discontinuation vs. continuation: Women who discontinued mood stabilizers experienced relapse rates of 70% to 80%; those who maintained pharmacotherapy had a significantly reduced rate of approximately 30% to 35%.

• Speed of relapse after discontinuation: The median time to first recurrence was more than fourfold shorter in the discontinuation group. Abrupt discontinuation (tapering over 1-14 days) carried a recurrence latency 11 times shorter than gradual taper (15-30 days). Women who stopped abruptly faced a 50% risk of recurrence within just 2.0 weeks (95% CI 1.0-6.0 weeks).

• Polarity of relapse: 74% of peripartum recurrences were depressive or dysphoric-mixed (a state combining depressed mood with agitation, irritability, and restless energy), and nearly half (47%) occurred during the first trimester.

Additional data on lithium specifically showed that recurrence rates during the 40 weeks after discontinuation were similar for pregnant (52%) and non-pregnant women (58%), demonstrating that pregnancy itself does not alter the baseline risk of stopping medication. However, postpartum recurrences were 2.9 times more frequent in pregnant women who had been off lithium compared to non-pregnant women (70% vs. 24%), highlighting the additional risk introduced by childbirth * (Viguera et al., 2000, American Journal of Psychiatry)*.

The consequences of untreated illness are not abstract. Depressive and mixed episodes result in poor prenatal care adherence, inadequate nutrition, increased substance use, and elevated circulating maternal cortisol that can adversely affect fetal neurodevelopment. Manic episodes introduce immediate risks from impulsivity, impaired judgment, and severe sleep deprivation. Experiencing a mood episode during pregnancy is the single strongest predictor of a severe postpartum psychotic or manic episode * (Bergink et al., 2012, American Journal of Psychiatry)*.

Modern clinical consensus, including the 2023 ACOG Clinical Practice Guidelines, explicitly recommends against withholding or discontinuing medications for mental health conditions due to pregnancy status alone.

6.2 Medication Safety: Agent-by-Agent

Lithium: recalibrated teratogenic risk

Lithium's use in pregnancy has been limited by decades of fear about Ebstein's anomaly (a congenital heart defect involving the tricuspid valve). The original risk estimates, derived from voluntary submissions to the International Register of Lithium Babies in the 1970s, suggested the risk was up to 400 times higher than in the general population. Those estimates were methodologically flawed: plagued by retrospective enrollment, ascertainment bias, and lack of control groups.

The contemporary data tells a different story. Patorno and colleagues (2017, New England Journal of Medicine) conducted a massive retrospective cohort study analyzing 1,325,563 pregnancies from the U.S. Medicaid database. The key findings:

• Overall cardiac malformations: 2.41 per 100 in lithium-exposed infants vs. 1.15 per 100 unexposed. Adjusted risk ratio: 1.65 (95% CI 1.02-2.68).

• Right ventricular outflow tract defects (the category encompassing Ebstein's anomaly): 0.60% in lithium-exposed vs. 0.18% unexposed. Adjusted RR: 2.66 (95% CI 1.00-7.06).

• Dose-dependency: Doses of 600 mg/day or less carried an RR of 1.11 (not statistically significant). Doses of 601-900 mg/day: RR 1.60. Doses above 900 mg/day: RR 3.22 (95% CI 1.47-7.02).

The absolute risk of Ebstein's anomaly is approximately 0.6% to 0.7%, roughly one additional case per 100 live births. This is vastly lower than the historical 3% to 4% estimates that drove decades of lithium avoidance. Major guidelines now support lithium maintenance during pregnancy, with a recommendation for targeted fetal echocardiography and a detailed Level II anatomical ultrasound between 18 and 22 weeks to screen for cardiac anomalies * (ACOG Clinical Practice Guideline No. 5, 2023)*.

The "lithium hold" debate during labor. Historical obstetric practice mandated stopping or reducing lithium by 30% to 50% at the onset of labor, driven by fears that postpartum fluid shifts would trigger maternal toxicity. Recent evidence challenges this. A pivotal 2021 retrospective cohort study by Molenaar and colleagues evaluated 233 peripartum lithium levels across 100 pregnancies and found no significant association between delivery timing and lithium level-to-dose ratio (Pearson r = -0.03, p = .63). Maternal levels remained stable across the delivery threshold (0.73 mmol/L pre-delivery, 0.70 mmol/L postpartum), and neonatal lithium levels showed no association with adverse outcomes. Discontinuing lithium at delivery exposes the mother to subtherapeutic levels at precisely the point of peak psychiatric vulnerability. Modern consensus argues against routine lithium discontinuation during labor, favoring stable dosing with twice-weekly therapeutic drug monitoring in the first two postpartum weeks to guide gentle dose reductions as renal function normalizes * (Molenaar et al., 2021, Bipolar Disorders)*.

Valproate: absolute contraindication

Valproate is the most potent teratogen (an agent that causes birth defects during fetal development) in the psychopharmacological formulary and is universally contraindicated as a first-line agent for women of childbearing potential.

• Structural malformations: First-trimester exposure carries a 6% to 9% risk of major congenital malformations, including neural tube defects, craniofacial abnormalities, and cardiac defects. Standard folic acid supplementation does not significantly reduce this risk.

• Neurodevelopmental toxicity: The NEAD study demonstrated that in utero valproate exposure is dose-dependently associated with an average IQ deficit of approximately 10 points compared to other antiepileptic monotherapies * (Meador et al., 2009, New England Journal of Medicine). Danish population data showed vastly elevated autism spectrum disorder rates * (Christensen et al., 2013, JAMA) and a subsequent registry study demonstrated a 4.5-fold increased risk of intellectual disability * (Daugaard et al., 2020, JAMA Network Open)*.

• Regulatory restrictions: 2023-2024 NICE updates mandate that valproate must not be started in anyone under 55 unless two specialists independently verify that no alternative exists. 2024 updates extended warnings to males regarding potential fertility and offspring risks.

Lamotrigine: the safest agent, but with pharmacokinetic challenges

Decades of prospective registry data indicate no significantly elevated risk for major structural malformations compared to the general population baseline. However, pregnancy dramatically alters lamotrigine pharmacokinetics.

Carbamazepine: significant teratogenicity, less commonly used

Carbamazepine carries a meaningful teratogenic risk, with major congenital malformation rates estimated at 2% to 3%, including neural tube defects and craniofacial abnormalities. While less teratogenic than valproate, these risks are clinically significant. Carbamazepine also carries the specific concern of HLA-B*1502-associated Stevens-Johnson syndrome risk in individuals of certain East Asian ancestry, requiring pharmacogenomic testing before initiation. In clinical practice, women of childbearing potential on carbamazepine should be transitioned to lamotrigine or an atypical antipsychotic before conception when feasible. If carbamazepine must be continued during pregnancy, targeted fetal screening is recommended, analogous to lithium monitoring * (ACOG Clinical Practice Guideline No. 5, 2023)*.

The estrogen-clearance problem. If you have a pregnant patient on lamotrigine, this is the pharmacokinetic concept that will matter most in your clinical management. The body eliminates lamotrigine primarily through a liver enzyme called UGT1A4, which is part of the glucuronidation system, one of the body's key drug-processing pathways. During pregnancy, rising estrogen levels powerfully activate this enzyme, accelerating lamotrigine clearance by up to 275% from non-pregnant baseline * (Karanam et al., 2024). This acceleration affects the vast majority of pregnant women (75% to 91%) and often begins as early as 5 to 10 weeks of gestation, sometimes before the pregnancy is recognized. Without dose adjustments, serum lamotrigine levels routinely drop below 65% of baseline, which is frequently insufficient for mood stabilization and triggers severe depressive or mixed relapses * (Pennell et al., 2008, Neurology; Karanam et al., 2024).

Management protocol. Establish a pre-pregnancy reference concentration while the patient is euthymic. During pregnancy, draw lamotrigine levels every 4 weeks. When plasma concentration falls below the reference, augment the daily dose by 20% to 25%. Some patients require dose increases totaling 200% to 300% of their preconception dose by the third trimester to maintain stability.

Critical postpartum danger. Immediately after delivery, estrogen crashes and UGT1A4 induction ceases. If the elevated third-trimester dose is maintained, the patient is at extreme risk of lamotrigine neurotoxicity (ataxia, severe dizziness, nystagmus, potentially life-threatening rash) within days. Best practice: reduce the dose by 20% to 25% within 1 to 3 days postpartum, with continued step-downs over 2 to 3 weeks guided by weekly blood draws until the pre-pregnancy dose is re-established * (Clark et al., 2013, American Journal of Psychiatry)*.

Atypical antipsychotics: metabolic risk, not structural teratogenicity

Large-scale data indicate that atypical antipsychotics are not major anatomical teratogens. 2024 neurodevelopmental assessments of preschool-aged children exposed in utero found no significant deficits compared to unexposed controls from similar psychiatric backgrounds * (Freeman et al., 2024)*. However, specific agents carry substantial maternal metabolic risk.

• Olanzapine: Highest metabolic hazard; adjusted RR of 1.61 (95% CI 1.13-2.29) for gestational diabetes, with absolute rates approaching 12% in some cohorts * (Park et al., 2018, American Journal of Psychiatry)*.

• Quetiapine: Moderate risk; adjusted RR of 1.28 (95% CI 1.01-1.62) for gestational diabetes.

• Aripiprazole, ziprasidone, risperidone: No statistically significant increase in gestational diabetes risk compared to women who discontinued.

The clinical implication: if using olanzapine or quetiapine during pregnancy, enhanced metabolic monitoring, dietary counseling, and close collaboration with obstetrics are essential.

| **In plain language for families: **

If you have bipolar disorder and are pregnant or planning pregnancy, the most important thing to know is that stopping your medication is usually more dangerous than continuing it. The old advice to "stop everything while pregnant" has been replaced by evidence showing that untreated bipolar disorder during pregnancy causes serious harm to both mother and baby. Most medications can be continued with appropriate monitoring and dose adjustments. The one clear exception is valproate (Depakote), which carries unacceptably high risks for the baby's brain development and should be switched to a safer medication before conception whenever possible. After delivery is the highest-risk time; having a plan in place before the baby arrives, including medication that starts the same day as delivery and a strategy for protecting your sleep, can prevent the severe psychiatric emergencies that otherwise affect roughly 1 in 4 to 5 bipolar women.

| **In plain language for families: **

The newer bipolar depression medications (cariprazine and lumateperone) work differently from older options and tend to cause less weight gain and fewer blood sugar problems. Cariprazine works best at a low dose for depression, which is the opposite of what you might expect. Its main side effect is a feeling of restlessness that can be mistaken for anxiety. Lumateperone is one of the best-tolerated options currently available and is notable because it is approved for both Bipolar I and Bipolar II depression. Your prescriber can help determine which medication makes the most sense based on your family member's specific situation, including whether other options have already been tried and how much weight and metabolic effects are a concern.

| **In plain language for families: **

The diagnostic bias section of this module addresses something that affects real patients: research shows that Black patients are systematically more likely to be misdiagnosed with schizophrenia when they actually have bipolar disorder. This matters because the two conditions require very different medications, and getting the wrong diagnosis means getting the wrong treatment, sometimes for years. If you or a family member is Black and has received a schizophrenia diagnosis, but the psychotic symptoms seem to come and go with mood changes rather than being constant, it may be worth asking the treatment team specifically about whether bipolar disorder with psychotic features should be considered.

PsychHQ

Bipolar in Pregnancy: Risk & Medication Safety