Key moments, in order
A mood stabilizer treats or prevents bipolar episodes without flipping someone into the opposite pole.
Make a greater effort to add mood stabilizers to long-standing antipsychotic regimens, even dual ones. When indicated, some patients improve in ways the antipsychotic class cannot achieve.
Bipolar specialists stress mood stabilizers, and the more patients you treat, the less you use antidepressants and the more mood stabilizers become the mainstay.
The core list: lithium, Depakote, lamotrigine, carbamazepine and oxcarbazepine.
It is not a mood stabilizer. The evidence is not there for primary mood stabilization, and on its own it just doesn't really work. Think of it as an adjunct: relatively safe, apart from a small risk of withdrawal and tolerance.
What turns a novice into an experienced bipolar clinician is focusing on maintenance. Treatment has three parts, maintenance, acute mania and acute depression, and maintenance is understated in a lot of practice.
Serious bipolar disorder rarely needs only one or two medications, so do not be afraid of polypharmacy in the service of getting someone better. But do not jump to it, and make sure the diagnosis is right first, or you give permission to over-medicate personality disorders.
Lithium scares people, and its prescription rate has gone down. It is scary the first time: all those labs and side effects to discuss.
Bipolar I and II in the DSM were created to give a reliable sample and definition for research, not to mark the only forms of the illness.
Do not be afraid of them as adjuncts beyond bipolar disorder. There is probably a lot of crossover between bipolar disorders, schizophrenia and sometimes OCD, ADHD and autism.
We are personally against using lamotrigine or other mood stabilizers for borderline personality disorder. Some people believe in it, but the signal from that small old UK study is not convincing.
Before treating something as bipolar, look for some evidence of episodes or stark changes in total functioning, not just mood reactivity in relationships.
Update your diagnoses and pay attention over time. Do not hang your hat on first impressions.
Describe what you are seeing, give options within certain parameters, and stay open to updating your formulation. If a patient does not want to accept lithium's risks, do not push it; move on to the next choice.
Or start around 450 mg nightly of the ER. Starting at 900 mg, and many of the feared side effects, belong to the older inpatient era; that is too high for outpatients.
The ER does not have to be given twice a day; you can dose it all at once. But level studies used immediate release two or three times a day, drawn 12 hours after a dose and before the next, so time your draws accordingly.
There is a narrow therapeutic window only when treating acute mania. In lower acuity conditions and long term, lower doses can bring clinical improvement, even if it is not perfect.
In a low-resource clinic where many patients refuse or cannot get labs, we will titrate to around 750 or even 900 mg, guided by reported adverse effects, before the first level.
If you know a patient will never follow up for labs, lithium is probably not the medication. Still, we will give up to 450 or maybe 600 mg long term without labs to a competent patient who understands the risks.
It is a salt. Too much salt lowers the level and too little raises it; dehydration raises it and overhydration washes it out; NSAIDs hold lithium in and can make it toxic.
Quite a few blood pressure medications interfere with lithium. Tell patients to let whoever prescribes their blood pressure medication know they are on lithium.
Keep hydration advice simple: keep it stable, do not get too dehydrated, and do not change your water intake on your own; bring up any changes at the next visit.
Walk through a one-page lithium sheet with the patient on a shared screen, especially the first few times you prescribe it.
Say it plainly: we don't do labs because this medication is unsafe; the labs are what make it extra safe.
The labs look at kidney function, thyroid function and calcium. Within the first month, check a lithium level, thyroid panel, CBC and comprehensive metabolic panel.
A slightly higher white count from demargination is fine and non-pathological.
At non-manic doses, the 20-year risk of renal dysfunction is much tinier than people realize, and renal damage is extremely unlikely as long as you watch for NSAIDs and antihypertensives and check levels.
A decline in kidney function is not renal failure. Given the choice of 30% worse kidney function or 30% less brain function over 20 years, take the better brain every time.
The two major adverse effects in practice are thirst with increased urination (more common, and different from the rarer diabetes insipidus) and acne. Increased urine can alarm clinicians who are new to lithium.
Dosed at the minimum effective level, it does not cause much weight gain. Acne, if it happens, has no great fix.
Spironolactone is an option for acne in a very compliant patient, but it raises lithium levels: check labs before and after starting it.
The combination is not completely contraindicated if the patient fully understands the risks, is well monitored, has good family support and you keep a close eye on them.
Amiloride has been written about in small case studies as reducing lithium's kidney damage and polyuria.
To tell true diabetes insipidus from simple lithium polyuria, you are supposed to do a 24-hour urine collection.
Ordering labs is not a big deal: make a lab requisition form and give it to the patient.
Treat the patient, not the level. Go by long-term clinical stability, and look carefully at sleep patterns.
For adjunctive lithium or mild bipolar disorder, levels of 0.4 to 0.7 are fine. The traditional target of about 0.75 comes from literature on severe, hospitalized bipolar I.
Maintenance is considered 0.6 to 0.8; 0.8 to 1.2 is for a patient who is actively manic. Once the mania breaks, bring the dose down: nobody should be stuck at 1.1 as maintenance without a clinical reason.
Lithium usually takes at least two weeks to start working. Many clinicians and patients want instant gratification, so set expectations.
Present it as the long game. Mood stabilizers decrease the number of episodes; they help depression and mania, but slowly.
Every untreated episode is burning brain, and if mood episodes are not treated early enough, cognitive functioning seems to decline into a kind of acquired ADHD.
High lithium levels guarantee a decline in kidney function; high valproate levels do not guarantee any end-organ damage. You still monitor, mainly for low platelets and hyperammonemia.
It is better for significant irritability and aggression, for people with some TBI history, and for people with a legitimate migraine condition, where it prevents headaches.
Consider avoiding lithium in people with genuine migraines (not just headaches); it seems to worsen them in a lot of people.
People with TBI-related neurocognitive or mood changes may benefit from lower valproate levels even without bipolarity.
The idealistic lithium patient has euphoric mania with intermittent episodes of depression. Depakote has more evidence in rapid cycling and mixed episodes.
Its anti-manic efficacy is not fantastic, but its antidepressant effect in bipolar depression is relatively robust. Studies show it is not effective for acute depression, yet our clinical experience contradicts that: there is a really big mismatch between experience with it and the trial data that guidelines rely on.
It is for prevention, not acute states. It helps prevent episodes of mania, but if a patient is already manic or hypomanic it will not work.
It is the most well-tolerated mood stabilizer by a long shot. Because SJS is rare, we phrase that risk to patients as an allergic reaction.
Absolutely follow the titration rules. The schedule is slower when you dose it with an inhibitor of its metabolism, such as Depakote. Follow the protocol and warn, warn, warn.
Tell patients that any new rash while starting it, one they have not had before and that is different, means stopping the medication and going to the ER. A same-day dermatologist works too and will know better than an ER doctor whether it is SJS.
Never tell a patient the rash is not a big deal. Give the proper response even if the system behind you is not built to respond to it appropriately.
Some ethnic groups have a higher risk of SJS with lamotrigine, oxcarbazepine and carbamazepine, and testing is available for the gene that puts them at particularly high risk. For Asian patients starting carbamazepine, the FDA recommends testing for HLA-B*15.
For an Asian patient, the FDA recommends testing for HLA-B 15 before carbamazepine.
Do not linger on lower doses. If you are going to use it, get it to 200 mg: it is so well tolerated and you get benefit. Do not be afraid to go above 200 either; you can go to 400.
Pregnancy reduces serum lamotrigine levels by about half, probably through estrogen.
Valproic acid jacks up lamotrigine levels, and rising lamotrigine is what we fear because of SJS. Whenever you combine them, halve the lamotrigine. Better still, avoid the combination, which is a huge pain, and review the protocol carefully if you must.
Because it has so few side effects, clinicians can keep patients on it and never offer what the correct treatment is.
Many people can take a lower daily dose of lithium alongside 200 of lamotrigine, which is much more tolerable than lithium alone.
If bipolar disorder does not respond to one mood stabilizer, the answer is two mood stabilizers, not an antipsychotic.
Too many drug interactions: it interacts with a huge number of other medications and with itself, so it takes forever to get to level. There are very few patients for whom you cannot justify a different medication.
You have to actually try it. Treat enough bipolar disorder and you will see people who do not respond to one mood stabilizer and do respond to another.
Probably less efficacious than Depakote, lamotrigine and lithium. At higher doses it may actually beat lamotrigine, and in the right patient it is not that bad.
For tolerability, the only real downside is that you have to give it at least twice a day.
Other anti-seizure drugs may work for bipolar disorder, but we will not know because no one has motivation to test them.
Levetiracetam is well known for causing irritability and worsening anger, yet there is some evidence it may have an anti-manic effect in some patients.
Not much is known about how they work, but anything that lowers the excitability of the brain intuitively should have some effect on bipolar disorder.
Assume the same SJS risk as lamotrigine. You cannot tell whether it is higher, because lamotrigine is used so much more that it collects most of the SJS cases.
Warn about SJS. You do not need oxcarbazepine levels; some people get them, but they are not very helpful clinically.
As an adjunct, it is a nice sleep and anxiety support while you truly treat with a mood stabilizer. In bipolar disorder we are much more likely to reach for it for sleep, though it is only ever adjunctive, for things like anxiety and insomnia.
We reserve benzodiazepines for refractory sleep-wake cycle disorders, catatonia and bipolar disorder. They are basically a mood stabilizer, reducing the brain's excitability, but they are a full shutdown medication. When you see catatonia, you should probably be using one.
Each moment is a point made in the episode, linked to the second it was said. The topic links open everything the podcast has said on that topic.