Lithium is the best mood stabilizer we have for bipolar I, and we don’t use it nearly enough. That’s the whole post, more or less. The rest is detail.

It’s been around since the 1940s, it’s a simple metal salt, you can’t patent it, no drug company has any reason to market it, which means there’s no sales rep showing up at the office handing out pens and lunch… which honestly explains a lot about our industry, because newer drugs with worse track records get prescribed more often just because somebody paid for the pens. Lithium has a sales rep budget of zero dollars. It is also the only mood stabilizer with consistent data showing it reduces suicide in bipolar patients independent of its mood effects. Nothing else has that data. Nothing.

What it does

The honest answer is that we don’t fully know. There are theories involving GSK-3 inhibition, inositol depletion, BDNF (a brain growth factor) going up, and a half-dozen other mechanisms researchers like to argue about, and none of that matters for what we’re doing here. What matters: it calms mood across both directions of bipolar, holds back mania more reliably than it pulls people out of depression, cuts relapse rates significantly, and reduces suicide. The last one is the kicker. Bipolar I has a roughly fifteen percent lifetime suicide rate untreated, meaning fifteen out of a hundred bipolar patients end up taking a road they can’t turn back from. Lithium drops that substantially. That’s not just some shit I’m saying, that’s the most robust finding in the bipolar literature.

The blood draw thing, which is what guys actually hate

Lithium has a narrow therapeutic window. Too low and it doesn’t work. Too high and it’s toxic, and lithium toxicity is a real medical emergency that can kill you or leave you with permanent neurological damage, which is a fact your prescriber should not be shy about telling you up front. The maintenance window is roughly 0.6 to 0.8 mEq/L for most patients, sometimes up to 1.0 for actively manic stretches.

This means blood draws. Initially every week or two while we titrate, then every three months for the first year, then every six to twelve months once things are quieter. We’re checking the lithium level, the thyroid (because lithium suppresses it over time in a chunk of patients), and the kidneys (because lithium is cleared by the kidneys and they’re slowly affected over decades). Most long-term lithium patients will at some point need thyroid supplementation, which is a daily pill that’s easy to manage. The kidney effects are slower and usually mild but they are real, which is the kind of thing you’d want a prescriber to sit down and walk you through honestly instead of skipping over.

The blood draw piece is what guys hate about lithium and what makes them refuse it before trying it. I get it. It’s also why we end up sending plenty of bipolar patients home on Depakote (valproate, a different mood stabilizer with worse long-term math) or Lamictal (lamotrigine, also a different mood stabilizer, mostly good for the depression side of bipolar II) when they would have done better on lithium, because lithium asks you to participate in your own treatment in a way the others don’t. The labs are the price of admission, and the price is real but the drug is worth it.

What’s nice to hear about it

I’m going to lead with the upside for once instead of burying it, because the lithium conversation usually opens with all the lab stuff and then patients never quite get to hear the part where the drug works. For the right patient, it works in a way nothing else really does. Mania quiets in a few weeks. The cycling slows. Sleep evens out. The relationship and the job stop getting periodically blown up by episodes nobody saw coming. The patients who do well on lithium tend to describe it not as feeling better in a chemical-vibes way but as their life finally being something they can plan for. That part doesn’t go on a rating scale, and it’s the part the drug delivers that newer agents mostly don’t.

Dosing, in plain language

Start at 300mg twice a day, or 600mg at night. Check a level five to seven days in, which is when the drug hits steady state and the number actually means something. Adjust from there. Most adults end up between 900 and 1500mg total daily, dosed once or twice a day. Once a day at night is fine for most people and cuts some of the daytime side effects.

Hydration matters a lot. Dehydration spikes lithium levels, so does ibuprofen and naproxen (the NSAIDs, the over-the-counter pain stuff) because they drop kidney clearance, so does the kind of stomach bug where you can’t keep fluids down. Any of those, hold the dose and call. That sounds like a lot. In practice it’s once or twice a year, you drink water like you actually like it because you’re going to need it, and you learn to manage. Tylenol (acetaminophen) is fine for pain because it doesn’t go through the same kidney pathway, ibuprofen and naproxen are a problem.

Side effects, in the order they come up

Tremor first. Fine tremor of the hands, usually mild, often resolves over the first few months. Caffeine makes it worse, which is its own issue for the guys who run on coffee. Propranolol (a blood pressure pill that also kills tremor) knocks it down if it’s bothersome.

GI stuff in the first weeks… nausea, sometimes loose stools, occasionally a metallic taste in the mouth. Mostly settles down. Taking it with food helps.

Thirst and peeing more. Lithium affects how the kidney holds onto water, so you pee more and you drink more, and some guys barely notice while some find it genuinely disruptive (the get-up-twice-a-night version is the most common complaint). Worth knowing going in.

Weight gain, modest, maybe five to ten pounds over a year. Less than Depakote or Zyprexa. More than Lamictal.

The cognitive thing is the one that bothers smart, high-functioning guys the most. Some patients on lithium describe feeling slightly less sharp, slightly out of it, slightly less themselves in a way that’s hard to put a finger on. Sometimes dropping the dose fixes it. Sometimes the trade-off just isn’t worth it for that particular person and we move to Lamictal or something else. The framing I try to use here is that the goal isn’t to feel exactly the way you did before, the goal is to keep your life from blowing up every couple of years… but I take the cognitive complaint seriously when patients raise it, because it’s real and dismissing it is how people end up quietly stopping their drug.

Acne and psoriasis can flare in people prone to either. Not common, real when it happens. Thyroid suppression over time, manageable with supplementation, expected enough that we’re checking labs anyway.

Who it’s actually for

Bipolar I, the classic version with full manic episodes… that’s the textbook fit. Mixed states. Bipolar with a family history that responded to lithium, because lithium-responsive bipolar tends to run in families (a brother or dad who did well on it is one of the better predictors that this patient will too). Bipolar with suicide risk, which is most of bipolar, which is the whole point about underuse.

For bipolar II, where the highs are mostly fine and the lows are the problem, Lamictal is usually the better first move. Lithium works for bipolar II, it’s just a heavier tool than the diagnosis usually needs.

For unipolar depression that’s failed multiple antidepressants, lithium at low doses (300-600mg added on top of the antidepressant, way below bipolar dosing) is one of the best-evidence augmentation moves in psychiatry, and almost nobody uses it because of the lab thing. Real shame, the data goes back to the 80s.

The kind of patient who tends to make the switch and not look back

Picture a guy in his late thirties who’s been carrying a bipolar I diagnosis since his mid-twenties, started after one big manic episode that involved buying something he couldn’t afford and driving somewhere he shouldn’t have. Has been on Depakote for years and is tired of the weight and the thinning hair, and has a brother who’s done well on lithium for a long stretch. He comes in willing to try the switch as long as somebody walks him through the trade-offs honestly… the labs, the hydration thing, the NSAID rule.

The cross-taper takes about six weeks. He lands somewhere around 1200mg of lithium at night with a level in the 0.7 range. Six months in he’s sharper than he’s been in a while, he’s lost most of the Depakote weight, his wife is asking him what’s different. Two years later he’s still on it, has had one mild hypomanic blip during a stretch of bad sleep with a newborn (which is just bipolar telling on itself, sleep loss is the most reliable trigger for mood instability in this population), tightened up sleep and rechecked the level, no full episode. He’s also alive, which the suicide-reduction data suggests is part of what the drug is doing for him in a way nothing else really can. That last sentence is the one I have to say out loud to patients sometimes, because it lands differently than the abstract version.

Lithium has a sales rep budget of zero dollars. Newer drugs with worse track records get prescribed more because somebody is handing out pens at the office, which honestly explains a lot about our industry.

What not to do

Don’t take ibuprofen or naproxen without checking with the prescriber, the kidney clearance thing is real… Tylenol is fine because it works through a different pathway, the rest of the over-the-counter pain shelf is a problem. Don’t get dehydrated, hot Saturday hike means you drink water like you actually like it because you’re going to need it, stomach bug means hold the dose for a day and call. Don’t skip the labs, the monitoring is the price of admission and the drug is worth it. Don’t stop cold turkey, especially in the first year, because lithium-responsive bipolar that gets stopped and restarted sometimes doesn’t respond the same way again on round two… this is one of the few drugs where the choice to stop has lasting consequences, which is worth knowing before you stop.

How it stacks up against the alternatives in bipolar I

Depakote (valproate) is the most-prescribed alternative because the monitoring feels less involved and the side effects feel more familiar to general medicine. It works for acute mania, sometimes works for maintenance, doesn’t have the suicide-reduction effect, and the long-term picture (weight, hair loss, tremor, liver effects, and the fact that it can wreck a fetus, which makes it inappropriate for any woman who might get pregnant) is meaningfully worse than lithium for most patients. Depakote is fine when it’s working. Lithium is better when both would work, which is most of the time in bipolar I.

The atypical antipsychotics (Abilify, Seroquel, Latuda, Vraylar, the newer antipsychotic class used at lower doses for bipolar maintenance) have real data for maintenance and are easier to start because there are no labs and no titration card. The catch is the metabolic side effects (weight, blood sugar, cholesterol going the wrong way) compound over decades, and being on one of these for twenty years isn’t free. For acute mania we’ll often use Seroquel or Zyprexa to get a patient down quickly while lithium loads in the background.

Lamictal is great for bipolar II where the depression is the problem. It does close to nothing for the manic side of bipolar I, so if you have bipolar I, Lamictal alone is not enough. Lamictal combined with lithium or an atypical is reasonable.

Where I land

Lithium is underused in modern American psychiatry, and the reasons are mostly economic and cultural rather than clinical. Nobody is marketing it because there’s no money in marketing it, the lab thing scares off prescribers who’d rather pick a drug without monitoring, and most patients haven’t been told that the boring metal salt is the one with the best long-term math for bipolar I. If you have bipolar I and you’ve never been offered lithium, the question to ask your prescriber is why not, and “we usually start with something newer” is not actually an answer.

If you want the prescription, you get the prescription. I’m a provider, not a parent. My job is the honest take, your job is the choice. With lithium specifically my honest take is more enthusiastic than with most things… not because it’s painless, it isn’t, but because for the patient who actually has bipolar I it’s the drug with the cleanest claim to keeping you alive.

Sources

Cipriani A, Hawton K, Stockton S, Geddes JR. Lithium in the prevention of suicide in mood disorders: updated systematic review and meta-analysis. BMJ. 2013;346:f3646. PMID 23814104. (Suicide-reduction meta-analysis)
Geddes JR, Burgess S, Hawton K, Jamison K, Goodwin GM. Long-term lithium therapy for bipolar disorder: systematic review and meta-analysis of randomized controlled trials. Am J Psychiatry. 2004;161(2):217-222. PMID 14754766.
Yatham LN, Kennedy SH, Parikh SV, et al. CANMAT and ISBD 2018 guidelines for the management of patients with bipolar disorder. Bipolar Disord. 2018;20(2):97-170. PMID 29536616.