Sometimes we have a window of opportunity to make a difference. In this series we discuss time-frames where lithium, clozapine, and metformin have the greatest benefits.

Publication Date: 02/17/2025

Duration: 19 minutes, 28 seconds

KELLIE NEWSOME: Lithium was the…. Let me see…. Yes, it was the third element to form in the universe, right after hydrogen and helium. But we have some new evidence that places it first-line in bipolar disorder. 

CHRIS AIKEN: Welcome to the Carlat Psychiatry Podcast, keeping psychiatry honest since 2003. I’m Chris Aiken, the editor-in-chief of the Carlat Report.  

KELLIE NEWSOME: And I’m Kellie Newsome, a psychiatric NP and a dedicated reader of every issue.  

CHRIS AIKEN: Last week, we talked about how to make a diagnosis of bipolar disorder faster with the Rapid Mood Screener. Without it, only 30% of people with bipolar get an accurate diagnosis when they first come to treatment, and this 2-minute screener can increase those odds to 80%. But even when the diagnosis is made in time, the right treatment doesn’t always get followed. Antidepressants are still the most common treatment for diagnosed bipolar disorder, followed by antipsychotics. Lithium, meanwhile, peaked in the 1980s and early 90s. Since then, prescriptions for lithium have fallen by more than 50%. But, some see lithium as a disease-modifying agent in bipolar, a medication that not only treats the symptoms but also reverses some of the pathophysiology behind bipolar. What that suggests a window of opportunity where starting lithium early can change the course of the illness. Today, we’re going to look at the evidence for that, none of which is definitive, but all of which is highly suggestive.  

KELLIE NEWSOME: But, before we get into the evidence, let’s cut to the chase. This isn’t some left-field idea. It’s in the latest practice guidelines. In 2023, the International Society for Bipolar Disorders published recommendations from the Taskforce on early intervention in bipolar. Usually, these guidelines are quite bland. They list a lot of FDA-approved options and say “All of them are good, just pick one”, but this one didn’t do that. For long-term prevention, it came down in favor of one medication: Lithium; after that, they recommend the anticonvulsants, and after that the antipsychotics.  

CHRIS AIKEN: I was shocked to see that kind of certainty in a practice guideline recommendation, and we're going to review the evidence they used to arrive there today so you can make up your own mind. But, part of the reason I was shocked was because we're talking here about early intervention, and in bipolar disorder, early means younger patients, usually in their teens, and already some guidelines on child and adolescent bipolar have come out with the opposite recommendation. They say that antipsychotics work better in this younger population. Better than lithium, better than anticonvulsants. So, what's going on here? Well, notice we said that the new guidelines favor lithium for long-term treatment, the maintenance phase. When it comes to acute trials, lithium is not the clear winner. The antipsychotics work faster, and in some trials, they work better. But if you look down the road at longer-term studies, people function better and have fewer episodes on lithium than they do if they had stuck with an antipsychotic. This is a well-known difference that has been written about since the 1970s, and it occurs in both children and adults. So, any guide that recommends antipsychotics first line in children is not thinking about the difference between long-term and short-term treatment, and when it comes to bipolar, it's the long-term that matters, especially with younger people who have a long life ahead.  

KELLIE NEWSOME: Some say that lithium works better if you start it early in the course of the illness, and while that’s true for most treatments, it may be especially true for lithium. Here are some studies that have compared lithium to other bipolar therapies in the early course. 

CHRIS AIKEN: Yeah, let's get to those, but before we review those studies, let's do a reminder of the kind of bipolar patient who generally responds well to lithium, and that is the classic textbook case of bipolar, whether they have Bipolar I or Bipolar II, we're talking about those who have pure euphoric manias or hypomanias. They have full recovery between episodes, so when the episodes are gone, they have a healthy personality, and their depressions come on after the manias or hypomanias, as if they're following the law of what comes up must come down. These really good lithium responders also tend to have fewer comorbidities, and about 1 in 3 people with bipolar have these classic traits, and whether you catch them early in the illness or later, they have a really good response to lithium, often having no future episodes after starting it. 

KELLIE NEWSOME: In 2020, Danella Hafemann and colleagues followed 340 children with bipolar for about 10 years. This wasn’t randomized, but when they adjusted for confounders, they found that the ones who took lithium had half as many suicide attempts, better functioning in school and relationships, and fewer problems with depression and aggression. 

CHRIS AIKEN: That ten-year child study gives a rich portrait of long-term success with lithium, but we need a randomized trial to know if that portrait is true, and that’s what Michael Berk and colleagues brought in 2017. They compared lithium to quetiapine over a year in 61 patients with first-episode mania. Over that year-long term, lithium was superior to quetiapine in depressive symptoms, overall functioning, and in cognitive abilities on testing, and had more neuroprotection on brain imaging with lithium than with quetiapine, and some of these differences are quite large and meaningful. These young patients were 18 times more likely to stay well on lithium than they were on quetiapine.  

KELLIE NEWSOME: So far, lithium is surpassing the antipsychotics, but what about lithium versus anticonvulsants in the early course of bipolar disorder? In 2011 and 2012, Lars Vedel Kessing ran two large studies that followed several thousand patients after hospitalization for their first episode of bipolar. Although the study wasn’t randomized, lithium came out superior to valproate and lamotrigine at preventing rehospitalization – both for mania and depression. We see this pattern in a lot of other studies where lithium prevents rehospitalization better than antipsychotics, anticonvulsants, and even better than antidepressants in unipolar depression, but we’re just focusing on the ones that looked at the early course of bipolar. 

CHRIS AIKEN: And we’re not cherry-picking the data here. These were the best-designed studies of early-onset bipolar that the task force found in their systematic review, and they explain why their guidelines came out in favor of lithium in that early course. Now, if we had time, we would review all of the long-term studies with lithium (early course or not), but let's do a quick summary here of that at least. If we look at all cases of bipolar disorder where we tested things over the long term,  

> Lithium’s preventative benefits have more support than any other med, with 21 controlled trials (in the maintenance phase), compared to 1-4 trials for most other mood stabilizers and antipsychotics. 

> In the controlled trials, lithium works about as well as the other options at preventing new episodes. 

> In the observational data, lithium works much better than other options– specifically, in 9 out of 10 large observational studies, patients fared better on lithium than other treatments in bipolar, including lower risks of hospitalization, suicide, as well as new episodes. 

KELLIE NEWSOME: Let’s pause for a preview of the CME quiz for this episode. Earn CME for each episode through the link in the show notes.  

1. TRUE or FALSE. Compared to other mood stabilizers, lithium is associated with a reduction in all-cause mortality. 

So, we’re seeing a signal that lithium might work better over the long haul than other options in bipolar, especially in the early course of the illness. But why is that? Is it doing something biologically to treat the underlying cause or prevent the progression of the disease? 

CHRIS AIKEN: Lithium has a few mechanisms that might explain these effects. One of the most interesting concluded that lithium calms the hyperexcitable state of bipolar neurons at the cellular level. 

KELLIE NEWSOME: Hyperexcitable. That sounds pretty close to the pathogenesis of bipolar. 

CHRIS AIKEN: Yes, and these were young neurons, suggesting that early exposure to lithium might do something to correct the course of illness at a cellular level, much as we see in studies of people who grow up exposed to lithium in the drinking water – those people have lower rates of suicide, crime, and mood problems. What they did in this study is a bit like science fiction. They created pluripotent stem cells from skin or blood cells of people with bipolar. Pluripotent means these cells are in an embryonic state and can turn into any kind of cell, so they programmed them to turn into neurons. So, in a way, they are growing new brain cells from bipolar patients…. Kind of ethically… 

KELLIE NEWSOME: I can see where this could get into uncomfortable territory, with a little too much neuronal complexity. 

CHRIS AIKEN: That's why they are sticking to a single neuron for now. No one is trying to clone a brain of your brain. Another mechanism is circadian rhythm. On a behavioral level, bipolar is a circadian disorder, and on a genetic level as well. The closest we have come to finding the genes for bipolar is the CLOCK genes that control the biological clock. Well, lithium normalizes the expression of these genes, and it normalizes the circadian rhythm at a behavioral level. So, if your patient is a night owl – as many with bipolar are – they may get more active in the morning and have a more regular sleep rhythm if they take lithium.  

KELLIE NEWSOME: Then the is the neuroprotection. Patients really appreciate this one. Most psychiatric meds are neuroprotective – well, except for benzos and stimulants and antipsychotics – but lithium simply does more. Its neuroprotective effects are broader than other meds – we see them throughout the entire brain – and it takes place through more mechanisms. 

CHRIS AIKEN: Neuroprotection might explain how lithium prevents dementia and the long-term outcomes that are better on lithium. Bipolar is a neuroprogressive disease. Each episode is like a head injury, and as those episodes build up, the patient has more cognitive problems, and we see more degenerative changes in the brain. Some studies have compared these biological effects to aging in the brain the theory here is that lithium accelerates brain aging. And that brings up another mechanism that patients like to hear about with lithium. Lithium has anti-aging properties. As we age, the ends of our DNA, the telomeres, crumble like the pages in an old book. Lithium prevents telomeric aging. That might explain some of its long-term benefits and its anti-mortality effects. We first got a sense of this in 2005 when psychiatrists at Oxford analyzed suicide rates in randomized controlled trials of lithium. They found not just a reduction in suicide but a reduction in all-cause mortality as well. This was repeated in 2015, when two large studies found lower mortality rates with lithium than other meds in bipolar, even after removing suicides from the data. Suicide is not the main cause of death in bipolar disorder – that belongs to stroke, and cancer is up there too. It’s medical illnesses like these that are the main reason why people with bipolar disorder die 10 years earlier on average. And this might explain some of lithium’s benefit – lithium lowers the risk of cancer by 30%. And along with lamotrigine, lithium is the only mood stabilizer that does not raise the risk of stroke, and in some studies may help prevent stroke.  

KELLIE NEWSOME: These hints at a longer life were confirmed last year by the largest study yet – 2 and a half million people, and they focused on those with first-episode mania! Here, lithium was the only medication with robust antimortality effects that survived statistical sensitivity analyses, not for suicide but for all medical causes.  

CHRIS AIKEN: A lot of psychiatrists avoid lithium because they are worried about its medical safety. Well, there’s no better measure of medical safety than death, but if we look at individual medical disorders, there is a new study that challenges that assumption. This is remarkable data – they looked at the entire population of Denmark – 6 million people – and compared all medical problems that developed in those people who took lithium or one of the anticonvulsants for bipolar disorder over 10 years. Other than hypothyroidism (which was higher on lithium), there was no difference in any medical outcome – not for kidneys, not for heart- even when lithium was compared to the medically neutral lamotrigine. So that's somewhat reassuring, and a similar study about two years ago compared lithium with valproate and found that both of these mood stabilizers had the same rate of kidney problems. Now, that was a surprise, but other studies have also pointed that way. So, putting it all together, all of these mood stabilizers have some poorly defined risks, and they're all kind of blurring into the same medical risk for patients, regardless of which one they take, except hypothyroidism with lithium. That's what the study is trying to tell us. And when I say that these mood stabilizers have risks, we haven't clearly found the cause to all of these, and in the case of Depakote, there is some pathophysiology for how it might harm the kidneys. But what might be going on is not that the medication is causing the problem, but that the medication may not be helping mental health as much as lithium is, and that indirectly might be harming physical health. So, all kinds of explanations here, but in the study where lithium looked the same as Depakote on the kidneys, I do want to point out that in a subset of patients who had had toxic lithium levels, there was a higher rate of kidney problems. So, if you can avoid toxic levels, you might avoid any kidney problems, and in some studies, there are no kidney problems on lithium as long as the level is kept at or below 0.8.  

KELLIE NEWSOME: And as we’ve talked about in recent podcasts, all these benefits don’t necessarily come at a cost of greater side effects. With the exception of the ever so tolerable lamotrigine, lithium causes less weight gain, sedation, and cognitive problems than other mood stabilizers and antipsychotics. You’ve put forth a lot of facts here – clinical, biological, epidemiological – and they all point in one direction. If you start lithium early in the course of bipolar your patient has a chance of staying well longer, living longer, and – at a biological level – you might undo some of the mechanisms that cause bipolar disorder and prevent some of the neurodegeneration that worsens its course. Sounds like a good deal, but are people taking it? Nope. The guidelines tell us to start with lithium, but only 1 in 13 patients get lithium as their initial drug for bipolar disorder, and those figures are from 20 years ago. I suspect it has sunk even lower (Baldessarini et al. 2007). 

CHRIS AIKEN: In a perfect world, people with mania would get recognized and treated right away, preferably with lithium. In our world, 1 in 3 are recognized when they first come to treatment, and 1 in 13 of those get lithium… which means that 1 in 40 patients with bipolar disorder get that ideal treatment. Use the Rapid Mood Screener, follow it up with a diagnostic interview, and consider lithium next time – if you’re worried that the patient won’t accept it, start low – 300mg or even 150 mg for a week, raise it as tolerated, and check the level once you get to 900 mg. Aim for 0.6-0.8 in most cases, but aim a few notches higher, 0.8-1.0, if they are actively manic. You can usually turn it down to the 0.6-0.8 maintenance range after a few months of recovery. 

KELLIE NEWSOME: Join the Carlat conversation online through the Daily Psych feed – search for Chris Aiken MD on LinkedIn, Twitter, Facebook, and that new one – Bluesky. Thanks for and helping us stay free of industry support.