Clear, engaging, and practical updates on clinical psychiatry.

Earn CME for listening to the podcast with a multimedia subscription. Listen for free here or using Apple Podcasts, Android, or Stitcher.

Previous Post
Next Post

How to Minimize Lithium’s Side Effects

Podcast, Volume , Number ,

Print Friendly, PDF & Email

Nausea, tremor, cognition, and renal function. We bring you our top tips for managing these side effects, including when to use instant release vs. controlled release and how to optimize the serum level for patient safety.

Published On: 9/7/2020

Duration: 20 minutes, 30 seconds


Lithium. The word conjures up thoughts of toxicity, dangerous drug interactions and intolerable side effects of most physicians. But patients rate lithium higher than the antipsychotics, and in this episode we’ll tell you how to minimize its side effects.

Welcome to The Carlat Psychiatry Podcast, keeping psychiatry honest since 2003. I’m Chris Aiken, the editor in chief of The Carlat Psychiatry Report. And I’m Kellie Newsome, a psychiatric NP and a dedicated reader of every issue.

Kellie: In 2013 Google surveyed 301 people with bipolar disorder with a simple question: What has helped your bipolar? Surprisingly, only one medication made the top 10 list, which was filled with lifestyle changes like exercise, yoga, mindfulness, journaling, and sunlight. The only medication on the 10-ten list was lamotrigine, but the medication that got the second highest rating was the original, the one that bipolar researchers call the “gold standard”: Lithium.

Dr. Aiken: Part of the reason is that lithium works. When Terence Ketter ranked all the mood stabilizers by their number needed to treat in 2015, lithium and quetiapine rose to the top for their preventative effects against both poles – they were the most effective and preventing mania and depression. Lithium’s usually doesn’t treat acute episodes of depression as well as atypical antipsychotics, but it does a good job of preventing them. The other reason is tolerability. When it comes to the side effects that matter most to patients – sedation, weight gain, and cognition ¬¬– lithium’s tolerability ranks right behind lamotrigine. 

Kellie: In this month’s Carlat Report we featured an interview with Janusz Rybakowski, one of the pioneering lithium researchers.  And as usual, one of my favorite parts about the article is the table. It lists all the signs that a bipolar patient will be a lithium responder. Dr. Rybakowski estimates that 1 in 3 bipolar patients are excellent lithium responders. How excellent? In the study he did he found that these patients stayed well on lithium without any major episodes for a whole decade. But if you’re going to keep patients on lithium for 10 years you need to know how to manage its side effects, and that’s what this podcast is about. So let’s start there Dr. Aiken with one that can make patients quit lithium even before it has a chance to work: Nausea.

Dr. Aiken: Nausea is a very aversive experience. It goes right to the hippocampus and once it locks in the patients memory there they are going to avoid whatever they associate the nausea with. So you want to avoid that at all costs. With most medications the controlled release version will reduce the nausea, as will raising it slowly, and that’s true with lithium too. Taking it just after a meal or – if it’s at night – after a glass of milk – also helps. I’ll usually offer an as-needed aid along with the first script – the patient doesn’t have to use it but it empowers them to know it’s there.

Kellie: What are your top nausea aids?

Dr. Aiken: I’ll ask the patient if they prefer a natural approach or a medication. For natural, ginger capsules 1-2,000 mg daily have evidence in various kinds of nausea from pregnancy to chemotherapy. It’s best to take the ginger about an hour before a big meal, then eat the meal, then take the lithium on top of it. For medication, ondansetron 4mg q12 hr prn is well tolerated. If that doesn’t work, Promethazine has more side effects but can be used 25-50mg q8hr prn.

Kellie: What brands of ginger do you recommend?

Dr. Aiken: The FDA has not stopped in to help us there, but the national institutes of health has an office of dietary supplements that recommends two independent labs that test supplements – ConsumerLabs and US Pharmacopeia. I keep a updated list of products the best-priced options that have passed their test in on my website:

Kellie: OK, tremor is another side effect that’s hard to live with. How do you treat that?

Dr. Aiken: First, lower caffeine. Next, lower lithium if you can – but you want to keep the level 0.6-0.8 for long term prevention. In the elderly you can aim a little lower like 30% lower. For antidotes, propranolol is the most popular. Start with instant release propranolol to find the right dose, which averages around 120 mg/day. Some patients may only need the propranolol as needed, but if you decide to convert to once-a-day ER propranolol, you have to raise the dose to do so because not all of the ER is absorbed.

Kellie: How much do you raise it by?

Dr. Aiken: 150%. So if they are taking instant release 60 mg/day, they’ll need 90 mg/day of the ER.

Kellie: Any other options for tremor?

Dr. Aiken: Some patients don’t want another medication, in which case vitamin B6 has some evidence for lithium tremor. The dose on this one is real high – 900-1200 mg/day – and I’ve only found a few brands that come in that high dose – they are on my website. B6 may improve akathisia and tardive dyskinesia if they are taking an antipsychotic. But B6 is not entirely benign – high doses can very rarely cause a temporary neuropathy. Another option is the calcium-channel blocker nimodipine. It has never been studied for lithium tremor but has good evidence for essential tremor at 120 mg/day. The reason to consider it here is it also has benefits in mania and rapid cycling. Anectodally, Robert Post and I have both found nimodipine helpful for anxiety in bipolar disorder.

Kellie: You mentioned you prefer the controlled release version for nausea. Are there any situations where you’d use the instant release?

Dr. Aiken: The instant release is less likely to cause diarrhea, which can also happen on lithium. Otherwise most side effects improve with the controlled release.

Kellie: There’s been some debate about giving lithium all at night or twice a day. Where do you stand on that?

Dr. Aiken: Lithium’s half life is 24 hours, so it can be given once a day, and if you give the patient will sleep through the peak levels and any side effects associated with that. However, intuition would suggest that twice a day dosing might be easier to tolerate, especially as lithium is not a very sedating medication – only 1 in 28 patients have problematic fatigue on it by the number needed to harm, compared to 1 in 5 for most other mood stabilizers. 

Kellie: What about with the kidneys though. I’ve heard two things – some say that the kidneys do better with a bolus of lithium, like an instant release given all at night, and others say you should avoid high levels to protect the kidneys, which suggests we should spread the dose out.

Dr. Aiken: There’s a lot of mixed messaging around that. Here’s what we know. First, keeping the serum level low is the best way to prevent renal damage. Toxic levels kill renal cells, and that damage builds up every time the level rises above the toxic line. This theory is confirmed by a handful of studies, including one that came out last month. They’ve found that lithium does not seem to harm the kidneys when kept below 0.8 mmol/L, but that renal impairments rise with the number of toxic exposures (again, for the elderly that max should be 0.6). So from this point of view, spreading the dose out and using a controlled release version seems wise. 

Kellie: What’s the other side of that story?

Dr. Aiken: The actual data tells a different story. Two studies have compared twice a day to all-at-night dosing. On by Morgans Schou in 1982 and the other by Rudy Bowen in 1991. These weren’t randomized trials, but they overcome that problem somewhat because they compared patients whose doctor’s habitually gave everyone twice a day vs. those who habitually dosed all at night. In both studies the patients on all-at-night lithium had better urine concentrating abilities – so less polyuria – and that’s thought to be a marker of renal toxicity. Animal studies also back this up. So the thinking here is that while we want to avoid toxic levels, it’s also important to give the kidneys a breather – some extended trough levels where they can rest and repair. 

Kellie: OK well I tend to trust the clinical data, so I’d give lithium all at night. But should we give it all as a controlled release or instant release to save the kidneys?

Dr. Aiken:That’s the unanswered question. When it comes to all at night dosing, no one has tested CR vs. IR. In the Schou study, they actually compared CR twice a day to IR all at night – so they changed two variables: the timing and the formulation. They found the IR was more favorable for the kidneys, but later studies suggest this difference has to do with the timing of the dose rather than the formulation. My suggestion is to use the CR, both because peak levels can be toxic, and because the CR reduces the overall rate of side effects by about 50%.

Kellie: What’s the max dose you can safely give all at night? I mean I have some patients on 1,500 mg of lithium a day.

Dr. Aiken: Another unanswered question. But I’d worry more about the lithium level than the dose. Since we know that levels of 0.8 and below are safer for the kidneys, what I’ll do is space the dosing out, giving a small amount of it in the morning, if the patient requires a higher serum level to maintain stability, like 1.0 or even 1.2.  I’ve never kept anyone’s lithium above 1.2 intentionally, and that’s about as high as the literature goes on this as well. But with lithium, keep in mind that what’s toxic to one patient is harmless to another. So a young adult may tolerate a level of 1.2 and have no toxicity, while an older adult may have symptoms of toxicity at a level of 0.7.  Here I’m just talking about toxic symptoms – like severe tremor, imbalance, visual disturbance, confusion, and vomiting. Unfortunately we don’t know what’s going on in their kidneys at those levels.

Kellie: And what can we do once renal function starts to decline?

Dr. Aiken: That’s tricky. In some studies the kidneys actually fared better when patients stayed on lithium as their creatinine rose, compared to switching to an anticonvulsant…. No one knows why that is but there’s some suggestion that valproic acid can impair the kidneys. If the creatinine rises the only thing you can do is lower the lithium dose. I’d refer to nephrology if the creatinine rises to 1.5 or higher. Of course you can stop lithium, but that’s not an easy choice if lithium is the only thing they’ve responded to. Renal failure may be fatal, but most cases of renal impairment on lithium do not progress to renal failure, and lithium discontinuation can be fatal as well – if we consider that lithium is the only mood stabilizer with robust antisuicide effects.

Kellie: So there are no medications for renal toxicity on lithium?

Dr. Aiken: None. People are exploring Amiloride, which treats nephrogenic diabetes insipidus on lithium and has some animal data to prevent renal toxicity. Another one with animal data is N-acetylcysteine. They’ve studied this antioxidant in mice whose kidneys have been damaged by all sorts of toxins, including lithium, and it seems to help repair the kidneys. 

Kellie:  N-acetylcysteine is also used in bipolar disorder – there’s a handful of studies showing it treats chronic bipolar depression. Are there any human studies of it in the kidneys?

Dr. Aiken: Not that I’m aware of, but as we have nothing else available, and this one is reasonable to use in bipolar anyway, I’ll sometimes add it in if renal protection is an issue. The dose in the mouse study equates to 1,000 mg a day of n-acetylcysteine in humans, which is a little below the dose used for bipolar depression: 2,000 mg a day.

Kellie: OK here’s another paradox. I often hear that lithium is good for the brain – that it prevents dementia and has more neuroprotective effects than any other psych med. So why do patients complain of cognitive problems on it?

Dr. Aiken: We don’t have a lot of studies on that. The largest one to date just came out, and it found that cognition usually improved – and did not worsen – on lithium, while other mood stabilizers fared worse when it came to cognition. Other studies have concluded the same thing. Lamotrigine is the most favorable for cognition, but lithium usually comes in 2nd or 3rd when the mood stabilizers are stacked up.  What we need are more controlled studies – there is one where they randomized patients to quetiapine vs lithium for mania, and 1 year later the ones on the lithium had better verbal abilities. But that’s a minor difference. 

Kellie: OK but back to my question – if the research says that lithium is so good for cognition, why do patient’s complain about it?

Dr. Aiken: I think lithium does some good and some harm. The good is long term – preventing cognition decline and dementia. But patients rarely appreciate preventative effects. The harm lithium does is in cognitive slowing and possibly reduced creativity – those have been documented in some studies and though those are only two cognitive domains they may be the ones that are most important to many patients. What you’ll hear people say when this happens is that lithium makes them feel dull. About 1 in 10 patients tell me that, and most of the time it improves with lowering the dose.

Kellie: For more tips on how to use lithium check out our interview with Dr. Rybakowski  in our September issue online. 

And now for the word of the day…. Turn taking

Turn taking is one aspect of speech that we look for in the mental status exam. Basically, the speaker pauses to allow the other person to take a term, and the conversation unfolds as the two respond to each other, comment, and ask questions. Besides pauses, vocal inflection and content can indicated that it’s time to pass the baton.

When patients don’t take their turn, it may be a sign of depression, schizophrenia, severe anxiety, cognitive problems, or guardedness from paranoia. On the other side, some patients never pause to allow a turn – as in the “verbal firehose” of mania, drug intoxication, frontal lobe deficits, and autism. Patients with narcissistic or histrionic tendencies may also monopolize the conversation, as can patients with anxiety when they aren’t frozen by it.

People with ADHD may speak with a jolting, fractured rhythm, pausing at random times as if distracted in the conversation. When interviewing someone with ADHD, you may feel unsure about when they are pausing to collect their thoughts or when it’s time for you to take your turn.

Turn taking also differs by culture and gender. Early studies found men interrupted more than women, but that gender bias hasn’t panned out in more recent research. New Yorkers Japanese have short pauses, almost talking over each other, while people from Northern Europe and California allow long pauses.

 Tune in next week where we’ll interview Michael Posternak on how to talk with delusional patients

Dr. Aiken: Lithium is one of few medications that has never received industry support, because it’s been generic since it first came out in the 1950’s. The Carlat Report has not been around that long, but it has been free of industry support since its inception, and we’re grateful to our subscribers who help us keep it that way.

Got feedback? Take the podcast survey.


Leave A Comment