Expert Consultations: Real-World Answers
Lithium: How to Manage Dosage and Side Effects and Persuade Patients to Take It – Interview with David Osser, M.D.
PUBLISHED: 10/01/2021
Psychopharmacology
Lithium: How to Manage Dosage and Side Effects and Persuade Patients to Take It – Interview
David Osser, M.D.
Guest
David Osser, M.D.
Mohan Gautam, D.O., M.S.
Interview by
Mohan Gautam, D.O., M.S.
In this interview, Dr. David Osser discusses the use of lithium for clinical practice. This includes practical insights on lithium dosing and monitoring strategies. He also discusses important drug–drug interactions, lithium’s effects on other systems, and management of adverse effects.
Start Learning
CME Information
Bipolar DisordersMood Stabilizers
Start Learning
Download MP3
blank div
Interview Highlights
Lithium has neuroprotective effects, which can be maximized if started early in the course of managing patients with bipolar disorder.
When titrating lithium doses, it is best to consider factors such as dose timing, medication formulation, and intake of other medications that could affect its clearance.
Laboratory tests that should be requested for patients taking lithium include lithium levels, renal parameters, and thyroid function tests.
Text Version
This edition of Psychopharmacology Institute’s Expert Consultations features Dr. David Osser from Harvard Medical School. In this interview, we will discuss how lithium is used in clinical practice.

Dr. Osser-Mohan Part A

Mohan Gautam, D.O.: My first question actually comes from one of my recent residency journal clubs where we talked about an article on lithium. I brought up lithium’s neuroprotective properties and was surprised that a lot of people didn’t really believe me. Now, I know you’re an advocate for the neuroprotective effects of lithium. Will you break this down for us?

David Osser, M.D.: Well, this is one of the ways that can be very helpful in persuading reluctant patients to take lithium. They don’t know about this, and it’s a real plus. As a matter of fact, I often give them a copy of an article titled “Effects of Lithium on Cortical Thickness and Hippocampal Subfield Volumes in Psychotic Bipolar Disorder.” This was written by one of our former residents at the Harvard South Shore Program, Iraklis Giakoumatos, in 2015 and demonstrated the neuroprotective effect of lithium.

To understand that, first, you have to explain to patients that bipolar disorder seems to be a condition where you get a gradual progressive shrinkage of some areas of the brain—particularly the cortical gray matter, some white matter tracts, and parts of the hippocampus. These losses occur gradually over the years and are associated with progressive neurocognitive impairment, such as memory difficulties, of which bipolar patients increasingly complain. Sometimes, a specific cause is found; sometimes, it seems to be due to the progression of the illness.

It has been shown by Giakoumatos and colleagues that this reduction in gray matter volume can be arrested or even reversed to some extent by lithium, and no other mood stabilizers have this effect. If anything, anticonvulsants and antipsychotics tend to exacerbate the reduction in these areas over many years. On the microcellular level, like in animal studies, lithium can protect against cell damage, neuronal apoptosis, or destruction that can occur over time. Furthermore, glutamate-induced cytotoxicity can be blocked or reduced by lithium.

By the way, this effect has been shown to extend to other neurological disorders. There has been interest in lithium as a way of preventing or slowing a variety of other conditions, like mild age-related cognitive impairment and Alzheimer’s disease. Even low doses of lithium have been found to have some beneficial effect on Alzheimer’s disease. In fact, there was a recent large study by Gerhard and colleagues from 2020 where they looked at a long-term effect on 42,000 bipolar patients. After 1 year of lithium treatment, the dementia risk was reduced by a hazard ratio of 0.77. These studies were published in the British Journal of Psychiatry and Acta Psychiatrica Scandinavica if you want to look them up.

This is really a benefit that we need to share with our patients. Because it occurs across the lifespan, it’s an argument for starting lithium as early as possible in bipolar patients because some of these structural losses may not be reversible over time. The earlier you can start lithium, the better. It has been demonstrated in clinical studies that the earlier you start lithium, the greater its long-term benefit.

Dr. Gautam: You described how lithium has these neuroprotective effects in the long run. What I’m curious about is, what about in the short run—for example, if a patient says it’s decreasing my creativity? Is this a known phenomenon that needs to be balanced with the long term?

Dr. Osser: Well, it seems to have at best only a slight effect on that. In a recent study involving 262 patients, serial neurocognitive testing showed no significant progressive impairments and actually some improvements compared with controls. So, we think that what patients think is neurocognitive impairment is actually probably loss of that highly creative thinking associated with mania. They love that, and they feel they’re much more productive. But this is one of the sacrifices they have to make and accept in exchange for the beneficial mood stabilizing effect that they get, which is the prevention of the depression and the trouble they get into during mania. There is a trade-off that you need to explain to them in advance. It should be part of your discussion from the beginning—“Look, here’s what you’re going to lose on lithium. You’re going to lose those highs and the great feeling that you get and the creativity.”

I just saw a patient this week who, when he gets into a manic spell, he starts wanting to write books and novels. In his recent manic episode, he wrote 120 pages—over 20,000 words—of an autobiography. He just suddenly woke up and felt like he was really able to get going on his writing. Over the next 4 days during his mania, he wrote that much, but then he went out of the mania and hasn’t written a word since. This is what they will lose, this charged-up excitement, creativity, and intensity, but it’s not measurable in terms of cognitive impairment.

The other issue is that they may be stuck in a chronic mild depression on lithium. Lithium isn’t as good for the depressive phase as it is for the manic phase. So, if they do wind up there, it will feel like cognitive impairment, and what it would call for is adding medication for the bipolar depression to try to get them back to euthymia.

Dr. Gautam: With respect to lithium initiation, you recommend starting with the immediate-release formulation either daily or twice daily on an outpatient basis. I’m curious about your rationale for this. Would you also recommend a different approach in an inpatient basis?

Dr. Osser: I recommend the immediate release for almost everybody initiating and maintaining lithium, and I’m not alone in that. Most experts support that, because the evidence shows that it is preferable to give lithium once a day, at night, because it has close to a 24-hour half-life. This is true even for the immediate-release version, which lasts all day pretty well. There is no evidence—in fact, there is evidence to the contrary—that multiple daily dosing increases compared with once a day.

There are also the advantages of once-daily dosing, which is that you get this opportunity to have a lower lithium level during the second 12-hour portion of the 24 hours between doses. So, you have a trough level that occurs 12 hours after you take the evening dose and that lower dose, that lower level has a kidney-sparing effect. It seems to be the reason for why the once-daily dose has been associated with lower urine volumes and probably less risk for long-term kidney damage. This is why you want to dose once daily, and you want to use the immediate-release formulation—because that would have the greatest further reduction in levels during the second 12 hours. Whereas if you use the sustained-release formulation or give multiple doses per day, the level wouldn’t drop that much more, and you’d lose some of that kidney-sparing advantage.

When you’re first starting lithium, though, you might want to give it more than once a day for inpatients. I don’t think you need to do this for outpatients because you tend to build it up more slowly for an outpatient. It’s not as urgent of a situation, typically. So, you’ll give 1 tablet daily for a week, one 300 mg tablet, then maybe 2 tablets for the second week, and then 3 tablets for the third week. It’s a typical gradual course for an outpatient introduction of lithium. But with inpatients, you’re in more of a hurry, so you may want to start with 300 mg 3 times a day. By giving it 3 times a day, you can monitor for immediate side effects—gastrointestinal effects, for example—and minimize those. You might arrange for patients to take it with food, if necessary, to deal with that. So, there is a slight advantage to dosing several times a day, just while you’re initiating it more rapidly, as you typically would want to in an inpatient level. You should convert over to a once daily at bedtime dose for the entire amount, perhaps by the end of hospitalization or soon after.

Dr. Gautam: I’d like to reiterate something that you said because it wasn’t something that I was aware of. That is that once we are using lithium immediate release and we get the patient to a steady state and we have a good dosage, we do want to convert this to once-daily dosing, preferably at night. And because lithium comes in different formulations, you are recommending to keep it immediate release, all at once and at night. Is that right?

Dr. Osser: Correct. That’s the safest and best way to give lithium as far as trying to minimize long-term kidney damage, and many people don’t seem to know this. I see in my consults all over that people are giving it more than once a day and using the sustained-release formulation as their routine. That’s not what the experts are recommending, and it’s not what the evidence supports. Somehow, the message hasn’t gotten out to the bedside that this is how lithium should be prescribed.

Dr. Gautam: I certainly wasn’t aware of this, and I’m learning something right now. I’m curious about one more component in this, and that’s that once-nightly dosing can be helpful for protecting renal function. We do know that renal filtration can be decreased at night or that it’s lower at night. Is this part of your calculation in any way?

Dr. Osser: Yes. Lithium is filtered more slowly while you’re asleep. So, when you take it all at night, you will wind up with a higher lithium level than the same dose spread out during the day or taken all at once in the morning. The difference seems to be approximately 20%. So, if you have a patient on multiple daily doses and you’re going to convert them to once-daily dosing, you’ll need approximately 20% lower dose when taking it all at once. So, let’s say they were taking 1,500 mg in divided doses during the day, and you’re going to convert that to a night dose. You would give 1,200 mg at night. Now, these are averages. It could vary with individuals depending on the peculiarities of their kidney function, and you still should check levels to be sure that you’ve correctly made the conversion to a night dose. After a week on the night dose, get another level just to be sure.

Dr. Gautam: When we’re titrating lithium, we generally think lithium takes 5 to 6 days to reach steady state and then we would get a lithium level. So, I’m curious: What factors prompt you to change the dosing before 5 days are up?

Dr. Osser: Well, certainly side effects. You know, you could be getting side effects even in the first couple of days, on the inpatient unit. You’ll then wonder if there’s something that you’ve overlooked or didn’t know about that is leading this person to excrete their lithium slowly. I don’t always wait 5 to 6 days to check a level, as side effects could lead me to do otherwise. But I also do not necessarily check a level at 6 days. Let’s say that they’re tolerating it well and they’re starting to improve; I’m in no rush to check a level. I might wait 1 to 2 weeks because everything is going well, with no side effects, and the patient is getting clinically better. But at some point, probably at least within the first 2 weeks, I’d get a lithium level at steady state.

Dr. Gautam: Patients can often be on multiple medications other than lithium, and there are agents that can influence lithium levels. Let’s talk about agents that can increase the lithium levels. We’re aware of some of these, like thiazides, non-steroidal anti-inflammatory drugs (NSAIDs), angiotensin-converting enzyme (ACE) inhibitors, angiotensin II receptor blockers, metronidazole, a low sodium diet, dehydration, being older, and having diminished renal function—all of these can increase lithium levels. I’m curious: What common medications are the most important to avoid? Like, are some NSAIDs worse than others?

Dr. Osser: They’re all about the same, but there is some suggestion that sulindac, which is an NSAID, is less likely to raise lithium levels. There can be variability with this, so some people may have an increased level and you probably should check. Now, of course, you could use acetaminophen instead of an NSAID, but there can be long-term potential problems with acetaminophen causing renal disease in itself. It’s one of the confounding factors in the studies of long-term lithium effects on the kidney. If someone was also on high doses of acetaminophen over many years, how do we know whether it was the lithium or the acetaminophen that could have caused that problem? That’s been tough to sort out, but occasional or as-needed acetaminophen is a good way to go for milder pain.

Now, one ACE inhibitor probably is a little worse than the others, and that’s lisinopril. It not only raises lithium levels, but the risk of toxicity seems greater for a variety of reasons. So, probably—that’s a blood pressure medication—if possible, you want to avoid combining it with lithium.

Dr. Gautam: Now, let’s talk about agents that can do the opposite, which is to lower the lithium level. I think, in general, we’re a little bit less aware of these. Things that can lower lithium levels are caffeine, pregnancy, and the state of mania itself. What do you think about the roles of these lithium-reducing agents? The one I’m most interested in is caffeine. I’m picturing this exaggerated scenario where a patient consumes zero caffeine on the day or day before she goes to get lithium levels checked, and then the next day drinks, like, 10 energy drinks. What do you think about this type of scenario?

Dr. Osser: Well, it’s definitely an issue. When I was working at Taunton State Hospital for many years, and we had people on lithium, we would see cases where they would run out and do just what you’ve said. They’d go to the canteen and we’d find out that they were guzzling caffeinated beverages all day and turning up with low lithium levels. So, was this nonadherence? Was it the caffeine? We would try to sort that out. We would restrict their caffeine or restrict their access to it while we recheck levels and went through the differentials for why their levels were low. In outpatient practice, you definitely have to keep track of this variable as well. Caffeine is one of the few things that directly lowers lithium levels.

Another one is mania. It’s been observed that when people become manic, lithium levels may drop even though they’re still taking lithium. This may be due to the manic state changing the status of sodium/potassium ATPase, leading to more lithium going into the cells and less lithium circulating in the plasma. So, we’re not actually getting a lower level where we want it, which is in the cells, but it will appear like a lower level in the plasma. This has led to some cases of lithium toxicity in inpatient units, as physicians see these low levels and they start pumping in more lithium. The patient becomes lithium toxic, and that can happen even though their levels in blood and plasma may not be that high. So, you should be careful and check whether there were adherence problems to explain the lower lithium levels or whether it was this phenomenon.

If they were taking their lithium and despite that had another manic episode, it would suggest that the intervention of choice would be adding a different antimanic agent rather than raising lithium levels potentially to toxic levels. Once toxicity happens, it tends to sour everyone’s enthusiasm for lithium even though it may have had an important role up to then, so you should be careful in evaluating that.

Dr. Gautam: We know that lithium is obviously a great antimanic agent. This has been proven over and over again. It’s also a good agent for the prevention of depression itself. In your lecture, though, I think you mentioned that at higher doses lithium can be associated with induction of depression. Did I understand that properly?

Dr. Osser: Yes, that’s right. A meta-analysis by Severus and colleagues found that high maintenance levels of lithium were associated with increased depression. Increasing the maintenance levels of lithium is something we often are inclined to do if someone’s mania is not responding well to standard maintenance levels, which I suggest should be 0.6 to 0.75. In a recent new review, the recommended maintenance level for routine use is 0.6 to 0.8. If breakthrough mania is present, clinicians tend to increase lithium and maintain it above 0.8. The downside of that is that the patient may have more depressive episodes, at least according to Severus’s review. His recommendation is to add a second antimanic agent if you’re not happy with how lithium is working at the 0.6 to 0.8 maintenance level.

Dr. Gautam: This ties in with another thing that I was surprised about. It was your recommendation, at least in the video lecture, to avoid a lithium level above 1 even if the patient is in acute mania. Is this why?

Dr. Osser: This is probably a relatively minor reason why. I think that the major reason is that cases of severe renal toxicity from lithium are among people who have hit levels of 1 or more at various points in the course of treatment. The current thinking is that we should try to avoid, whenever possible, going over levels of 1 and especially over levels of 1.2 because even brief periods above that may be associated with a process that could lead to renal impairment. So, again, we’d try to treat with lower levels and if needed, add a second antimanic agent. Lithium isn’t that great for all kinds of mania. It’s great for classic and euphoric mania. But mixed mania, where there are several depressive symptoms along with the mania, does not respond that well to lithium. In fact, the second-generation antipsychotics and some of the anticonvulsant mood stabilizers like valproate and carbamazepine seem to work better in mixed mania. We don’t always expect lithium to be the best thing we have, and we shouldn’t be pushing it to toxic levels, especially in mixed cases, where our expectations probably should be lower.

Dr. Osser-Mohan Part B

Dr. Gautam: It’s generally recommended that we monitor lab parameters for patients taking lithium every 4 to 6 months. Among these, which lab value changes would alarm you almost immediately?

Dr. Osser: I think the main one would be kidney function tests, with creatinine and estimated glomerular filtration rate (eGFR) being the 2 usual parameters of that. Sometimes, we have creatinine levels that just gradually go up and it’s 1.1, then 1.2, then 1.3. If that’s happening and you haven’t already switched to once-a-day immediate-release lithium taken at bedtime, this would definitely be the time to do it to see whether you can arrest this creeping creatinine process. I can’t guarantee it will help, but it certainly would be the first move I would make.

Then there’s debate about what the upper limit is before you really need to consider going off lithium. Some have said it’s 1.5, and some have said 1.6. It depends on the renal specialist with whom you’re collaborating. This would be a good time to bring in that specialist to assess the person’s kidney function and see whether they believe this is as far as you can go. And then you should try to see what alternatives you have to lithium.

We’re always reluctant to do that because lithium is the best thing we have, number 1. It’s not like we have 10 other equally good options that we can just try. Even if we taper down the lithium very slowly, which we do want to do, patients just may not do as well on the other product.

The other parameter to watch is the eGFR. When the eGFR drops below 60, that’s when you start to get concerned. There’s more consensus on this item such that if the eGFR is 50 or less, you really have to see what your alternatives are to lithium. I think those are the most alarming numbers, and they do sometimes suddenly go way up. They may have been perfectly normal, then you get your next level and suddenly the creatinine level is 2. That is very alarming, and you have to swing into action, transition off lithium, and try to ensure no further damage occurs. Collaborating with your renal specialist would be well advised at that point as you transition the patient to another regimen.

Dr. Gautam: In addition to the concerns with renal function with lithium, another classic organ system that comes into question is the thyroid. Lithium is associated with thyroid concerns, because if I understand properly, it prevents the release of T4, which eventually causes thyroid stimulating hormone (TSH) elevations over time. So, would hypothyroidism really be expected over time?

Dr. Osser: It is. Hypothyroidism is quite common, and if untreated, patients can develop goiter. They may even develop goiter in the absence of abnormalities in their thyroid function test because compensatory activity may have taken place to counteract the lithium effect, and the result of that compensatory stimulation can be goiter. So, you do want to be watching this, though it is almost never a reason to discontinue lithium. We usually can manage it by keeping thyroid function tests as close to normal as possible. Actually, recent thinking has suggested that lower baseline levels of TSH are optimal for preventing treatment-resistant depression and mood cycling. It isn’t extremely well proven, but some have said you need to keep the TSH at 2 or less while the patient is on lithium. So, we’re more liberal about adding thyroxine now. We don’t wait until the TSH is 10, as it used to be, to start bringing it down, as that can result in both physiologic and psychiatric complications.

Dr. Gautam: Do you think it would be reasonable for the psychiatrist to be the one to add the thyroxine? Or do you think we should refer and have an endocrinologist or some other specialty weigh in here?

Dr. Osser: Prescribing thyroxine is not in our specialty, but we sometimes do it as an augmentation for depression. So, I think that is within the repertoire of a psychiatrist, but it definitely is a good idea to get approval or support from your collaborating primary care physician regarding your decision. Now, they are much less inclined to prescribe this, as they wait longer when treating non-lithium-related hypothyroidism than we’re inclined to. So, you’ll have to have a dialogue and achieve consensus on starting it and what dose to use. I’d say it’s probably not a decision to completely make in a silo. You should at least be conferring with your collaborating primary care physician.

Dr. Gautam: I see. In addition to lithium’s association with thyroid dysfunction, do you have thoughts about its relationship with the parathyroid, either in terms of hyperparathyroidism or hypoparathyroidism?

Dr. Osser: Recent studies are suggesting that we have some reason to be worried about hyperparathyroidism. As I mentioned in the lecture, a review showed that 18% of 423 patients treated with lithium for 13 years developed hyperparathyroidism. That’s in comparison to 1% to 4% in the general population. Some of the patients needed a parathyroidectomy, and there may be more out there that have this problem, so it should become more routine to measure calcium, parathyroid hormone, and vitamin D yearly. This hasn’t found its way into the practice guidelines yet, but it has been recommended in a number of papers, including this paper by Meehan in 2015 that I mentioned.

Recently, one of my residents asked me if I was doing this. I realized that I hadn’t been doing it, even though I mentioned it in my lectures. I’ve changed my behavior lately, and I am checking it. But there may be a lot of unnecessary testing going on for people who are not going to have a problem. But I am in accordance with what they are saying in the literature.

Dr. Gautam: Another concern may be weight gain. Lithium generally is associated with less weight gain compared with valproic acid, which I learned from you. However, one way in which lithium can cause weight gain is through increased thirst and increased water retention. I’m curious if you’d recommend the use of the diuretic amiloride and at what point you would consider starting it.

Dr. Osser: There are several mechanisms through which lithium can cause weight gain. The net increase from lithium is lower than our commonly used medications for bipolar disorder, especially valproate but also olanzapine and quetiapine. Yes, there can still be significant weight gain from lithium, but it is a little easier to deal with compared with other medications that cause weight gain. For example, olanzapine can cause insulin resistance right away, just after 1 dose. A single dose of olanzapine, 24 hours later, causes a significant and measurable change in insulin resistance and neuroinflammatory markers. This is before you’ve gained any weight at all, so it is a metabolic disturber.

With lithium, there are multiple mechanisms that can cause weight gain, some of which may be easier to deal with. The first is the diabetes insipidus-like syndrome, where patients get thirstier, leading them to drink and urinate more. This is a very common side effect. If they’re doing that, what beverages are they consuming? For example, a patient of mine loves Pepsis and was consuming these caffeinated and sugared Pepsis at a rate of 20 cans a day. So, you can imagine that he was quickly gaining weight from all those calories. Fruit juices are other people’s preference, and they have just as many calories and a lot of sugar. So, you have to find out what they’re drinking and get them to change their lifestyle to noncaloric beverages if you possibly can. Remember that caloric intake from beverages does not change the amount of food you eat in a meal. So, that is a very easy to fix cause of weight gain. If you can change your beverage choice, you could still eat just as much and save significant calories.

Now, the second possibility is the diabetes insipidus-like syndrome, which may result in some fluid retention along with all of the salt that patients eat. So, there is added water weight. This kind of weight comes on really quickly. You can get abdominal swelling within a week or 2 and put on a large number of pounds. This is not adipose at this short interval. It’s often likely to be water, and there may even be edema, such as pitting edema in the legs or even the hands. If you’re getting this more serious aspect of the syndrome, then you have a situation where you might be able to help them excrete that water with an appropriate diuretic.

The appropriate diuretic is amiloride because it spares rising lithium levels. It usually does not raise lithium levels the way other diuretics, like hydrochlorothiazide, can and hence is safer to use. There are mechanistic advantages. Amiloride can actually prevent lithium entry into certain cells, such as in the kidneys, where it might induce this diabetes insipidus-like syndrome. So, it is the diuretic of choice for this indication. You can start with 5 mg and can raise it weekly or every 2 weeks, as high as 20 mg. Ten milligrams is the usual dose.

Again, this is a medication with which we psychiatrists and prescribing clinicians are not normally that conversant. So, you may want to get input from primary care to assist you with this, in terms of whether it’s a reasonable choice and if there any side effects that you should look out for in your particular patient. But, yes, amiloride is the diuretic of choice to help them excrete that water and avoid that kind of weight gain.

Third, there’s also some long-term weight gain from hypothyroidism, as we discussed earlier. This comes over the very long term and is usually accompanied by the other symptoms of hypothyroidism, such as lethargy and cognitive impairment.

The fourth reason for weight gain is the same thing that you see very prominently with the quetiapines and olanzapines of the world, which is a craving for carbohydrates. I do warn patients about that as well when they’re starting lithium. I tell them, “Look, this may happen to you. And if you’re craving carbohydrates now, it may get worse.” So, they’re prepared for this and ready to institute measures to change their snacking preferences to prevent weight gain.

All of these should be discussed with the patient before you start lithium. Weight gain shouldn’t come as a surprise to them. You want to prepare them for this so they can take proactive steps before it gets out of hand. Through discussion, you reassure people that you know about this problem, you’ve encountered it, you have solutions, they’re workable, and hopefully you can head off some, if not all, of the weight gain that is associated with lithium.

Dr. Gautam: I’d like to spend a little bit more time on side effects, in addition to those we just discussed, because it can lead to issues with noncompliance if patients are troubled by things that emerge after starting lithium. Would you be able to discuss some of these other side effects, such as hair loss or acne?

Dr. Osser: Hair loss is fairly common with lithium and affects up to 10% of patients. For comparison, the prevalence of hair loss with valproate was 12%, just for comparison. Carbamazepine probably has a lower rate of hair loss. Hair loss is more common in women. The first thing to do is to make sure it’s not due to hypothyroidism. Hair loss is 1 of the symptoms of hypothyroidism. Then, you should get the lithium level down if it isn’t in the lower end of the usual dosing maintenance range, which is 0.6 to 0.8. Get it down there.

What else could help after that? We don’t have anything by way of evidence-based treatments. There’s lore out there that zinc and selenium can help with this because those minerals may be involved in hair generation. I reviewed the literature, and I couldn’t even find any case reports of this actually working. You could try it, though, and the proposed dosing would be 50 mg of zinc and 200 mg of selenium. These are actually found in multivitamins with minerals in similar quantities, so you might not need separate treatment with each of those. That’s about all we have.

There’s also a question of using minoxidil for hair loss. The package insert says that we shouldn’t use it for medication-related hair loss, but there are case reports that have found that it improves lithium-related hair loss. Now, you have to continue it indefinitely because as soon as you stop it, the problem comes back. So, it’s not a wonderful solution.

Sometimes, the problem will go away after a while, but it’s one of those cosmetic issues that can be a real game changer for people on lithium. It’s totally intolerable. They’d rather have any other medication you can think of, no matter how toxic, than have this side effect. So, you’ll have to do your best to use your art of persuasion that lithium’s benefits are worth these risks, this particular harm, even though it is an unpleasant one.

Dr. Gautam: How about the management of gastrointestinal (GI) disturbances? Does the once daily at night dosing help with that?

Dr. Osser: Not necessarily. This is one of the times when you may have to deviate from that plan. Nausea is probably the most common GI side effect, and it’s typically seen early, soon after starting it. It can be a game changer right away unless you have a solution for it. The usual solution is to try taking it with meals, in smaller divided doses. This leads to lower peak levels after each dose. The evidence suggests that, over time, people adjust to this and it doesn’t persist. When that happens, you can then transition them back to once-a-day dosing. It might be that once a day could be with their evening meal, at least at first, if there’s been benefit from having it on top of a meal. Maybe after that, you’ll be able to get it to actual bedtime.

Now, vomiting is more problematic. That’s more likely to be due to lithium toxicity, though, so definitely check levels. But this is probably going to be an unworkable side effect if it’s truly happening at more moderate levels. Diarrhea is in the same category of being an unpleasant and game-changing side effect if you can’t do something about it. Some 10% of people will get diarrhea from lithium. Again, high levels are the culprit in many cases, so get the levels down. The sustained release is more likely to cause diarrhea because the sustained release is absorbed more distally and therefore can be irritating in the colon and the small intestine, whereas the immediate-release and liquid formulations are absorbed proximally and cause less diarrhea. So, you definitely want to avoid the sustained release in anyone who’s getting diarrhea. And certainly, you want to avoid toxic lithium levels. Combining with alcohol can also be a real problem with diarrhea and lithium.

Dr. Gautam: What about lithium-induced tremors?

Dr. Osser: Well, they are common, and some 25% of patients will develop a tremor. The tremor is a typical action-type tremor. It’s irregular and much more rapid than a parkinsonian tremor, which is a 4- to 6-beat/second regular rhythmic tremor. The tremor of lithium is an irregular and much more rapid tremor, too quick to count the number of beats per second. It’s similar to many other causes of action tremor. There are many other causes, and you need to see whether you can eliminate those. Caffeine causes that kind of a tremor, as well as alcohol withdrawal if they happen to be going through that. There’s also familial essential tremor, which gives an identical-appearing rapid, irregular tremor. If they have a family history of tremor, they probably are more prone to get that from lithium. Anticonvulsants, such as divalproex and carbamazepine, can cause a similar tremor. If they’re on those as well as the lithium, you’ll have to try to sort out which deserves the most credit for the tremor.

The tremor also correlates with lithium level. So, first of all, you want to get the levels down, if possible, to the lower end of the range. You definitely should not let them get to toxic levels because that’s when the tremors are going to be really wicked.

After you’ve adjusted all of those factors, you can consider treatments. We do have the beta blockers, which can be used in a dose anywhere from 20 mg a day up to even 300, which there are reports of. There’s the immediate-release propranolol which only lasts a few hours, and may have to be given 3 to 4 times a day to cover the tremor all day. There’s also a sustained-release version that you can use. There are also other beta blockers that have been reported to be useful. Metoprolol, atenolol, and nadolol also work. All of these beta-blockers raise lithium levels a little, like 20%, so you have to watch out for those.

There’s also possible treatment with vitamin B6. In 1 study, 900 mg to 1200 mg a day was reportedly helpful. Primidone, which is used for essential tremor, has also been tried for lithium tremor. The downside of it is that primidone is an enzyme inducer, and it may induce the metabolism of second-generation antipsychotics and lower blood levels of these medications. Benzodiazepines can also treat these tremors. There are various reasons why you may want to avoid a benzodiazepine, but they can be used.

Dr. Gautam: This has been Dr. David Osser joining us from Boston, Massachusetts. Thank you so much, Dr. Osser. As always, here are 3 take-home messages from today’s discussion:First, after lithium has been optimally titrated, it is best to administer it once daily, at night.
Second, when converting the lithium doses from 3 times daily to once nightly, we need to consider that renal filtration decreases at night. So, to convert 3 times daily to once nightly, we use approximately 80% of the total dose. Remember that it’s best to use the immediate-release formulation.
Third, an important selling point for lithium is that unlike any other medicine for bipolar disorder, it’s the only one that has evidence for neuroprotection!