Lithium Risks: Thyroid, Kidney, and Weight Gain Problems
(Written 12/2005; updated 1/2006)

Lithium remains one of the most widely used medications for the treatment of bipolar disorders, but has long been known to cause problems in these three respects.  What have we learned about these problems?

Most people taking lithium may not really want to know the "how" part, they just know that we have to watch these three potential problems: by measuring thyroid function and kidney function with a lab test, and weight gain with a bathroom scale. 

But others will want to know as much as they can about what is going on inside their bodies when they take this stuff.  For that group -- and because trying to write it out is a great way for me to make sure I understand it -- here is a plain-English translation of a 2005 review of these three problem areas, by a couple of British doc's.Livingstone and Rampes As they do in their article, we'll go through these three risks areas one at a time. 

Thyroid Disorders and Lithium

This section is rather long, going first into the chemistry of the lithium interaction with thyroid hormone production which occurs in some people; then into the factors that affect whether you might be one of them, and finally into the issues of autoimmune thyroid reactions. Here are the main concepts in this section: these reactions are very common, especially in women; they can be easily treated with thyroid hormone; but a few people will continue to require thyroid hormone treatment after lithium has been stopped -- they will have become hypothyroid, perhaps permanently (although that might have been their fate even without lithium). 

Why Care? What Do Lithium-induced Thyroid Problems Look Like?
From a very nice review on this subject from the UCLA Mood programKleiner, here is a list of symptoms to watch out for with lithium-induced thyroid hormone reductions: 

You can see that these symptoms are not specific for thyroid problems.  There are lots of causes of these kinds of symptoms, including the very condition for which you might be taking lithium.  So, when you see them, don't assume they indicate a thyroid problem. But they do suggest a need for a thyroid test, if you and your doctor remember to think of this issue -- which can be difficult sometimes, as there are so many reasons for these kinds of symptoms which all must be considered at once. 

How Often Does This Happen?
Different studies have produced very different numbers, ranging up to 23% of all patients taking lithium.Kleiner  The studies vary because, for one reason at least, different definitions for "hypothyroidism" have been used.  Rates in people not taking lithium range from 2.5 to 10%, so lithium is certainly not the only way to have this happen to you, and some people are clearly more vulnerable than others, but it's hard to say who. Update 2012: Men are more vulnerable than women, it appears: a recent study of people who'd been on lithium for and average of 16 years found reduced kidney function in 38% of men but only 16% of the women.Rybakowski 

Who Has These Reactions To Lithium?

The review article from which I'm paraphrasingLivingstone and Rampes includes a list of factors which make having a thyroid problem of any kind, temporary or lasting, more likely when taking lithium. Here are the most common:

How Lithium Lowers Thyroid Hormone Production
Lithium interferes with thyroid hormone production in several ways. As a result, some people become "hypothyroid" -- too little thyroid hormone -- while taking lithium. How often does this happen?  Most studies say this affects around 1 person in 10, but a few have found that almost half of the patients studied became hypothyroid.  At least we can say that this problem is very common.  It is also clearly affected by gender: women are much more likely than men to become hypothyroid while taking lithium; in two studies, the women were found to be five times more susceptible to this than men. (Guys, if you're following this, I hope you're downsizing your overall risk, just a little anyway, in the face of these numbers: remember the 1 in 10 rate, or maybe at a maximum 1 in 2 rate, is for both genders together.  So if women are as much as 5 times more likely to be in that group, us guys must be at a proportionally lower risk.  Still at risk, but lower).

How does lithium interfere with thyroid hormone production? In several ways, apparently. First, it gets in the way of building the hormone.  Thyroid hormone is made by sticking iodine molecules onto a basic dietary building block (amino acid) called tyrosine.  Four iodines are placed on each tyrosine, which is why the completed thyroid hormone molecule is often labeled T4. When other parts of your body use the T4, they begin by stripping off one of those iodines, converting the hormone to its active form, T3.  The thyroid gland (located at the base of your neck, just above your collarbone) also makes some T3, as well as T4.  But lithium interferes with the "iodination" right at the beginning of this process.  This interference is reversible: it would not continue after lithium is no longer being used.

Secondly, lithium inhibits the release of fully-formed thyroid hormone from the thyroid gland into the bloodstream. Third, it affects the conversion of T4 to T3 out in the cells to which the T4 travels in the blood (liver and muscle, for example). 

Will My Thyroid Go Back To Normal If I Get This Reaction and Then Stop Lithium?
If this reaction was temporary, and thyroid function would rebound to normal after lithium was stopped (for whatever reason), then this whole story would not be so concerning.  You could just take thyroid hormone while you're taking lithium.  This is not complex; it's nothing like having to take insulin for diabetes. The pills are cheap; the tests to get the dose right are pretty simple; going a little too high on the dose is not terribly risky overall; and there is some reason to think that thyroid hormone itself may have some mood-improving properties.  So, if there was only a temporary risk of lithium treatment, then these thyroid affects of lithium would be relatively minor risks compared to some of the other risks of mood stabilizer medications. 

But unfortunately, a some people become permanently hypothyroid from taking lithium and have to take thyroid hormone for the rest of their lives.  Remember, this is not a disaster like developing diabetes from an "atypical antipsychotic" (these common medications are described on the mood stabilizer page:  see olanzapine, risperidone, and quetiapine in particular as potential causes of this problem). Just as with the thyroid problems from lithium, diabetes may not go away when the atypical antipsychotic is stopped, but diabetes is a far more concerning problem -- much more difficult to control, requiring extensive life-style changes (unlike just taking one pill a day of thyroid hormone), and potentially causing severe long-term effects including early death or severe disability from heart attacks and strokes. But still, if lithium can permanently mess up thyroid production, even if that is not as serious as diabetes -- can we tell who's more likely to have this outcome?

From the list of factors which affect your risk of becoming hypothyroid on lithium, note the last one: antibodies in your bloodstream directed against thyroid tissue. Because triggering such reactions often leads to lasting "autoimmunity", this reaction is probably the basis of staying hypothyroid after lithium. Therefore the risk factors for permanent hypothyroidism from lithium include markers for this autoimmune reaction -- including having antibodies already before lithium.  I'll quote the British experts, as this next idea is not one many psychiatrists, or most primary care doc's, will have heard of (so you might have to quote it if you want to make this point):

In the case of patients found to have elevated thyroid autoantibodies [before lithium], it is likely that in the event they go on to develop hypothyroidism, they would require to continue [thyroid hormone] even upon discontinuation of lithium.

Drs. Livingstone and Rampes, our British guides here, note that 8% of all women, and 3% of men, have evidence of thyroid abnormalities (with no lithium around). So this potential risk of lithium causing a probably-permanent hypothyroidism is not small. 

The authors go on to recommend that blood tests for these autoantibodies be obtained before starting lithium treatment. Until I read this review I'd not encountered this idea.  I'm sure these tests are expensive. Sometimes we turn to lithium is because it's cheap. If we add this kind of testing for everyone, it may not be so cheap anymore. Our health care system is already way over budget. Can we afford this? (On the other hand, if it's you, you might want to know.)  A cheap way to guess about your immune status is to look for other family members with thyroid problems, as these autoimmune things often run in families. (They run together with bipolar disorder also, for reasons yet unclear, but more so with rapid cycling bipolar disorders.)  So perhaps a cheap way to manage this issue would be to avoid lithium, unless we really have to consider it, for anyone with a family history of thyroid problems and rapid cycling of symptoms and near-hypothyroidism. (The UCLA review on this issueKleiner wonders aloud as to whether these tests have value in predicting who might have trouble with lithium (pg 251); but then goes on to recommend the tests later (pg. 252)). 

I personally have a hunch that lithium is not really the culprit when this happens, at least not all the time. Some of those who end up hypothyroid after lithium treatment might have ended up there anyway, and lithium just tipped them earlier. After all, their family members often have become hypothyroid without lithium. Their family members also often have rapid cycling bipolar symptoms (sometimes diagnosed, sometimes not), suggesting that perhaps the thyroid problem is related to the bipolar problem -- and not necessarily to lithium!  (there are other reasons to make this connection, as you can see on my Thyroid and Bipolar page).  But, this will take a lot more research to determine; research of the kind not easily funded, because there is no pharmaceutical company in a position to make money on the research. So that means we could be waiting a long time for an answer to this one.  I don't often see recommendations in the standard writings on bipolar disorder saying "we should warn people that they may become permanently hypothyroid if they take lithium".  Whether this warning really should be issued depends on how often this happens, which we don't seem to really know. We can say that it is not rare, and we can tell people about these factors which raise the risk.

There may be mood experts who think it is irresponsible of me to put this kind of information about lithium on the internet.  "Patients should be educated about these risks by their own doctors, who are in a position to help the patient weigh the risks and benefits of lithium and other treatments."  Okay, sounds good. Or, sounded good until about the year 2000 when patients and their families started looking all over the internet for information about treatment options and started -- some of them, anyway -- to want much more information about what they were being offered than had been the routine for doctors for years.  The world of medicine has changed.  Information first, then we talk, is how it works now for some -- like you, who have read all this way. Shall we move on, then?

If I Stay on Lithium, Can I Ever Try Stopping Thyroid?
(Paraphrasing from the UCLA review again, pg. 253)  This has not been studied. Usually doctors just continue it. But for people who do not have thyroid antibodies, this might be worth trying. Here's why. Some people show signs of hypothyroidism, after starting lithium -- but only for a month or two, and then their thyroid status (e.g. their lab tests) go back to normal, even while continuing lithium. So, how would you know if you might be one of these people, if thyroid hormone had already been started? The only way to know would be to taper off later. Since you can taper very slowly (and the long lifetime of thyroid hormone in your bloodstream helps make this taper smooth), this is not a huge risk. But of course you wouldn't do this without working closely with your doctor. You both want to watch closely for signs of hypothyroidism such as from the list above, and watch your blood tests (e.g. TSH). So you need to do this together.  People who have thyroid autoantibodies are much more likely to have to stay on thyroid replacement, so this taper-off idea may not be such a great idea in those folks. 

Less Common:
Hyperthyroidism and Hyperparathyroidism From Lithium Treatment
Becoming hyperthyroid -- making too much thyroid hormone, without receiving any from the outside -- is a rare effect of lithium. In most of the few reported cases there are reasons to think something else was going on in the patient besides just lithium (e.g. most such cases occurred after years of lithium treatment).

Parathyroid hormone problems are really a completely separate hormone problem. I've just lumped them in this section to keep things a little simpler. In this case, lithium causes an increase in the action of the parathyroid (literally "next to the thyroid") glands, which sit right on top of the thyroid gland itself. These glands secrete a hormone which regulates the amount of calcium in your blood. This hormone is called, excitingly, parathyroid hormone (PTH for short).

Minor increases in calcium, which could result from the increased PTH caused by lithium, may be very mild and thus unnoticed. If your calcium was high, this would show up if your doctor ordered the full chemistry panel for your blood test, not just the cheaper mini-panel which is all that's required to check kidney function. So, according to this review, at least once during your career on lithium you should have the larger panel and check your calcium.  However, having a problem due to increased calcium, from a lithium effect, is very uncommon (these include having a kidney stone, so if that ever happened on lithium, it would be worth checking your calcium level).


There are 4 known ways to have kidney problems due to lithium. One is very common; the rest are very uncommon. These involve three different parts of the kidney.  I never quite got these three straight before attempting to write the following, which is based on the Livingstone review and another review by Gitlin. I am still not sure I've got it just right but this will give you curious types a sense of what's going on with lithium and the kidney. 

I'll oversimplify first and then acknowledge that it's more complicated.  The "bottom line": anyone who's done so well on lithium as to consider staying on it should watch, over the next 10 years, for decreasing kidney function.  This is a common risk which affects at least 1 in 10 people who take it long-term, perhaps 1 in 2.  Otherwise, there are additional risks but these are very unusual and much like other very uncommon risks you'd face with nearly any medication you consider.

1. Common: an effect on the ability of the kidney to concentrate urine ("distal tubule" effect, for those of you who know kidneys)

When taken at full doses for a decade or more, lithium can interfere with the ability of the kidney to do one of it's main jobs: make urine concentrated.  If you can't concentrate urine, then you will make dilute urine -- a lot of it.  It would take a much larger volume of urine to get rid of all the waste products we get rid of, with urine, if you can't make it concentrated.  Think of it like making orange juice from one of those frozen cans -- in reverse.  You start with the orange juice, and your kidney turns it into a much smaller amount of fluid.  No, it doesn't then freeze it, the analogy stops before that part. If you couldn't do this concentration process, you'd be pee-ing out a whole jug of orange juice instead of the much smaller can's worth. If you currently produce about 5-10 orange juice cans worth of urine per day, and lost you concentrating ability, you'd be producing 5-10 big jugs worth of urine per day. That's a lot. You'd be urinating every hour or two around the clock. That's what can happen with long-term lithium if precautions (like stopping lithium before things reach this point) are not taken. 

Okay now, notice the phrase "when taken at full doses for a decade or more" part up there. This is not a short-term risk. It really only applies to people who do so well on lithium that we choose to continue it at full doses (high blood levels, at the top of the therapeutic range) for years.  In the Livingstone review, the authors say this affects about 10% of patients who have been on the medication for 15 years or more being affected. Other authors pick a shorter interval, like 10 years.Gitlin   Another review says 20-70% of patients will be affected.Gabutti But you get the point:  this is a common risk.  The likelihood of this problem is higher for anyone who's ever had "lithium toxicity", where the blood level of lithium got much too high for some reason.  Such episodes increase the likelihood of having kidney problems years later. Thus the ways you avoid this problem are:

2. Renal tubular acidosis (failure to excrete acids)

This is a serious kidney problem which occurs rarely with lithium treatment. It's like the other rare events that can happen when you take nearly any medication.  We doctors don't routinely tell patients about events with this frequency, although one could argue we should.  At some point there is a trade-off between telling the patient every possible bad thing that could happen, no matter how unlikely; and making the medication sound so scary no one would want to take it, thereby depriving the patient of the possible benefits.  We do this nearly every time we prescribe.  If you think "whoa, I should stick with herbs", that's understandable; yet most herbal treatments are even less studied regarding their possible risks. For example, flavonoids in some herbal preparations have been associated with "severe nephropathy" -- in other words, the same kidney problems we're looking at here.  There's just no way to look at treatments without looking at risks -- some known, some unknown. 

Renal tubular acidosis is a disease that occurs when the kidneys fail to excrete acids into the urine, which causes a person's blood to remain too acidic. This causes all sorts of other problems that slowly lead to big trouble, including dying. Big trouble.  Good thing it's not common. I've never seen nor heard of a colleague who's seen a case. But it has been seen enough to be recognized as clearly associated with lithium.

3. "Glomerular" changes (the problem is with the glomerulus, part of the basic unit of kidney structure)

This gets lumped in with #1 above as a reason for inability to concentrate urine.  It's a separate problem; but as you'll see at the end of all this, maybe not so separate. 

4.  "Interstitial" changes (the problem is with the space between the units of kidney structure)

This also gets lumped in with #1 above. Gitlin says that unlike #1, this one is "not progressive with time".  Most other reviews seem to regard this one as very uncommon.

So, how much decrease in kidney function should you allow to occur before it really makes more sense to switch off of lithium? As you can imagine, that depends on how bad the symptoms were before lithium. The next medication you try, probably an anti-seizure medication like divalproex or carbamazepine, is not guaranteed to work as well as lithium. Indeed, it might now work at all. If it didn't, then you're comparing the risk of worsening kidney function versus having your symptoms come back. So, you can see, it depends on how bad your symptoms were before lithium. If they were really bad, that justifies continuing lithium longer -- in theory. Here's the formal (very formal) analysis of that decision process: Werneke


Before we leave this subject, I have to admit that although the 4 kinds of kidney problems are routinely separated in the psychiatric literature, the reality appears to be more complicated, as shown in one study which looked directly at a bunch of kidneys from patients at autopsy. These patients had been taking lithium for an average of 13 years.Markowitz  Their kidneys revealed a mixture of tubule, glomerulus, and interstitial problems in varying degrees.  In other words, it's not as simple as outlined above.  

A slow loss of function is the common thread in all these problems (except the rare #2).  Therefore we monitor kidney function routinely while people are taking lithium, watching their creatinine.  This is a simple marker of kidney function requiring only a standard blood test. 

The bottom line again: a loss of concentrating ability over many years of full-dose use is very common and must be watched for.  Low dose lithium for shorter periods of time does not carry this risk. It does carry the risk of the relatively rare and more immediate problems, #2-4 above.  

Weight Gain

Of course weight gain happens even without lithium for a lot of people. But no question, lithium can cause it.  Unlike valproate (Depakote), which announces weight gain is coming by increasing appetite to obvious extremes, lithium causes a slow but steady accumulation. But at least two of the ways lithium causes this can be avoided.

How often does lithium-induced weight gain happen? One study compared the weight trends for people taking lithium by itself versus those taking a placebo. To make lithium the only medication involved is obviously a very necessary study design, as most of the other mood medications, including antidepressant medications, also have been associated with weight gain.  This study showed that about half of the lithium group gained over 10 pounds in a year, where only about 10% of the placebo patients did.Peselow  In a records-review of patients in "natural" treatment, not a randomized trial, weight gain over the course of treatment (2 to 10 years) was over 20 pounds for 45 out of 70 patients (65%).Vendsborg  Are you kidding? More than a 50/50 chance of gaining 20 pounds, if treatment "works" and I stay on it? Sorry, that looks to be the case.  

Who gets this weight gain
Women are much more likely to get it than men:  for example, in one recent study, weight gain during the first year of treatment occurred in 47% of the women but only 18% of the men.Henry  Those who are heavier to start are also at greater risk, as are those who are "young" (that tidbit is from a 1976 publication I did not chase to find out what "young" meant in their study).

What causes the weight gain?
Looks like there are several possible causes, two of which are easily prevented.. First, remember that lithium can cause hypothyroidism and hypothyroidism can cause weight gain. So your monitoring system to make sure your thyroid status stays well within the normal range will allow you to prevent this cause.  Secondly, lithium causes thirst (by acting in your system rather like the sodium in table salt). If you respond to that thirst with sugared drinks like Pepsi or lemonade, and do that most every day, you have added a very significant source of calories to your daily intake. So, that problem can be avoided as well: carry plain old water with you wherever you go, and do not use soft drinks except as a rare treat (which is how most people ought to use them, whether taking lithium or not, so you're not really depriving yourself here, just being smart like the average person needs to be smart).

Everyone will gain some weight initially while taking lithium because it acts like sodium: it makes you thirsty and it makes you hang on to more of the water you drink. Water is heavy (a pint's a pound, as we non-metric rebels are forced to remind ourselvesj). So when you first start lithium there will be a weight gain of several pounds very quickly.  This is not the long-term weight gain described above. And that long-term weight gain does not happen to everyone.

How can this weight gain be avoided -- at least in part -- ?
Here's your plan to avoid weight gain while taking lithium. First, expect to gain up to 5 pounds in the first week or so when you get to your full dose.  This is water. Relax -- as much as you can while gaining 5 pounds. Remind yourself that this is not the big problem with lithium's long-term weight gain. I'm sure you found that sentence very reassuring.  Sorry.  We have to try to keep the big picture in mind here: giving lithium a fair chance. 

Okay, now it's time to start monitoring.  How much further weight gain are you willing to accept before you decide it's time to pull the plug on lithium? 1 pound? 2 pounds? Give yourself a little breathing room so as not to stop a possibly important medication:  say 4 or 5 more pounds? Okay, whatever weight that brings you to, that is your threshold. Make sure your doctor knows that if you go over this weight, you want to stop lithium. Weigh yourself once a week. If you go over your threshold, tell the doctor and get from her/him a tapering-off plan and a plan for the next approach you're going to try. (DO NOT STOP LITHIUM SUDDENLY. This can cause your symptoms to rebound, more intense than they were before you started. You'll probably take a month or several to taper off lithium while starting something else).

In the meantime, watch out for sugared drinks (that means none, for now; and as a rare treat, later). And of course you're going to be using the exercise plan explained in Exercise--Not the Usual Rap, right?  That would be a good idea no matter what treatment approach you're planning, but it also may give you some protection from weight gain. 

Beyond that, there is a whole essay on Weight Gain and Bipolar Treatments, but these are the lithium-specific steps. (Exercise is specific for everything!)

Thanks to Drs. Livingstone and Rampes for all the work they put into their review.