About Lithium NDI
by Dr. David Marples, Department of Physiology, University of Leeds, England

Lithium is often given to patients with manic-depressive disorder  (also sometimes called bipolar depression). It is usually very effective in such conditions, but can be toxic if the dose is not carefully monitored and controlled. Lithium can have a number of side effects, of which the most prominent are increased drinking (polydipsia) and increased urine production (polyuria). These occur  because lithium interferes with the way the kidney responds to the hormone ADH, which normally controls our urine production.This makes it an acquired form of nephrogenic (kidney-related) diabetes insipidus (NDI).

Up to half of patients on lithium notice that they drink and urinate more. In around 1/5 of patients, this increase is enough to be quite troublesome, and occasionally it is very severe: in extreme cases it can be as bad as in inherited forms of NDI.

The situation is complicated by the fact that many psychiatric patients drink more than they need anyway - so-called psychogenic polydipsia - which of course also increases urine production. Therefore it is not always easy to know if the changes are due to the lithium treatment or not.

The effect is dose related, but the dose really has to be determined by what is needed to manage the manic depression for which the drug is given. (The blood levels of lithium are closely monitored because it is easy to get into the toxic range, which can be fatal, because of effects on the brain. This is more important than NDI).

The increased urine production usually returns to normal when lithium treatment is stopped, but this may take a long time (months), and particularly after long periods of lithium treatment (years) it may be irreversible. The evidence in the literature is unclear about whether there is a structural change in the kidney (some studies report interstitial fibrosis - in effect a kind of scarring).

The NDI caused by lithium can be treated, usually with bendrofluazide diuretics (a type of water tablet, ironically, which reduces urine production in some forms of NDI, probably by altering the body's salt balance), or with non-steroidal anti-inflammatory drugs (NSAIDs) such as ibuprofen, which are more often used as pain-killers. No-one understands how these drugs work in NDI, but they seem to help in some patients. Neither of these treatments works for all patients, and although they may help, but not cure the problem completely.

About 1 in 1000 Americans are on Lithium treatment. It is the mainline treatment for bipolar depression, but I don't know what fraction of patients are on it. About 1/4 have a big enough increase
in urine output for it to be a problem to them.

As far as I know, there are no known predisposing factors. the incidence of NDI with lithium is related to the level of lithium in the blood (which of course is also what is important in controlling the depression). Therefore it is not so much the dose of lithium you are taking, but your circulating lithium levels that are important.  These should be measured routinely by the doctor anyway, because of the risk of lithium toxicity. The aim is to use as little lithium as possible while controlling the symptoms, which should also minimise the problem with NDI.

I don't know who makes lithium preparations for the US market. In the UK a range of companies provide lithium, as either the carbonate or the citrate. The latter can be used as a syrup.

Quite a lot of other drugs, some of them also psychoactive drugs, can also cause NDI, and potentially may interact with the effects of lithium, but as far as I know none actually predispose you to NDI. However, a number of drugs reduce lithium excretion, and so cause a rise in lithium levels in the blood, which will worsen NDI. Lithium also interacts with drugs in many other ways irrelevant to NDI (but of course not to the patients!). A good pharmacology book would help here.

Treatment of lithium-induced NDI is not easy, but two groups of drugs are sometimes used. The first are thiazide diuretics - it seems crazy to use a diuretic to reduce diuresis, but for some reason it often works - probably because the body loses sodium, and in an attempt to conserve that, it also retains some water. The other group are called NSAIDs - usually used a painkillers. These include drugs like indomethacin and ibuprofen. Their mechanism of action is unknown.  Neither group of drugs works for all patients, and even if they do they will not cure the problem completely, but they are worth a trial if the NDI is troublesome.

Lithium-induced NDI usually resolves over a few weeks or months after treatment is stopped. It remains controversial whether there is permanent damage, but there is reasonable evidence that treatment for many years leads to some interstitial fibrosis (effectively, scarring) in the kidneys, and that this may affect tubular function. Of course, this is potentially a problem, since many patients will take lithium for very long periods, often for the rest of their lives, and this is important to control their depression.

Physiology of lithium induced NDI: hysiology of lithium induced NDI:
Vasopressin binds to receptors on the kidney tubular cells, and in response to this another chemical, called cyclic AMP, is produced by an enzyme linked to the receptor. This cyclic AMP then activates a number of other proteins within the cells, and leads ultimately to the delivery of water channels (AQP2) to the cell surface, which makes the cells water permeable, so water can be reabsorbed from the urine. Lithium is thought to interfere with the production of cyclic AMP, and so blocks the effect of vasopressin.

Cyclic AMP may also act as a signal for the production of more AQP2 water channels, and this process may also be blocked, which would explain why lithium causes a decrease in the number of AQP2 water channels in the cells. Of course, if you have fewer water channels, it is harder to reabsorb the water through the cells. However, it looks as though lithium also affects AQP2 production in other ways as well, which we are trying to discover.

David Marples  (NDI Research)
Department of Physiology
University of Leeds
Leeds
LS2 9NQ
England.