Comparison of the action of amiloride and thiazide in the reduction of lithium-NDI in patients with affective disorders

Lithium is the drug of choice for treating bipolar disorders. It is successful
in reducing both manic and depressive symptoms in 70-80% of the patients.
Lithium is also used to treat schizoaffective disorders and depression, and is
thus an indispensable pharmaceutical component of current psychiatric therapy.
Unfortunately, lithium often causes the development of Nephrogenic Diabetes
Insipidus (NDI), a disorder in which the kidney is unable to concentrate urine
in response to vasopressin, leading to polyuria and polydipsia.
About 0.1% of humans in Western countries are on lithium therapy. Because
lithium causes a decrease in urine concentrating ability in 50% of patients,
and in 20% leads to NDI, lithium NDI is of high clinical relevance.
In lithium-induced NDI, thiazide diuretics have been used successfully in
selected patients to reduce both polyuria and polydipsia and thiazide diuretics
are at the moment used to reduce excessive urine output in patients receiving
lithium. However, thiazide treatment may lead to hypokalemia, because thiazides
cause an increased potassium excretion.
Amiloride is a potassium-sparing diuretic, not causing hypokalemia. Amiloride
blocks sodium uptake through the epithelial sodium channel expressed in the
apical membrane of renal connecting tubules and collecting ducts. This channel
is probably a major pathway for lithium accumulation in the cell. By this
action, amiloride could directly block lithium entry into the cell, and in this
way prevent or cure NDI. In studies with a limited number of patients it was
suggested that amiloride indeed counteracts lithium-NDI to a great extent. In
view of amiloride*s potassium-sparing and less natriuretic (thus less toxicity
risk) properties, it may be advantageous over thiazide. However, formal
comparisons between the drugs are lacking and it is not common practice to
treat lithium-NDI in patients in the Netherlands with amiloride.

To compare the effect of amiloride on lithium-induced Nephrogenic Diabetes
Insipidus with the effect of hydrochlorothiazide, measured as urine volume and
maximal urine osmolality.

A double blind, randomized, parallel group, multi-center study. Amiloride will
be compared with hydrochlorothiazide

Patients will be treated with 2 x 5 mg/ day amiloride or 2 x 12.5 mg / day
hydrochlorothiazide. After 4 weeks the dose will be increased to 2 x 10 mg/day
for amiloride and 2 x 25 mg/day for hydrochlorothiazide, if no side effects are
present and potassium levels and blood pressure are still normal. After 3
months, treatment with amiloride or hydrochlorothiazide will be ended. All
patients will continue to use lithium during the study.

Patients will visit the hospital 7 times. At all these occasions, blood samples
(10 ml) wil be collected. At 4 times, a dDAVP test will be performed; dDAVP
will be administered intranasally and urine will be collected for 6 hours.
Besides this, patients are 4 times asked to collect 24 hrs urine.

Risks for the patient include side effects of the medication, or side effects
caused by changes in the plasma-lithium concentration. To avoid these risks,
patients are first treated with a lower dose of the medication, which will be
increased after a month, if no side effects are present. Besides this, the
lithium levels will be measured during the study, and the dose of lithium will
be adjusted if necessary. Other parameters, like blood pressure and potassium
concentration are also measured during the study.

The study will increase the knowledge about the best treatment for
lithium-induced NDI, and will in the study population probably also result in a
decrease of the polyuria.