Treatment of severe lithium-induced polyuria with amiloride

Nrf2 activation protects against lithium-induced nephrogenic diabetes insipidus

Lithium (Li) is the mainstay pharmacotherapeutic mood stabilizer in bipolar disorder. Its efficacious use is complicated by acute and chronic renal side effects, including nephrogenic diabetes insipidus (NDI) and progression to chronic kidney disease (CKD). The nuclear factor erythroid-derived 2-related factor 2 (Nrf2) pathway senses and coordinates cellular responses to oxidative and electrophilic stress. Here, we identify that graded genetic activation of Nrf2 protects against Li-induced NDI (Li-NDI) and volume wasting via an aquaporin 2-independent mechanism. Renal Nrf2 activity is differentially expressed on functional segments of the nephron, and its activation along the distal tubule and collecting duct directly modulates ion transporter expression, mimicking paradoxical effects of diuretics in mitigating Li-NDI. In addition, Nrf2 reduces cyclooxygenase expression and vasoactive prostaglandin biosynthesis. Pharmacologic activation of Nrf2 confers protective effects, confirming this pathway as a potentially novel druggable target for the prevention of acute and chronic renal sequelae of Li therapy.

Drug Interactions with Lithium: An Update

Lithium has been used for the management of psychiatric illnesses for over 50 years and it continues to be regarded as a first-line agent for the treatment and prevention of bipolar disorder. Lithium possesses a narrow therapeutic index and comparatively minor alterations in plasma concentrations can have significant clinical sequelae. Several drug classes have been implicated in the development of lithium toxicity over the years, including diuretics and non-steroidal anti-inflammatory compounds, but much of the anecdotal and experimental evidence supporting these interactions is dated, and many newer medications and medication classes have been introduced during the intervening years. This review is intended to provide an update on the accumulated evidence documenting potential interactions with lithium, with a focus on pharmacokinetic insights gained within the last two decades. The clinical relevance and ramifications of these interactions are discussed.

Renal aquaporins and water balance disorders

BACKGROUND

Aquaporins (AQPs) are a family of proteins that can act as water channels. Regulation of AQPs is critical to osmoregulation and the maintenance of body water homeostasis. Eight AQPs are expressed in the kidney of which five have been shown to play a role in body water balance; AQP1, AQP2, AQP3, AQP4 and AQP7. AQP2 in particular is regulated by vasopressin.

SCOPE OF REVIEW

This review summarizes our current knowledge of the underlying mechanisms of various water balance disorders and their treatment strategies.

MAJOR CONCLUSIONS

Dysfunctions of AQPs are involved in disorders associated with disturbed water homeostasis. Hyponatremia with increased AQP levels can be caused by diseases with low effective circulating blood volume, such as congestive heart failure, or osmoregulation disorders such as the syndrome of inappropriate secretion of antidiuretic hormone. Treatment consists of fluid restriction, demeclocycline and vasopressin type-2 receptor antagonists. Decreased AQP levels can lead to diabetes insipidus (DI), characterized by polyuria and polydipsia. In central DI, vasopressin production is impaired, while in gestational DI, levels of the vasopressin-degrading enzyme vasopressinase are abnormally increased. Treatment consists of the vasopressin analogue dDAVP. Nephrogenic DI is caused by the inability of the kidney to respond to vasopressin and can be congenital, but is most commonly acquired, usually due to lithium therapy. Treatment consists of sufficient fluid supply, low-solute diet and diuretics.

GENERAL SIGNIFICANCE

In recent years, our understanding of the underlying mechanisms of water balance disorders has increased enormously, which has opened up several possible new treatment strategies. This article is part of a Special Issue entitled Aquaporins.

Lithium reduces aquaporin-2 transcription independent of prostaglandins

Vasopressin (AVP)-stimulated translocation and transcription of aquaporin-2 (AQP2) water channels in renal principal cells is essential for urine concentration. Twenty percent of patients treated with lithium develop nephrogenic diabetes insipidus (NDI), a disorder in which the kidney is unable to concentrate urine. In vivo and in mouse collecting duct (mpkCCD) cells, lithium treatment coincides with decreased AQP2 abundance and inactivation of glycogen synthase kinase (Gsk) 3β. This is paralleled in vivo by an increased renal cyclooxygenase 2 (COX-2) expression and urinary prostaglandin PGE(2) excretion. PGE(2) reduces AVP-stimulated water reabsorption, but its precise role in lithium-induced downregulation of AQP2 is unclear. Using mpkCCD cells, we here investigated whether prostaglandins contribute to lithium-induced downregulation of AQP2. In these cells, lithium application reduced AQP2 abundance, which coincided with Gsk3β inactivation and increased COX-2 expression. Inhibition of COX by indomethacin, leading to reduced PGE(2) and PGF(2α) levels, or dexamethasone-induced downregulation of COX-2 both increased AQP2 abundance, while PGE(2) addition reduced AQP2 abundance. However, lithium did not change the prostaglandin levels, and indomethacin and dexamethasone did not prevent lithium-induced AQP2 downregulation. Further analysis revealed that lithium decreased AQP2 protein abundance, mRNA levels and transcription, while PGE(2) reduced AQP2 abundance by increasing its lysosomal degradation, but not by reducing AQP2 gene transcription. In conclusion, our data reveal that in mpkCCD cells, prostaglandins decrease AQP2 protein stability by increasing its lysosomal degradation, indicating that in vivo paracrine-produced prostaglandins might have a role in lithium-induced NDI via this mechanism. However, lithium affects also AQP2 gene transcription, which is prostaglandin independent.

Lithium: updated human knowledge using an evidence-based approach. Part II: Clinical pharmacology and therapeutic monitoring

After a single dose, lithium, usually given as carbonate, reaches a peak plasma concentration at 1.0-2.0 hours for standard-release dosage forms, and 4-5 hours for sustained-release forms. Its bioavailability is 80-100%, its total clearance 10-40 mL/min and its elimination half-life is 18-36 hours. Use of the sustained-release formulation results in 30-50% reductions in peak plasma concentrations without major changes in the area under the plasma concentration curve. Lithium distribution to the brain, evaluated using 7Li magnetic resonance spectroscopy, showed brain concentrations to be approximately half those in serum, occasionally increasing to 75-80%. Brain concentrations were weakly correlated with serum concentrations. Lithium is almost exclusively excreted via the kidney as a free ion and lithium clearance is considered to decrease with aging. No gender- or race-related differences in kinetics have been demonstrated. Renal insufficiency is associated with a considerable reduction in renal clearance of lithium and is considered a contraindication to its use, especially if a sodium-poor diet is required. During the last months of pregnancy, lithium clearance increases by 30-50% as a result of an increase in glomerular filtration rate. Lithium also passes freely from maternal plasma into breast milk. Numerous kinetic interactions have been described for lithium, usually involving a decrease in the drug's clearance and therefore increasing its potential toxicity. Clinical pharmacology studies performed in healthy volunteers have investigated a possible effect of lithium on cognitive functions. Most of these studies reported a slight, negative effect on vigilance, alertness, learning and short-term memory after long-term administration only. Because of the narrow therapeutic range of lithium, therapeutic monitoring is the basis for optimal use and administration of this drug. Lithium dosages should be adjusted on the basis of the serum concentration drawn (optimally) 12 hours after the last dose. In patients receiving once-daily administration, the serum concentration at 24 hours should serve as the control value. The efficacy of lithium is clearly dose-dependent and reliably correlates with serum concentrations. It is now generally accepted that concentrations should be maintained between 0.6 and 0.8 mmol/L, although some authors still favour 0.8-1.2 mmol/L. With sustained-release preparations, and because of the later peak of serum lithium concentration, it is advised to keep serum concentrations within the upper range (0.8-1 mmol/L), rather than 0.6-0.8 mmol/L for standard formulations. It is controversial whether a reduced concentration is required in elderly people. The usual maintenance daily dose is 25-35 mmol (lithium carbonate 925-1300 mg) for patients aged <40 years; 20-25 mmol (740-925 mg) for those aged 40-60 years; and 15-20 mmol (550-740 mg) for patients aged >60 years. The initial recommended dose is usually 12-24 mmol (450-900 mg) per day, depending on age and bodyweight. The classical administration schedule is two or three times daily, although there is no strong evidence in favour of a three-times-daily schedule, and compliance with the midday dose is questionable. With a modern sustained-release preparation, the twice-daily schedule is well established, although one single evening dose is being recommended by a number of expert panels.

Amiloride blocks lithium entry through the sodium channel thereby attenuating the resultant nephrogenic diabetes insipidus

Lithium therapy frequently induces nephrogenic diabetes insipidus; amiloride appears to prevent its occurrence in some clinical cases. Amiloride blocks the epithelial sodium channel (ENaC) located in the apical membrane of principal cells; hence one possibility is that ENaC is the main entry site for lithium and the beneficial effect of amiloride may be through inhibiting lithium entry. Using a mouse collecting duct cell line, we found that vasopressin caused an increase in Aquaporin 2 (AQP2) expression which was reduced by clinically relevant lithium concentrations similar to what is seen with in vivo models of this disease. Further amiloride or benzamil administration prevented this lithium-induced downregulation of AQP2. Amiloride reduced transcellular lithium transport, intracellular lithium concentration, and lithium-induced inactivation of glycogen synthase kinase 3beta. Treatment of rats with lithium downregulated AQP2 expression, reduced the principal-to-intercalated cell ratio, and caused polyuria, while simultaneous administration of amiloride attenuated all these changes. These results show that ENaC is the major entry site for lithium in principal cells both in vitro and in vivo. Blocking lithium entry with amiloride attenuates lithium-induced diabetes insipidus, thus providing a rationale for its use in treating this disorder.

Amiloride restores renal medullary osmolytes in lithium-induced nephrogenic diabetes insipidus

In lithium-induced nephrogenic diabetes insipidus (NDI), alterations in renal medullary osmolyte concentrations have been assumed but never investigated. Amiloride can modify lithium-induced NDI, but the impact of amiloride in lithium-induced NDI on renal medullary osmolytes, aquaporins, and urea transporters is unknown and is the basis of this study. Rats fed lithium (60 mmol/kg dry food) over 4 wk developed NDI. Urine osmolality fell to 287 ± 19 mosmol/kgH2O (controls 1,211 ± 90 mosmol/kgH2O). Organic osmolytes in the renal medulla showed significant decreases compared with controls [inositol 221 ± 35 to 85 ± 10 mmol/kg protein; sorbitol 35 ± 9 to 3 ± 1 mmol/kg protein; glycerophosphorylcholine (GPC) 352 ± 80 to 91 ± 20 mmol/kg protein; and glycine betaine 69 ± 11 to 38 ± 38 mmol/kg protein]. Medullary urea content fell from 2,868 ± 624 to 480 ± 117 mmol/kg protein. Concurrent administration of amiloride (0.2 mmol/l) in the drinking water restored urine osmolality (1,132 ± 154 mosmol/kgH2O), and reduced urine volume. Medullary osmolyte content were restored to control values (inositol, 232 ± 12; sorbitol 32 ± 6; GPC, 244 ± 26; glycine betaine, 84 ± 5 mmol/kg protein). Medullary urea rose to 2,122 ± 305 mmol/kg protein. Reduced AQP2, AQP3, and urea transporter (UT-A1) expression was significantly reversed following amiloride therapy. Data presented here provide further understanding of how amiloride may substantially restore the lithium-induced impaired renal concentrating mechanism.

[Intravenous ketoprofen for severe lithium-induced polyuria]

INTRODUCTION

Although lithium has a narrow therapeutic range, it is widely used in psychiatry because of its antipsychotic and antidepressant properties. During long-term treatment, the onset of nephrogenic diabetes insipidus is common, but few cases of severe hypotonic polyuria, which would be an aggravating factor, have been reported. Appropriate treatment in such cases is an open question.

CASE

We report a case of acute lithium poisoning in a 42-year-old man, due to chronic lithium treatment (plasma lithium=2.6 mmol/L). This patient, admitted to our intensive care unit, presented neurological disorders complicated by the early emergence of severe nephrogenic diabetes insipidus. After perfusion of hypotonic solution and intravenous treatment with ketoprofen (100 mg x 3/24 h), the polyuria improved rapidly.

COMMENTS

The beneficial action of nonsteroidal antiinflammatory drugs lies in their capacity to inhibit prostaglandin synthesis. Lithium causes excess production of prostaglandins, which decrease the ability of kidneys to reabsorb free water. Some publications report indomethacin to be effective in this case. Because it is available only in oral or rectal forms, however, its effect may be delayed. Our case suggests that intravenous ketoprofen, with its rapid onset of action, is effective in the treatment of severe lithium-induced nephrogenic diabetes insipidus. Rehydration must be strictly monitored because of the risk of renal failure connected with nonsteroidal antiinflammatory drugs.

Causes of reversible nephrogenic diabetes insipidus: A systematic review

BACKGROUND

In nephrogenic diabetes insipidus (NDI), the kidney is unable to produce concentrated urine because of the insensitivity of the distal nephron to antidiuretic hormone (arginine vasopressin). In settings in which fluid intake cannot be maintained, this may result in severe dehydration and electrolyte imbalances. The risk for conversion of reversible to irreversible NDI seems to be a potential complication. This review summarizes the reversible causes of acquired NDI to facilitate earlier recognition and more effective treatment by clinicians.

METHODS

Two reviewers independently searched MEDLINE, Experta Medica (EMBASE), and ISI bibliographic databases. Human studies that described NDI caused by drugs, substances, or metabolic disturbances were included. To evaluate the causal role of the risk factor, data were abstracted according to Koch's postulates.

RESULTS

One hundred fifty-five studies published between 1957 and March 2004 described 30 risk factors. Of 155 studies, 58 studies provided a "definite" diagnosis of NDI; 83 studies, a "probable" diagnosis; and 14 studies, a "possible" diagnosis. Nine factors were considered "definite" causes of NDI; 15 factors, "probable" causes; and 6 factors, "possible" causes. The most reported risk factors were lithium (84 studies), antibiotics (16 studies), antifungals (11 studies), antineoplastic agents (9 studies), antivirals (8 studies), and metabolic disturbances (8 studies). Duration of NDI reversal, as well as conversion to irreversible symptoms, seemed to depend on the duration of exposure.

CONCLUSION

Most risk factors for reversible NDI were medications, and their identification and removal resulted in resolution of the condition. Long-term treatment with lithium seemed to result in irreversible NDI.

Treatment of lithium-induced diabetes insipidus with amiloride

A 63-year-old African-American woman was admitted to the hospital with urosepsis and altered mental status. She had a history of schizophrenia and was treated with olanzapine 5 mg/day and lithium carbonate 300 mg 3 times/day. During her hospital stay, her sodium level and serum osmolality increased and her urine osmolality decreased, whereas her lithium levels remained within normal limits. Based on these findings, the patient was diagnosed with diabetes insipidus secondary to lithium therapy and was treated successfully with amiloride. Clinicians have been aware of lithium toxicity for many years and traditionally have administered thiazide diuretics for lithium-induced polyuria and nephrogenic diabetes insipidus. Recently, amiloride, a potassium-sparing diuretic, has been reported as a successful treatment for nephrogenic diabetes insipidus.

Differential pharmacokinetics of lithium in elderly patients

The pharmacotherapeutic use of lithium in the elderly as acute and maintenance therapy in bipolar disorder and augmentation therapy for major depression is well documented. Differences in the response to lithium are explained, in part, by the effect of age-related physiological changes, comorbid conditions, and concurrent medications on the pharmacokinetics of lithium in the elderly. The pharmacokinetic profile of lithium has been studied for many years, primarily in younger adult populations. Lithium pharmacokinetics may be influenced by a number of factors including age. It was first noted several years ago that elderly individuals required lower doses of lithium to achieve serum concentrations similar to those observed in younger adults. This is due to the combination of a reduced volume of distribution and reduced renal clearance. The composition of the human body changes with aging producing an increase in body fat, a decrease in fat-free mass and a decrease in total body water. Lithium clearance decreases as the glomerular filtration rate decreases with increasing age. The effects of other medical conditions in the elderly on the pharmacokinetics of lithium are less well delineated. Reduced lithium clearance is expected in patients with hypertension, congestive heart failure or renal dysfunction. Larger lithium maintenance doses are required in obese compared with non-obese patients. The most clinically significant pharmacokinetic drug interactions associated with lithium involve drugs which are commonly used in the elderly. Thiazide diuretics, ACE inhibitors, and nonsteroidal anti-inflammatory drugs can increase serum lithium concentrations. The tolerability of lithium is lower in the elderly. Neurotoxicity clearly occurs in the elderly at concentrations considered 'therapeutic' in general adult populations. There are no placebo-controlled randomised trials of lithium in old age and recommendations for clinical use are based on extrapolations from pharmacokinetic studies, anecdotal reports from mixed age populations and clinical experience in old age psychiatry. Serum concentrations of lithium need to be markedly reduced in the elderly population and particularly so in the very old and frail elderly.

Clinical relevance of drug interactions with lithium

Although lithium continues to be regarded as the treatment of choice for bipolar disorders, the clinical use of this mood stabiliser is associated with an extremely narrow therapeutic range. Relatively minor increases in serum concentrations may induce serious adverse sequelae, and concentrations within the therapeutic range may result in toxic reactions. The safety of combining lithium with other medications, therefore, is a major concern, and extensive clinical experience has served to identify several significant drug interactions. Lithium removal from the body is achieved almost exclusively via renal means. As a result, any medication that alters glomerular filtration rates or affects electrolyte exchange in the nephron may influence the pharmacokinetic disposition of lithium. Concomitant use of diuretics has long been associated with the development of lithium toxicity, but the risk of significant interactions varies with the site of pharmacological action of the diuretic in the renal tubule. Thiazide diuretics have demonstrated the greatest potential to increase lithium concentrations, with a 25 to 40% increase in concentrations often evident after initiation of therapy. Osmotic diuretics and methyl xanthines appear to have the opposite effect on lithium clearance and have been advocated historically as antidotes for lithium toxicity. Loop diuretics and potassium-sparing agents have minor variable effects. Nonsteroidal anti-inflammatory drugs (NSAIDs) have also been associated with lithium toxicity, although the relative interactive potential of specific NSAIDs is difficult to determine. Small prospective studies have demonstrated large interindividual differences in lithium clearance values associated with different NSAIDs. A growing body of evidence also suggests that ACE inhibitors may impair lithium elimination, but further investigations are needed to identify patients at risk. Anecdotal reports have linked numerous medications with the development of neurotoxicity without an apparent effect on the pharmacokinetic disposition of lithium. Antipsychotics, anticonvulsants and calcium antagonists have all be implicated in a sufficient number of case reports to warrant concern. As these medications have all been commonly coadministered with lithium, the relative risk of serious interactions appears to be quite low, but caution is advised.

Nephrotoxicity of xenobiotics

Nephrotoxicity can be grouped by the xenobiotics place of action, by the clinical presentation or by the generic toxic effect. The latter can be dose related, indirect, idiosyncratic or allergic. Nephrotoxicity of lithium, demeclocycline, aminoglycosides, cyclosporine, mercuric ion, nonsteroidal anti-inflammatory drugs, methoxyflurane, ethylene glycol, D-penicillamine and methicillin is reviewed in light of all these three viewpoints, but emphasis is on toxic mechanisms.

The consulting psychiatrist and the polydipsia-hyponatremia syndrome in schizophrenia

OBJECTIVE

The authors seek to extend understanding and treatment of hospitalized schizophrenics presenting with complications of polydipsia and dilutional hyponatremia. Attending physicians may ask the consultation/liaison psychiatrist to see schizophrenics with hyponatremically-induced delirium or other psychiatric syndromes. The referring physician may or may not have identified polydipsia and dilutional hyponatremia and their complications. This article will help the consultation/liaison psychiatrist recognize early evidence of water imbalance, describe evaluation, and provide somatic and behavioral treatment approaches to this life-threatening syndrome.

METHOD

Over the past ten years, the authors have treated more than 100 patients with the polydipsia-hyponatremia syndrome. The authors discuss their and others' experience with drugs that help and hinder patients suffering from dilutional hyponatremia. They review current key articles from the polydipsia-hyponatremia syndrome literature including articles identified via Medline search 1985-94.

RESULTS

Schizophrenics with the polydipsia-hyponatremia syndrome most commonly present with polydipsia, polyuria, urinary incontinence, cognitive, affective, and behavioral changes, seizures, or coma. Quantitating polydipsia, hyponatremia, and diurnal changes in body weight facilitate therapeutic interventions. Treatment include patient and caregiver education, drug therapies to better treat psychosis and better treat osmotic dysregulation, behavioral interventions to interdict polydipsia, and diurnal weight monitoring.

CONCLUSIONS

Once recognized, acute, subacute, and chronic complications of the polydipsia-hyponatremia syndrome are readily treatable. Besides treating the patient, consultation/liaison psychiatrists can teach their medical colleagues about this syndrome. In so doing, they will enhance the quality of their patients' lives and help the internist and surgeon feel more comfortable when working with schizophrenics.

Ziskind-Somerfeld Research Award 1993. Biochemical, behavioral, and clinical studies of the role of inositol in lithium treatment and depression

Lithium (Li) reduces brain inositol levels by inhibiting the enzyme inositol monophosphatase. The enzyme inositol-1-phosphatase was measured in human red blood cells of controls, Li-free bipolar patients, and Li-treated bipolar patients and was found to be reduced by 80% in Li-treated bipolars, thus supporting the concept that chronic Li at therapeutic concentrations inhibits this enzyme. Two behaviors in rats caused by Li, reduction of rearing, and Li-pilocarpine seizures, are reversed by intracerebroventricular replenishment of inositol. The reversal is stereospecific to the naturally occurring myo-inositol; whereas the stereoisomer L-chiro-inositol is ineffective. The reversal is dose-dependent, requiring a dose consistent with known quantities of brain inositol depletion; and is time-dependent, as inositol must be given 1-8 h before stimulation. High-dose peripheral inositol also reverses the limbic seizures induced by Li-pilocarpine, and using gas chromatography was shown to increase brain inositol levels that had been reduced by Li treatment. Low-dose inositol could be shown to reverse a peripheral Li-induced side effect, polyuria/polydipsia, in rats and in patients treated with Li. A higher dose of inositol markedly reduced Hamilton Depression Ratings in 9 of 11 unipolar major depressive disorder patients previously unresponsive to tricyclics, in an open design, but had no effect on chronic schizophrenics in a controlled double-blind randomized crossover trial. A new inositol monophosphatase inhibitor, a fungal product originally discovered as a complement inhibitor, was found to act like Li and lower the seizure threshold for subconvulsant doses of pilocarpine. These data suggest that inositol monophosphatase inhibition is a key mechanism of Li's therapeutic action and that design of new inositol monophosphatase inhibitors may be a practical strategy to create new compounds with Li-like therapeutic effects.

Psychotropic drug use in the medically ill. Part II

Underlying medical illness and drug interactions may make the use of psychotropic agents problematic in some physically ill patients. This overview, published in two parts, discusses six major classes of psychotropic medications (cyclic antidepressants, monoamine oxidase inhibitors, benzodiazepines, neuroleptics, lithium, psychostimulants, and carbamazepine) and examines their use in the setting of specific types of medical illnesses (e.g., cardiovascular, pulmonary, hepatic, and renal disease). Practical considerations in using psychotropic medications in medical-surgical patients--particularly those who are elderly or medically debilitated--will receive special emphasis.