Brief Summary

The majority of this text was written in 1999 and almost all of it remains important and topical as I write this update of it (in Nov. 2014). The observations I made concerning the generally poor management of lithium treatment by doctors have been borne out a number of times in various subsequent publications demonstrating that lithium treatment is very poorly administered and monitored. I can only suggest that patients, their relatives and carers need inform themselves from reliable sources, such as this, and make sure that their doctors do the right thing. Lithium is an amazing and underused drug (strictly speaking it is a chemical element, the word drug usually infers a synthesised compound).

Evidence that it is helpful to attribute greater clinical significance to the toxic effects (on thyroid, parathyroid, and kidney) resulting from 'therapeutic' levels of lithium has continued to accumulate.

A majority of cases of serious toxicity result primarily from insufficient medical care and supervisory management of treatment, not from over-dose or drug interactions.

NDI (nephrogenic diabetes insipidus) is common (10%+), may be progressive as well as irreversible, and increases risk of serious central nervous system toxicity.

Hypothyroidism is especially common in females over 40 yrs (~20%)

Formal written procedures / guidelines / instructions for all those clinics / hospitals / medical practices taking responsibility for care of patients on lithium are a priority and a benchmark for 'good clinical practice'.

Aim to see the patient, and check they are attending (specialist) follow up and having regular tests, before issuing a Lithium script. Prescribe sufficient lithium to last till the next essential monitoring visit / blood test.

Levels between 0.4 and 0.8 mmol/L are usually most suitable for prophylaxis; but major side effects, and even toxicity, can occur within that range.

The information about lithium on may be printed / copied / distributed freely. It is important to date any printout / copy and also to give the web address so the latest version can be accessed by anyone reading the information.

Lithium update: Introduction

Note the information on this site ( is not a basic primer, nor is it a comprehensive overview of all aspects of lithium; it is an update on specific recent developments. I will update and re-reference the rest of this commentary as soon as I have time.

The information about lithium on may be printed / copied / distributed freely. It is important to date any printout/ copy and give the web address so the latest version can be accessed by anyone reading the information.

Lithium has been widely used for 40 years, yet there is still much we do not know about its therapeutic and toxic effects. Recent advances, outlined below, are a reminder of the slow and hard won advances in knowledge and also of the ephemerality of accepted wisdom and dogma. Both the latest evidence, and consensus guidelines, reinforce the appropriateness of specialist consultation / management for patients on lithium (Ref 3, 5, 6, 8, 9, 13, 26). In my opinion that is especially so because of this continually advancing knowledge. Toxic effects from Lithium (see below) are a major cause of negligence actions against doctors. Whilst levels between 0.4 and 0.8 mmol/L are usually suitable for prophylaxis toxicity does occur within that range. Above 0.8 close and expert supervision is recommended. Some laboratories and guidelines still suggest a therapeutic range as high as 1.0 or 1.2 which may be appropriate in some cases, but preferably only with close supervision by an experienced specialist.

General supervision of management.

Formal written procedures / guidelines / instructions for clinics / hospitals / medical practices taking responsibility for care of patients on lithium are a priority and a benchmark for 'good clinical practice' (Ref 5, 6).

For those GPs who undertake sole responsibility of caring for patients on lithium, and do not already have such documentation / guidelines, one current web site to consider with pdf file downloads re lithium is (Ref 5)

Doctor and patient information is available on this site.

Routine plasma estimations are done 12 hours after the last dose. Regular blood levels are only an indication of appropriate prescribing, other risk factors require consideration. Lithium is a toxic drug and evidence indicates avoidance of toxicity and side effects is promoted by close medical supervision, if possible through specialists and specialist hospital clinics (Ref 5, 6, 13).

Mogens Schou, who often attracts the epithet of the world’s leading authority on Lithium therapy, suggests ‘laboratory monitoring of Lithium treatment is important: clinical monitoring of Lithium treatment is essential’ (Ref 9).

Some principals / procedures from the clozapine monitoring system can be usefully applied to lithium, eg one may adopt a system such as below.

Do not issue a Lithium script without:--

Checking who is responsible for issuing scripts and monitoring tests.

Seeing the patient / referring them on to who is responsible

Checking they have been / are attending (specialist) follow up

Checking patients have written information / instructions

Only prescribe sufficient lithium to last till the next essential monitoring visit / blood test.

**Note:- agree with the supervising specialist for only one person to issue scripts and monitor tests. Check you have formal written procedures / guidelines / instructions.

I strongly advise all patients be given written information / instructions and that all involved in Lithium monitoring, including the practice receptionist, appreciate the toxic nature of the drug. A clear understanding with the laboratory concerning urgent reporting of significant results is a good idea.

Causes of toxicity

Some 'current' (2003) guidelines on Lithium will benefit from updating as a result of useful new toxicology studies from the Newcastle (Australia) group (Ref 8). Until now it has been presumed that toxicity is precipitated by drugs like ACE inhibitors, NSAIDs and diuretics; but it seems almost all severe cases of neurotoxicity result from chronic therapeutic intoxication, which is itself the result of failure of therapeutic monitoring by medical supervision. They are not usually the result of drug interactions or overdoses. This may be because, especially with overdoses, lithium is usually renally cleared sufficiently quickly to prevent accumulation in cerebral tissues to a degree that causes serious toxicity. It may be that the 'brain' half-life is much longer than the plasma half-life.

ACE inhibitors, NSAIDs and diuretics may tip the balance in vulnerable individuals but they rarely cause serious toxicity by themselves.

ACE inhibitors, NSAIDs and diuretics (especially thiazide) elevate Lithium by about thirty per cent. Take care when they are added to lithium; compensatory dose reduction and monitoring are needed.

Lithium excretion

Lithium passes freely through the glomerulus of the kidney, so the rate of excretion is proportional to glomerular filtration rate (GFR). It is then extensively re-absorbed (about 75%) by the proximal tubule. The proportion of lithium excreted is referred to as the fractional excretion of lithium (FE Li) and is about 25%. Reducing either GFR or FE Li will lead to retention of lithium and rises in plasma lithium levels. Hence, estimated creatinine clearance seems a better index of toxicity risk that serum creatinine.

The Cockroft-Gault formula (Ref 15) may be used:

Clcr = ((140 - age) x body weight)/(72 x Scr), x 0.85 for females

Clcr = creatinine clearance, Scr = serum creatinine

Factors increasing lithium levels

Reduction of glomerular filtration rate (GFR). Age, various medical conditions, NB including hypothyroidism.

Reduction of fractional excretion of lithium (FE Li), ACE inhibitors, NSAIDs and diuretics and thyrotoxicosis

NDI (nephrogenic diabetes insipidus) and thyroid disorder and hyperparathyroidism reduce both GFR and FE Li

NDI (nephrogenic diabetes insipidus)

NDI (nephrogenic diabetes insipidus) occurs in ~ 10% of persons on lithium (probably more with higher plasma levels) and it can precipitate serious plasma volume depletion and hence both reduced glomerular filtration rate (GFR) and reduced fractional excretion of lithium (FE Li). It may be a harbinger of more serious renal disease. In a recent series of patients with biopsy established lithium tubulointerstitial nephropathy (biopsied because of renal insufficiency and proteinuria): 87% had nephrogenic diabetes insipidus, 33% hypertension. This renal dysfunction may be irreversible despite lithium withdrawal so early detection is essential to prevent progression.

Older people on lithium are at risk of severe hypernatraemia after an acute illness or if their fluid intake is restricted. Those at risk of NDI may benefit from screening before any elective major surgery.


Recent papers (Ref 20, 27) sampled over 700 patients on Lithium; a prevalence of 'clinical hypothyroidism' of 10.4% (female 14%, male 4.5%) was found and the risk in females who started Lithium after the age of 40 was more than 20% . Whether or not to cease it and try a different treatment or continue with Lithium and add thyroxine is an important decision. It is appropriate to monitor both T4 and TSH levels. Females over forty may be advised that their risk is probably >20%. Hypothyroidism is associated with anti-thyroid peroxidase (TPO) antibodies, so screening this high risk group before and during lithium treatment may be useful. TPO-Abs and lithium appear to be independent risk factors for the development of hypothyroidism.

Hyperthyroidism (Ref 15, 23, 27) may be more common in females with affective disorder and may be uncovered on lithium cessation; perhaps by altering thyroid function it may mask the signs of hyperthyroidism by inducing cellular unresponsiveness.

Hyperparathyroidism effects ~ 3% of those on lithium (Ref 33) and may be a risk factor for lithium toxicity.

Guidelines on Lithium

Guidelines on Lithium therapy suggest generally:-

Three monthly: lithium levels

6-12 monthly: T4 and TSH (? in females > 40 yrs: anti-thyroid peroxidase (TPO) antibodies)

12 monthly: creatinine (and estimated creatinine clearance), E & U, Ca++ and early morning urine osmolality / 24 hr urine volume.

12 monthly: Weight, BP and pulse, urine dipstick.

Variations in these intervals will be influenced by eg: age, gender, medical history of risk factors, illness severity, previous illness course.

Evidence also indicates that:--

Taking Lithium for anything less than 2-3 years may increase, rather than decrease, overall morbidity, especially if Lithium is stopped suddenly, because of the risk of rebound mania.

What does ‘suddenly’ mean? No-one really knows. I suggest reducing by 25% of the dose, or 250 mg, every 6 - 8 weeks.


1. 2002 The clinical manifestations of lithium intoxication. Meltzer E, Steinlauf Se. Isr Med Assoc J;4:265-7.

2. 2002 Brain-to-serum lithium ratio and age: An in vivo magnetic resonance spectroscopy study. Moore CM, Demopulos CM, Henry ME, Steingard RJ, Zamvil L, Katic A, et al. Am J Psychiatry;159:1240-2.

3. 2002 Bipolar disorder. Muller-Oerlinghausen B, Berghofer A, Bauer M. Lancet;359:241-7.

4. 2002 Effect of abrupt change from standard to low serum levels of lithium: A reanalysis of double-blind lithium maintenance data. Perlis RH, Sachs GS, Lafer B, Otto MW, Faraone SV, Kane JM, et al. Am J Psychiatry;159:1155-9.

5. 2002 Monitoring patients on lithium-a good practice guideline. Nicholson J, Fitzmaurice B. Psychiatric Bulletin;26:351-353.

6 2001 Preventing lithium intoxication. Guide for physicians. Delva NJ, Hawken ERe. Can Fam Physician;47:1595-600.

7. 2001 Lithium-induced nephrogenic diabetes insipidus in older people. Mukhopadhyay D, Gokulkrishnan L, Mohanaruban K. Age Ageing;30:347-50.

8. 2001 Lithium toxicity: An iatrogenic problem in susceptible individuals. Oakley PW, Whyte IM, Carter GLe. Aust N Z J Psychiatry;35:833-40.

9. 2001 Lithium treatment at 52. Schou Me. J Affect Disord;67:21-32.

10. 2001 Brain lithium concentrations in bipolar disorder patients: Preliminary (7)li magnetic resonance studies at 3 t. Soares JC, Boada F, Spencer S, Mallinger AG, Dippold CS, Wells KF, et al. Biol Psychiatry;49:437-43.

11. 2000 Does lithium treatment still work? Evidence of stable responses over three decades. Baldessarini RJ, Tondo L. Arch Gen Psychiatry;57:187-90.

12. 2000 Sixteen-year mortality in patients with affective disorder commenced on lithium. Brodersen A, Licht RW, Vestergaard P, Olesen AV, Mortensen PB. Br J Psychiatry;176:429-33.

13. 2000 Lithium monitoring before and after the distribution of clinical practice guidelines. Eagles JM, McCann I, MacLeod TN, Paterson N. Acta Psychiatr Scand;101:349-53.

14. 2000 Lithium nephrotoxicity: A progressive combined glomerular and tubulointerstitial nephropathy. Markowitz GS, Radhakrishnan J, Kambham N, Valeri AM, Hines WH, D'Agati VD. J Am Soc Nephrol;11:1439-48.

15. 2000 Lithium: Thyroid effects and altered renal handling. Oakley PW, Dawson AH, Whyte IM. J Toxicol Clin Toxicol;38:333-7.

16. 2000 Brain lithium measurements with (7)li magnetic resonance spectroscopy (mrs): A literature review. Soares JC, Boada F, Keshavan MSe. Eur Neuropsychopharmacol;10:151-8.

17. 2000 Reduced suicide risk during lithium maintenance treatment. Tondo L, Baldessarini RJ. J Clin Psychiatry;61:97-104.

18. 2000 Risk of recurrence of bipolar disorder in pregnant and nonpregnant women after discontinuing lithium maintenance. Viguera AC, Nonacs R, Cohen LS, Tondo L, Murray A, Baldessarini RJ. Am J Psychiatry;157:179-84.

19. 1999 Drug-induced diabetes insipidus: Incidence, prevention and management. Bendz H, Aurell M. Drug Saf;21:449-56.

20. 1999 Lithium-associated clinical hypothyroidism. Prevalence and risk factors. Johnston AM, Eagles JM. Br J Psychiatry;175:336-9.

21. 1999 Antidepressants and suicidal risk. Muller-Oerlinghausen B, Berghofer A. J Clin Psychiatry;60:94-9; discussion 111-6.

22. 1999 Lithium-induced nephrogenic diabetes insipidus. Stone KA. J Am Board Fam Pract;12:43-7.

23. 1999 A case of lithium-associated hyperthyroidism. Yamagishi S, Yokoyama-ohta M. Postgrad Med J;75:188-9.

24. 1998 Use of psychotropic medications in breast-feeding women: Acute and prophylactic treatment. Austin MP, Mitchell PB. Aust N Z J Psychiatry;32:778-84.

25. 1998 Psychotropic medications in pregnant women: Treatment dilemmas. Austin MP, Mitchell PB. Med J Aust;169:428-31.

26. 1998 The expert consensus guidelines for treating depression in bipolar disorder. Frances AJ, Kahn DA, Carpenter D, Docherty JP, Donovan SL. J Clin Psychiatry;59:73-9.

27. 1998 Thyroid disorders in lithium-treated patients. Kirov G. J Affect Disord;50:33-40.

28. 1998 Treating recurrent affective disorders during and after pregnancy. What can be taken safely? Schou M. Drug Saf;18:143-52.

29. 1998 Lithium treatment and risk of suicidal behavior in bipolar disorder patients. Tondo L, Baldessarini RJ, Hennen J, Floris G, Silvetti F, Tohen M. J Clin Psychiatry;59:405-14.

30. 1997 Irreversible lithium neurotoxicity: An overview. Kores B, Lader MH. Clin Neuropharmacol;20:283-99.

31. 1997 Emergency treatment of lithium-induced diabetes insipidus with nonsteroidal anti-inflammatory drugs. Lam SS, Kjellstrand C. Ren Fail;19:183-8.

32. 1996 Renal function on and off lithium in patients treated with lithium for 15 years or more. A controlled, prospective lithium-withdrawal study. Bendz H, Sjodin I, Aurell M. Nephrol Dial Transplant;11:457-60.

33. 1996 Hyperparathyroidism and long-term lithium therapy--a cross-sectional study and the effect of lithium withdrawal. Bendz H, Sjodin I, Toss G, Berglund K. J Intern Med;240:357-65.

34. 1996 Lithium toxicosis in a cow. Wallace MA, Blodgett DJ. Vet Hum Toxicol;38:99-100.

35. 1990 Lithium treatment during pregnancy,delivery, and lactation: An update. Schou M. Journal of Clinical Psychiatry;51:410.