Editor’s note: We know that our reviews of the withdrawal literature are incomplete, and we urge readers to help us add to them. Please send study citations that are relevant to the withdrawal literature for mood stabilizers to [email protected].
The class of drugs known as “mood stabilizers” is a disparate group of medications with different theorized mechanisms of action and effects. The classification of mood stabilizer is itself contested, with no standardized definition. This document will review the drugs lithium and the “antiepileptic” or “anticonvulsant” medications carbamazepine, lamotrigine, and valproate/divalproex/sodium valproate. As the longest prescribed drug in the class, lithium is the most thoroughly researched, although areas of needed research exist across this class of drugs as a whole.
Mechanism of Action
Various theories of the mechanisms of action of lithium and other mood stabilizers have been proposed, but no prevailing or unifying theory has been established. Thus is it unknown how the neurochemical effects of these drugs result in any therapeutic effects, nor how their withdrawal might affect users.
1) Balon R, Yeragani VK, Pohl RB, Gershon S. Lithium discontinuation: withdrawal or relapse? Compr Psychiatry 1988;29:330-4. PubMed link
Balon et al. review animal and human studies of lithium discontinuation, including the possible pathophysiological explanation for rebound phenomena. The authors conclude that the research is mixed, and that no research has specifically addressed the neurochemical basis of withdrawal from lithium.
2) Williams R, Cheng L, Mudge A, Harwood A. A common mechanism of action for three mood-stabilizing drugs. Nature 2002;417(6886):292-295. PubMed link
In this article, the authors propose that a common mechanism of action in which lithium, carbamazepine, and valproic acid “inhibit the collapse of sensory neuron growth cones and increase growth cone area.”
3) Harwood A, Agam G. Search for a common mechanism of mood stabilizers. Biochem Pharmacol 2003;66(2):179-189. PubMed link
Authors review the search for a common mechanism of action among mood stabilizers, as a means of understanding their therapeutic effects and justification for being a unified class of drugs. While reviewing some possibilities, the authors conclude that no common mechanism has been established, nor theory of common action.
4) Post R. Kindling and sensitization as models for affective episode recurrence, cyclicity, and tolerance phenomena. Neurosci Biobehav Rev 2007;31(6):858-873. PubMed link
In this article, Post outlines his kindling hypothesis of treatment of bipolar disorder, including the hypothesis and supporting research that discontinuation of lithium treatment can lead to a refractory period. In the refractory period, Post proposes that outcomes may be worse than before the introduction of treatment and that reintroduction may not achieve the same effects, although this phenomenon only occurs in 10-15% of patients.
5) Moncrieff J. The Myth of the Chemical Cure: A Critique of Psychiatric Drug Treatment. 2009; New York: Palgrave Macmillan. Publisher link
Moncrieff’s critical text on psychiatric drugs includes two chapters on drugs used to treat bipolar disorder. Moncrieff discusses the lack of consensus on a biological theory or animal model of bipolar disorder that would explain or justify the effects of mood stabilizers. She proposes that theoretically the withdrawal effects of lithium could be accounted for by a rebound from its toxic effects, resulting in excitability of the nervous system. She also reviews research on effectiveness of lithium and other mood stabilizers.
6) Schloesser RJ, Martinowich K, Maji HK. Mood stabilizing drugs: mechanisms of action. Trends Neurosci 2012;35:36-46. PubMed link
The authors discuss recent research on the effect of mood stabilizing drugs and that, “at least some of the therapeutic effects of mood-stabilizing drugs appear to be induced by activating neurotrophic and neuroprotective pathways, and related intracellular signaling pathways.” The reason for their therapeutic effects, or how to identify more effective drugs, are unknown.
7) Malhi GS, Tanious M, Das P, Coulston CM, Berk M. Potential mechanisms of action of lithium in bipolar disorder. CNS Drugs 2013;27:135-153. PubMed link
The authors review research on the various theories of mechanisms of action of lithium from microscopic to macroscopic levels, including that lithium affects enzymes involved in second messenger systems that then modulate neurotransmission, resulting in generally inhibitory effects, and that lithium may also be neuroprotective. However, it is unknown why or how lithium’s varied neurochemical effects are implicated in the treatment of mania and depression.
Animal studies have attempted to explain treatment efficacy of mood stabilizers, as well as account for any “rebound” effects of withdrawal from these drugs. In many cases, withdrawal rat studies suggest that some effects of these medications are reversible whereas others persist after discontinuation.
8) Ahluwalia P, Singhal R. Effect of low-dose lithium administration and subsequent withdrawal on biogenic amines in rat brain. Br J Pharmacol 1980;71(2):601-607. PubMed link
The authors found that lithium administration and withdrawal affected levels of tyrosine, tryptophan, noradrenaline, dopamine, 3-Methoxy-4-hydroxyphenylglycol, 3,4-dihydroxyphenylacetic, and 5-Hydroxytryptamine. Effects varied across different brain regions. Two days of withdrawal of lithium did not result in a straightforward return to baseline levels.
9) Ahluwalia P, Singhal R. Monoamine uptake into synaptosomes from various regions of rat brain following lithium administration and withdrawal. Neuropharmacology 1981;20(5):483-487. PubMed link
The authors examined noradrenaline, dopamine, and 5-hydroxytryptamine (5-HT) uptake across different brain regions during lithium administration and 2 days of withdrawal. Lithium administration resulted in variable effects on uptake across different regions; withdrawal resulted in return to control levels in all regions except the dopamine and 5-HT uptake in the striatum and 5-HT uptake in the midbrain. The authors propose that “rebound” mania following lithium withdrawal may be a result of this persistent effect on monoamine uptake.
10) Ahluwalia P, Singhal R. Effect of lithium treatment and withdrawal on uptake of noradrenaline into rat brain synaptosomes: a kinetic study. Prog Neuropsychopharmacol Biol Psychiatry 1982;6(4-6):339-342. PubMed link
The authors report on a rebound phenomenon related to noradrenaline uptake following lithium withdrawal. In control animals, two different uptake mechanisms for noradrenaline were found, one low-capacity mechanism and one high-capacity mechanism. Following withdrawal of lithium, uptake returned to control levels in the low-capacity mechanism but further increased uptake in the high-capacity mechanism. The authors posit that the noradrenaline uptake increase seen in the high-capacity mechanism may help explain “rebound” mania seen after lithium withdrawal.
11) Christensen S, Hansen B, Faarup P. Functional and structural changes in the rat kidney by long-term lithium treatment. Ren Physiol 1982;5(2):95-104. PubMed link
The authors observed renal concentrating ability is impaired during long-term (21 week) administration of lithium in rats and that these effects were completely reversed upon drug withdrawal, despite reports that impairment can persist in humans.
12) Ahluwalia P, Singhal R. Comparison of the changes in central catecholamine systems following short- and long-term lithium treatment and the consequences of lithium withdrawal. Neuropsychobiology 1984;12(4):217-223. PubMed link
This study provides support that withdrawal of lithium, in this case following either short-term or long-term treatment, does not lead to a return to normal states in central catecholamine systems in the rat brain.
13) Lerer B, Globus M, Brik E, Hamburger R, Belmaker R. Effect of treatment and withdrawal from chronic lithium in rats on stimulant-induced responses. Neuropsychobiology 1984;11(1):28-32. PubMed link
The authors found inhibition of hyperactivity induced by a lower dose of stimulants but not by a higher dose during administration of lithium. After withdrawal from lithium, rats showed a subsensitivity to this hyperactivity response; the authors theorize a relationship between this subsensitivity and reports of psychotic symptoms in humans following withdrawal.
14) Ahluwalia P, Singhal R. Kinetics of the uptake of monoamines into synaptosomes from rat brain. Consequences of lithium treatment and withdrawal. Neuropharmacology 1985;24(8):713-720. PubMed link
Results of this study suggest increased uptake of dopamine during lithium treatment across observed brain regions, whereas withdrawal resulted in decreased uptake below control levels.
15) Berggren U. The effect of chronic lithium administration and withdrawal on locomotor activity and apomorphine-induced locomotor stimulation in rats. J Neur Transm 1988;71(1):65-72. PubMed link
Berggren observed no change in in apomorphine-induced locomotor stimulation in rats following administration of lithium, but reported an increase in stimulation following withdrawal. This effect was short-term, with no difference found 4 days after withdrawal, suggesting a temporary increased sensitivity of dopamine receptors.
16) Barros H, Tannhauser S, Tannhauser M, Tannhauser M. Effect of sodium valproate on the open-field behavior of rats. Braz J Med Biol Res 1992;25(3):281-287. PubMed link
Rats were observed for 14 days following interruption of sodium valproate treatment, with no changes in behavior observed.
17) Carli M, Morissette M, Hébert C, Di Paolo T, Reader T. Effects of a chronic lithium treatment on central dopamine neurotransporters. Biochem Pharmacol 1997;54(3):391-397. PubMed link
The authors found some effects of lithium on dopamine systems, supporting the role of dopamine in affective disorders, but did not find any prolonged effects following 48 hours of lithium withdrawal.
18) Miki M, Hamamura T, Kuroda S, et al. Effects of subchronic lithium chloride treatment on G-protein subunits (Golf, Ggamma7) and adenylyl cyclase expressed specifically in the rat striatum. Eur J Pharmacol 2001;428(3):303-309. PubMed link
The authors found an increase in G-protein after 2 weeks of lithium administration, and that levels did not return to baseline levels until 1 week after withdrawal. The authors discuss the possible relationship between these results and “rebound” mania phenomena after lithium withdrawal.
19) Sattin A, Senanayake S, Pekary A. Lithium modulates expression of TRH receptors and TRH-related peptides in rat brain. Neuroscience 2002;115(1):263-273. PubMed Link
The authors report the observed effects of acute and chronic lithium administration and its withdrawal on Thyrotropin-releasing hormone (TRH) receptors and TRH-related peptides. Lithium administration resulted in varied effects (both increases and decreases) in different brain regions, and opposite effects in the 48-hours of observed withdrawal. Observed withdrawal effects were not equivalent to a return to baseline levels.
20) Pekary A, Sattin A, Meyerhoff J, Chilingar M. Valproate modulates TRH receptor, TRH and TRH-like peptide levels in rat brain. Peptides 2004;25(4):647-658. PubMed link
The authors report on the effects of valproate on TRH levels, finding that TRH levels increased with valproate administration and persisted after two days of withdrawal from treatment. The authors discuss these results in light of theories of this drug’s mood regulating potential.
21) Ferrie L, Young A, McQuade R. Effect of chronic lithium and withdrawal from chronic lithium on presynaptic dopamine function in the rat. J Psychopharmacol 2005;19(3):229-234. PubMed link
The authors found in this study that lithium administration decreased presynaptic dopamine release in rats and that dopamine release levels returned to normal once lithium was withdrawn. The authors conclude that lithium’s effects on dopamine release are not related to rebound mania phenomena.
22) Ferrie L, Young A, McQuade R. Effect of lithium and lithium withdrawal on potassium-evoked dopamine release and tyrosine hydroxylase expression in the rat. Int J Neuropsychopharmacol 2006;9(6):729-735. PubMed link
The authors found that lithium treatment attenuated release of dopamine in treated rats and that this effect persisted 3 days after withdrawal of lithium. Thus, these results do not suggest a “rebound” effect of withdrawal on dopamine release, as has been theorized due to recurrent mania in humans.
The main concern in withdrawal of mood stabilizing drugs is potential “relapse,” principally in the form of manic episodes. Research related to relapse is discussed in the “Discontinuation Success Rate” section below, although some researchers have suggested that the experience of mania after withdrawal from mood stabilizers is a withdrawal reaction rather than relapse. Reports of other withdrawal symptoms are mixed, from no effects to more typical drug- withdrawal symptoms such as anxiety and irritability to kidney-related effects related to lithiumwithdrawal. Some research suggests that mood stabilizers are protective against suicidality, such that withdrawal may increase this risk.
23) Rifkin A, Quitkin F, Howard A, Klein D. A study of abrupt lithium withdrawal. Psychopharmacologia 1975;44(2):157-158. PubMed link
Twelve participants were prescribed lithium for 6 weeks and then abruptly withdrawn to placebo. The authors compared reported side effects in the last week of lithium and the first week on placebo. Reported symptoms were similar, leading the authors to conclude that lithium does not produce withdrawal effects.
24) Rabin E, Garston R, Weir R, Posen G. Persistent nephrogenic diabetes insipidus associated with long-term lithium carbonate treatment. Can Med Assoc J 1979;121(2):194-198. PubMed link
This case report describes persistent urine concentration ability of the kidneys of a woman for 4 years following discontinuation of lithium. The authors conclude that the persistent renal effects were likely related to lithium administration.
25) Christodoulou G N, Lykouras E P. Abrupt lithium discontinuation in manic-depressive patients. Acta Psychiatr Scand 1982:65:310-314. PubMed link
Eighteen patients were discontinued from lithium and followed for 15 days. The authors found reduced side effects and no withdrawal symptoms in their sample, although 3 individuals relapsed within the first four days of discontinuation.
26) King JR, Hullin RP. Withdrawal symptoms from lithium: four case reports and a questionnaire study. Br J Psychiatry 1983;143:30-5. PubMed link
This questionnaire study surveyed lithium users about withdrawal symptoms. Users reported short-term anxiety as a symptom, as well as longer-term effects such as increased emotional responsiveness, improved concentration, and decreased thirst.
27) Bendz H. Kidney function in a selected lithium population. A prospective, controlled, lithium-withdrawal study. Acta Psychiatr Scand 1985;72(5):451-463. PubMed link
By studying withdrawal effect, the author found that long-term use of lithium effects kidney functioning in both reversible and irreversible ways.
28) Goodnick P. Clinical and laboratory effects of discontinuation of lithium prophylaxis. Acta Psychiatr Scand 1985;71(6):608-614. PubMed link
Twelve patients who had been taking lithium for at least a year and were in remission discontinued lithium for three weeks and completed weekly rating scales regarding mood symptoms and side effects. No significant changes in mood symptoms or relapses were found. Side effects decreased after two weeks of discontinuation, particularly renal functioning improvement.
29) Balon R, Yeragani VK, Pohl RB, Gershon S. Lithium discontinuation: withdrawal or relapse? Compr Psychiatry 1988;29:330-4. PubMed link
The authors conclude that little evidence documents “true” withdrawal (i.e., symptoms not attributable to relapse) but that its existence is probable, with symptoms of anxiety, irritability, and disturbed sleep.
30) Duncan J, Shorvon S, Trimble M. Withdrawal symptoms from phenytoin, carbamazepine and sodium valproate. J Neurol Neurosurg Psychiatry 1988;51(7):924-928. PubMed link
This withdrawal study was conducted in order to determine withdrawal symptoms from these medications in the treatment of seizures and seizure disorders. Faster and slower rates of withdrawal were compared to a control group that maintained the medications. No significant differences in symptoms were found between groups.
31) Souza F, Mander A, Foggo M, Dick H, Shearing C, Goodwin G. The effects of lithium discontinuation and the non-effect of oral inositol upon thyroid hormones and cortisol in patients with bipolar affective disorder. J Affect Disord July 1991;22(3):165-170. PubMed link
The authors monitored hormone levels in 14 individuals who were withdrawn from lithium treatment. Significant changes in hormones were found among the participants, which the authors related to the research supporting decreased thyroid functioning during lithium treatment. Seven of the participants relapsed following withdrawal, and relapse was not associated with changes in hormones.
32) Suppes, T, Baldessarini, RJ, Fredda, GL, Tohen, M. Risk of recurrence following discontinuation of lithium treatment in bipolar disorder. Arch Gen Psychiatry 1991;48:1082–1088. PubMed link
This article reviews research on the risks to users following discontinuation from lithium, first and foremost the risk of relapse. The possibility of a risk of users becoming refractory to lithium after discontinuing and then reintroducing it, as well as increased suicide risk, are discussed in terms of a being areas of concern that require further research.
33) Post R, Leverich G, Altshuler L, Mikalauskas K. Lithium-discontinuation-induced refractoriness: preliminary observations. Am J Psychiatry 1992;149(12):1727-1729. PubMed link
The authors present a case study of four individuals diagnosed with bipolar disorder who experienced relapses following discontinuation of long-term treatment, followed by ineffectiveness of the treatment once reinstated. The authors suggest that this refractoriness may be a withdrawal effect.
34) Schou M. Is there a lithium withdrawal syndrome? An examination of the evidence. Br J Psychiatry 1993;163:514-518. PubMed link
After reviewing research evidence, the author concludes that little quality evidence of a withdrawal syndrome for lithium had been produced to date. In the review, Schou explores other interpretations of reported symptoms and methodological weaknesses of studies that have led to inconclusive results.
35) Ketter T, Malow B, Flamini R, White S, Post R, Theodore W. Anticonvulsant withdrawal-emergent psychopathology. Neurology 1994;44(1):55-61. PubMed link
The authors studied psychopathology symptoms that occurred during withdrawal of anticonvulsant (carbamazepine, valproic acid, or phenytoin) drugs used to treat seizures among 32 participants. Tapering ranged between 5 and 45 days. The authors found increases in reported symptoms in the final week of tapering, followed by dramatic increases once the drugs were discontinued. Moderate to severe pathology was documented in 12 participants, including 2 with psychotic symptoms. The authors conclude that this symptomatology may have been due in part to withdrawal effects.
36) Swartz C, Dolinar L. Encephalopathy associated with rapid decrease of high levels of lithium. Ann Clin Psychiatry 1995;7(4):207-209. PubMed link
This case study documents neurotoxicity following rapid withdrawal from high doses of lithium, which the authors distinguish from toxicity resulting from high doses of lithium alone.
37) Bendz H, Sjödin I, Aurell M. Renal function on and off lithium in patients treated with lithium for 15 years or more. A controlled, prospective lithium-withdrawal study. Nephrol Dial Transplant 1996;11(3):457-460. PubMed link
Results supported decreased kidney functioning in long-term lithium patients that persisted for 9 weeks after withdrawal.
38) Darbar D, Connachie A, Jones A, Newton R. Acute psychosis associated with abrupt withdrawal of carbamazepine following intoxication. Br J Clin Pract 1996;50(6):350-351. PubMed link
This case study discusses the incidence of psychotic symptoms, including agitation and paranoid delusions, after withdrawal from a toxic dose of carbamazepine in a patient with no history of psychosis.
39) Tondo L, Baldessarini R, Floris G, Rudas N. Effectiveness of restarting lithium treatment after its discontinuation in bipolar I and bipolar II disorders. Am J Psychiatry 1997;154(4):548-550. PubMed link
The authors studied the effects of withdrawing and restarting lithium treatment, finding no evidence to support any “refractoriness” after interruption of treatment.
40) Tondo L, Jamison K, Baldessarini R. Effect of lithium maintenance on suicidal behavior in major mood disorders. Ann N Y Acad Sci 1997;836:339-351. PubMed link
In this review, the authors discuss the evidence of an increased risk of suicidality following discontinuation of lithium.
41) Tondo, L, Baldessarini RJ, Hennen J, et al.: Lithium treatment and risk of suicidal behavior in bipolar disorder patients. J Clin Psychiatry 1998, 59:405–414. PubMed link
This article found an increased risk of suicidality in the year following discontinuation of lithium when compared to those who maintained treatment.
42) Baldessarini R, Tondo L, Hennen J. Effects of lithium treatment and its discontinuation on suicidal behavior in bipolar manic-depressive disorders. J Clin Psychiatry 1999;60 Suppl 2:77-84. PubMed link
From their review, the authors conclude that lithium discontinuation, and particularly abrupt discontinuation, is associated with increased risk of suicidal ideation and death by suicide.
43) Bowden C. The ability of lithium and other mood stabilizers to decrease suicide risk and prevent relapse. Curr Psychiatry Rep 2000;2(6):490-494. PubMed link
In this review article, the author discusses evidence of suicide risk reduction via use of lithium, divalproex, and carbamazepine. Evidence for lithium’s effects were mostly drawn from naturalistic studies and suggest that length of lithium use may be a factor in the lower rates of suicidality found. The author also discusses two trials that compared lithium to other mood stabilizers, and that it remains unclear the extent to which any medication reduces suicide risk versus other psychosocial interventions delivered in the course of medication management.
44) Faedda G, Tondo L, Baldessarini R. Lithium discontinuation: uncovering latent bipolar disorder? Am J Psychiatry 2001;158(8):1337-1339. PubMed link
In this letter to the editor, the authors comment on a recent study of discontinuation of adjunctive lithium treatment among individuals with unipolar depression. The authors point out that the incidence of manic episodes after lithium withdrawal, and thus rediagnosis to bipolar disorder, is greater than would be expected than would be expected statistically. These results suggest that manic episodes experienced after discontinuation are withdrawal-related rather than relapse.
45) Gelisse P, Kissani N, Crespel A, Jafari H, Baldy-Moulinier M. Is there a lamotrigine withdrawal syndrome? Acta Neurol Scand 2002;105(3):232-234. PubMed link
This case reports describes the development of psychomotor inhibition in a patient withdrawn from lamotrigine abruptly. The authors discuss the possibility of a withdrawal syndrome, with typically minor and less typically severe reactions.
46) Carmaciu C, Anderson C, Lawton C. Thyrotoxicosis after complete or partial lithium withdrawal in two patients with bipolar affective disorder. Bipolar Disord 2003;5(5):381-384. PubMed link
In this case study the authors describe the emergence of thyrotoxicosis in two patients following withdrawal from lithium, one having been withdrawn fully and the other partially. The authors discuss possible explanations for the relationship between withdrawal and this condition and the need for further research.
47) Yerevanian B, Koek R, Mintz J. Bipolar pharmacotherapy and suicidal behavior. Part I: Lithium, divalproex and carbamazepine. J Affect Disord 2007;103(1-3):5-11. PubMed link
The authors compared individuals maintained on lithium, divalproex, and carbamazepine to those withdrawn in rates of suicidality. For all three medications, rates of suicidality were higher after withdrawal of the medication than with treatment.
48) Frey L, Strom L, Shrestha A, Spitz M. End-of-dose emergent psychopathology in ambulatory patients with epilepsy on stable-dose lamotrigine monotherapy: a case series of six patients. Epilepsy Behav 2009;15(4):521-523. PubMed link
The authors identified six individuals via retrospective chart review who experienced distressing psychiatric symptoms during late-dose withdrawal from lamotrigine. The principal symptoms reported were anxiety and irritability.
49) Grünfeld J, Rossier B. Lithium nephrotoxicity revisited. Nat Rev Nephrol 2009;5(5):270-276. PubMed link
This article reviews literature on the effects of lithium use on renal functioning. The authors discuss the costs and benefits of discontinuing lithium treatment, especially given that discontinuation can sometimes but not always improve renal functioning.
50) Howland R. Potential adverse effects of discontinuing psychotropic drugs. Part 3: Antipsychotic, dopaminergic, and mood-stabilizing drugs. J Psychosoc Nurs 2010;48(8):11-14. PubMed link
Howland reviews research on potential withdrawal symptoms of mood instability and risk of mood episode relapse, with the highest risk associated with more sudden discontinuation. The anticonvulsants have different side effects, but generally discontinuation is also associated with mood instability, as well as anxiety, agitation, and sleep disturbance. These drugs, and thus their discontinuation, also affect metabolism of other medications.
51) Werneke U, Ott M, Renberg E, Taylor D, Stegmayr B. A decision analysis of long-term lithium treatment and the risk of renal failure. Acta Psychiatr Scand 2012;126(3):186-197. PubMed link
The authors discuss relative risks and benefits of lithium continuation and discontinuation in response to kidney disease. The authors conclude that lithium continuation is still recommended in most cases given risk of relapse and suicide upon discontinuation.
52) Chen M, Zhang W, Guo Z, Zhang W, Chai Y, Li Y. Withdrawal reaction of carbamazepine after neurovascular decompression for trigeminal neuralgia: a preliminary study. J Neurol Sci 2014;338(1-2):43-45. PubMed link
Ninety patients were followed after carbamazepine withdrawal; 26 patients reported withdrawal symptoms within 4 days of withdrawal. Symptoms included insomnia, dysphoria, hallucination, hand fremitus, and headaches, and symptoms alleviated within 1 week.
Discontinuation Success Rates
The incidence of withdrawal-related relapses is fairly well-established for lithium, although less is known about discontinuation of other drugs in this class. Differentiation of symptoms that result from drug withdrawal from a recurrence of illness is not systematic in the research literature, making it difficult to assess withdrawal (see Peter Breggin’s book Psychiatric Drug Withdrawal: A Guide for Prescribers, Therapists, Patients, and Their Families and Guy Chouinard’s article “Issues in the clinical use of benzodiazepines: potency, withdrawal and rebound” in the Journal of Clinical Psychiatry for further discussion of this issue). Dropouts from clinical studies also complicate this research. Shorter term studies favor drug-treated groups over those taking placebo in terms of risk of relapse, but longer term studies suggest that the risk of relapse is comparable for those who maintain drug treatment and those who withdraw. In addition, the time spent withdrawing the drug treatment may affect the risk of relapse, with abrupt withdrawal increasing that risk. (See “tapering speed” section below.”)
53) Baastrup P, Poulsen J, Schou M, Thomsen K, Amdisen A. Prophylactic lithium: double blind discontinuation in manic-depressive and recurrent-depressive disorders. Lancet 1970;2(7668):326-330. PubMed link
The authors found that 21 of 39 patients discontinued from lithium relapsed during the trial, whereas none of the patients who continued taking lithium relapsed. The authors reported that the relapses were distributed across the period of the trial (5 months) and did not indicate “rebound” effects.
54) Small JC, Small IF, Moore DF. Experimental withdrawal of lithium in recovered manic-depressive patients. Am J Psychiatry 1971:127:1555-1558. PubMed link
In this case study five patients were withdrawn from lithium; the authors found that 4 of the 5 patients relapsed within seven weeks of withdrawing from the drug.
55) Lapierre Y D, Gagnon A, Kokkinidis L. Rapid recurrence of mania following lithium withdrawal. Biol Psychiatry 1980:15:859-864. PubMed link
In this study of 20 patients who were withdrawn from lithium after a long period of mood stability, 4 were found to relapse within one week. The authors suggest that the manic relapses were rebound phenomena, reflecting a reaction to lithium withdrawal.
56) Klein H, Broucek B, Greil W. Lithium withdrawal triggers psychotic states. Br J Psychiatry 1981;139:255-256. PubMed link
In this study of lithium withdrawal of 21 patients, 11 patients relapsed within two weeks of discontinuation. Among the other 10 participants, discontinuation symptoms of anxiety, irritability, disturbed sleep, and some mood lability were reported.
57) Margo A, McMahon P. Lithium withdrawal triggers psychosis. Br J Psychiatry 1982:141:407-410. PubMed link
The authors of this case study of lithium withdrawal concluded that all 4 of 4 patients relapsed with manic episodes within 2 weeks of discontinuation.
58) Christodoulou G N, Lykouras E P. Abrupt lithium discontinuation in manic-depressive patients. Acta Psychiatr Scand 1982:65:310-314. PubMed link
Eighteen patients were discontinued from lithium and followed for 15 days. The authors found reduced side effects and no withdrawal symptoms in their sample, although 3 individuals relapsed within the first four days of discontinuation.
59) Sashidharan S, McGuire R. Recurrence of affective illness after withdrawal of long-term lithium treatment. Acta Psychiatr Scand 1983;68(2):126-133. PubMed link
The authors followed 22 patients who had discontinued lithium in this observational study and found that 16 experienced mood episode recurrence within 67 months of discontinuation. The observed that manic episodes seemed to occur closer to discontinuation than depressive episodes, and 4 of the 16 experienced a mood episode within 3 months of discontinuation.
60) Mander A. Is there a lithium withdrawal syndrome? Br J Psychiatry 1986;149:498-501. PubMed link
The author compared a control group of participants who had never taken lithium, or had taken it for less than three months, to those who were treated with lithium. All participants had been discharged from care and had been stable for three months afterward. Mander found that the risk of recurrence was higher for those who had taken lithium and discontinued use than in those who had never taken lithium, concluding that these results support the existence of a lithium withdrawal syndrome. Most relapses within the first three months of discontinuation were manic rather than depressive episodes.
61) Heh C, Sramek J, Herrera J, Costa J. Exacerbation of psychosis after discontinuation of carbamazepine treatment. Am J Psychiatry 1988;145(7):878-879. PubMed link
Twenty individuals diagnosed with schizophrenia were discontinued abruptly, with an exacerbation of psychotic symptoms for two individuals. The authors review several hypotheses that could account for the increase in symptoms following discontinuation.
62) Suppes T, Baldessarini R, Faedda G, Tohen M. Risk of recurrence following discontinuation of lithium treatment in bipolar disorder. Arch Gen Psychiatry 1991;48(12):1082-1088. PubMed link
The authors analyzed risk of recurrence from 14 studies in which participants were discontinued from lithium. They found that 50% of mood episodes that occurred following discontinuation occurred within the first 10 weeks after treatment was stopped.
63) Suppes T, Baldessarini R, Faedda G, Tondo L, Tohen M. Discontinuation of maintenance treatment in bipolar disorder: risks and implications. Harv Rev Psychiatry 1993;1(3):131-144. PubMed link
This review article discusses recurrence risk following discontinuation, as well as other potential risks such as treatment refractoriness and suicidality.
64) Scull D, Trimble MR. Mania precipitated by carbamazepine withdrawal. Br J Psychiatry 1995; 167:698. PubMed link
This case study of a woman being treated for epilepsy with carbamazepine describes the appearance of symptoms of mania following withdrawal, suggesting a possible relationship between withdrawal and “rebound” mania.
65) Johnson R, McFarland B. Lithium use and discontinuation in a health maintenance organization. Am J Psychiatry 1996;153(8):993-1000. PubMed link
The authors compared rate of mental health service use among patients who maintained lithium treatment and those who discontinued it, finding that those who discontinued had higher rates of psychiatric hospitalization and emergency services.
66) Coryell W, Winokur G, Solomon D, Shea T, Leon A, Keller M. Lithium and recurrence in a long-term follow-up of bipolar affective disorder. Psychol Med 1997;27(2):281-289. PubMed link
In this study, patients with bipolar disorder were followed for 5 years, with one group of patients continuing lithium prophylaxis and another group discontinuing and patients taking lithium were compared to those who were not. The authors suggest that lithium prophylaxis may be helpful in preventing relapse but not recurrence, as there was no significant difference between the two groups in rates of recurrence in the long term.
67) Kennebäck G, Ericson M, Tomson T, Bergfeldt L. Changes in arrhythmia profile and heart rate variability during abrupt withdrawal of antiepileptic drugs. Implications for sudden death. Seizure 1997;6(5):369-375. PubMed link
The authors studied cardiac symptoms in ten patients following abrupt withdrawal from carbamazepine and phenytoin in the last day of treatment and four days following withdrawal. The authors conclude that the cardiac symptoms associated with abrupt withdrawal may contribute to sudden unexpected deaths among patients with epilepsy.
68) Baldessarini R, Tondo L. Recurrence risk in bipolar manic-depressive disorders after discontinuing lithium maintenance treatment: an overview. Clin Drug Investig 1998;15(4):337-351. PubMed link
In this review, the authors conclude that discontinuation carries a risk of recurrence of mood symptoms, particularly within the first year of discontinuation, but that gradual discontinuation attenuates this risk. The authors also conclude that individuals risk only minor loss of effectiveness of lithium if it is reintroduced after discontinuation.
69) Baldessarini R, Tondo L, Viguera A. Discontinuing lithium maintenance treatment in bipolar disorders: risks and implications. Bipolar Disord 1999;1(1):17-24. PubMed link
This review of the research evidence highlights the increased risk of recurrence and suicidality after lithium discontinuation, particularly abrupt discontinuation, and the authors’ hypothesis that this phenomenon reflects a reaction the body’s adaptations to long-term treatment. The authors caution interpretation of discontinuation studies and any conclusion that the heightened recurrence risk resulting from discontinuation in placebo arms is comparable to non-treatment.
70) Davis J, Janicak P, Hogan D. Mood stabilizers in the prevention of recurrent affective disorders: a meta-analysis. Acta Psychiatr Scand 1999;100(6):406-417. PubMed link
In this meta-analysis, Davis et al. conclude that maintenance lithium reduces relapses when compared to no treatment, and they present their argument against evidence of lithium withdrawal-related relapse.
71) Bowden C. The ability of lithium and other mood stabilizers to decrease suicide risk and prevent relapse. Curr Psychiatry Rep 2000;2(6):490-494. PubMed link
The author discusses evidence of reduced risk of relapse when patients are maintained with lithium or divalproex when compared to placebo, while also discussed the limitations of the designs of earlier studies from the 1970s that overinflated the apparent risk of relapse upon discontinuation.
72) Calabrese J, Suppes T, Monaghan E, et al. A double-blind, placebo-controlled, prophylaxis study of lamotrigine in rapid-cycling bipolar disorder. Lamictal 614 Study Group. J Clin Psychiatry 2000;61(11):841-850. PubMed link
This withdrawal study found that 23 of 89 participants who withdrew to placebo had not relapsed after the 26-week study period. Forty-nine required some intervention, and the remaining participants withdrew before the end of the study. The authors did not find a significant difference in the time to additional intervention between those withdrawn to placebo and those maintained on lamotrigine.
73) Macritchie K, Hunt N. Does ‘rebound mania’ occur after stopping carbamazepine? A pilot study. J Psychopharmacol 2000;14(3):266-268. PubMed link
This pilot study followed 6 individuals who had withdrawn carbamazepine treatment and did not support a “rebound” effect as has been reported with lithium discontinuation.
74) Bowden C, Calabrese J, DeVeaugh-Geiss J, et al. A placebo-controlled 18-month trial of lamotrigine and lithium maintenance treatment in recently manic or hypomanic patients with bipolar I disorder. Arch Gen Psychiatry 2003;60(4):392-400. PubMed link
This study compared patients who were all treated with lamotrigine before being maintained on lamotrigine, lithium, or placebo. Participants in the two maintenance treatment groups had significantly longer time to a subsequent mood episode than those who had discontinued.
75) Calabrese J, Bowden C, DeVeaugh-Geiss J, et al. A placebo-controlled 18-month trial of lamotrigine and lithium maintenance treatment in recently depressed patients with bipolar I disorder. J Clin Psychiatry 2003;64(9):1013-1024. PubMed link
Participants in this study were stabilized on lamotrigine in an open-label phase and then maintained on lamotrigine, lithium, or placebo. For those participants stabilized on lamotrigine and then withdrawn to placebo, the median time to treatment was significantly shorter than for those maintained on lithium or lamotrigine. Adverse events reported by placebo participants were headache, nausea, dizziness, tremor, rash, somnolence, diarrhea, insomnia, and infection or influenza.
76) Jess G, Smith D, Mackenzie C, Crawford C. Carbamazepine and rebound mania. Am J Psychiatry 2004;161(11):2132-2133. PubMed link
The authors describe a case study of patient with no history of mood disorders who experienced manic symptoms after withdrawing from carbamazepine.
77) Biel M, Peselow E, Mulcare L, Case B, Fieve R. Continuation versus discontinuation of lithium in recurrent bipolar illness: a naturalistic study. Bipolar Disord 2007;9(5):435-442. PubMed link
This 2-year study followed 159 individuals who maintained treatment and 54 patients who discontinued lithium. Illness recurrence occurred in both groups, with higher odds of recurrence among the discontinued group.
78) Viguera A, Whitfield T, Cohen L, et al. Risk of recurrence in women with bipolar disorder during pregnancy: prospective study of mood stabilizer discontinuation. Am J Psychiatry 2007;164(12):1817-1824. PubMed link
Pregnant female participants who had discontinued mood stabilizer treatment were compared to women who maintained drug treatment up to a year postpartum. Most participants were also taking adjunctive antidepressants. The authors reported higher risk of recurrence of bipolar symptoms among those who had discontinued medication, as well as longer duration of episodes. Postpartum findings were not reported.
79) Franks M, Macritchie K, Mahmood T, Young A. Bouncing back: is the bipolar rebound phenomenon peculiar to lithium? A retrospective naturalistic study. J Psychopharmacol 2008;22(4):452-456. PubMed link
This study found that the “rebound” of bipolar disorder symptoms after discontinuation of medication occurred in the majority (74%) of cases reviewed and included those discontinuing all medications in this class. Rates of relapse were highest among those who had withdrawn from lithium and anticonvulsants versus those discontinuing antidepressants or antipsychotics.
80) Moncrieff J. The myth of the chemical cure: a critique of psychiatric drug treatment. 2009; New York: Palgrave Macmillan. Publisher link
Moncrieff critically reviews clinical trials on prophylactic treatment with lithium and other mood stabilizing drugs, noting the many methodological issues in this research that make it difficult to determine if long-term treatment does in fact reduce the risk of mania and/or depression in comparison to no treatment.
81) Sharma P, Kongasseri S, Praharaj S. Outcome of mood stabilizer discontinuation in bipolar disorder after 5 years of euthymia. J Clin Psychopharmacol 2014;34(4):504-507. PubMed link
This study followed 23 individuals withdrawing from lithium, valproate, or carbamazepine after at least 5 years of prophylactic treatment. Discontinuation was planned over 3 to 12 months. Twenty individuals had a recurrent manic episode following discontinuation; the time to the recurrent episode was a median of 10 months.
82) Simhandl C, König B, Amann B. A prospective 4-year naturalistic follow-up of treatment and outcome of 300 bipolar I and II patients. J Clin Psychiatry 2014;75(3):254-262. PubMed link
The authors followed 300 patients over 4 years and found that stopping medication, either by the patient or the prescribing psychiatrist, decreased the time to relapse in comparison to those who maintained medication, including those who reduced their dosage. No differences in risk to relapse were found in relation to various demographic and clinical variables or bipolar I versus bipolar II diagnoses.
Most researchers agree that gradual tapering (2 weeks or longer) of lithium is safer than abrupt tapering (less than two weeks). This conclusion has been broadened to other mood stabilizers, although less research supports the superiority of a particular tapering speed among the anticonvulsant drugs.
83) Baldessarini R, Tondo L, Faedda G, Suppes T, Floris G, Rudas N. Effects of the rate of discontinuing lithium maintenance treatment in bipolar disorders. J Clin Psychiatry 1996;57(10):441-448. PubMed link
In this trial comparing of rapid (1-14 days) and gradual (15-30 days) discontinuation of lithium, the time to recurrence was 5 times faster with rapid discontinuation and specifically in the first year after discontinuation. Participants were 20 times more likely to be stable 3 years post-discontinuation following a gradual versus rapid taper.
84) Darbar D, Connachie A, Jones A, Newton R. Acute psychosis associated with abrupt withdrawal of carbamazepine following intoxication. Br J Clin Pract 1996;50(6):350-351. PubMed link
Due to the incidence of psychotic symptoms following withdrawal from a toxic dose of carbamazepine, the authors suggested slow tapering of carbamazepine treatment given the risk of these symptoms after abrupt withdrawal.
85) Faedda G, Tondo L, Baldessarini R, Suppes T, Tohen M. Outcome after rapid vs gradual discontinuation of lithium treatment in bipolar disorders. Arch Gen Psychiatry 1993;50(6):448-455. PubMed link
In comparing gradual versus rapid discontinuation, the authors found that risk of recurrence was higher among those who had discontinued rapidly, and that this risk was highest within the first year of discontinuation. Beyond the first year, recurrence rates were not significantly different between groups.
86) Baldessarini R, Tondo L, Floris G, Rudas N. Reduced morbidity after gradual discontinuation of lithium treatment for bipolar I and II disorders: a replication study. Am J Psychiatry 1997;154(4):551-553. PubMed link
This replication study (Faedda et al., 1993) included a 2-year follow-up of participants withdrawn rapidly (less than 2 weeks) or gradually (2+ weeks) from lithium. Slower discontinuation was associated with increased latency to first relapse; in addition, significantly more of the gradual taper group had remained stable at the 2-year follow-up than the rapid taper group. Thus, the authors conclude that gradual discontinuation could reduce recurrent episodes as well as delaying them.
87) Baldessarini R, Tondo L. Recurrence risk in bipolar manic-depressive disorders after discontinuing lithium maintenance treatment: an overview. Clin Drug Investig 1998;15(4):337-351. PubMed link
In this review of studies on lithium discontinuation, the authors provide support for their hypothesis that gradual discontinuation (over 2-4 weeks) reduces risk of relapse versus more abrupt discontinuation (2 weeks or less). The authors propose that the brain’s need to adapt gradually to the withdrawal of lithium accounts for this reduced risk.
88) Baldessarini R, Tondo L, Viguera A. Discontinuing lithium maintenance treatment in bipolar disorders: risks and implications. Bipolar Disord 1999;1(1):17-24. PubMed link
In this study, the authors reviewed research on clinical effects of lithium discontinuation, concluding that a) discontinuation increases suicide risk and risk of relapse and b) gradual tapering is associated with less risk.
89) Gelisse P, Kissani N, Crespel A, Jafari H, Baldy-Moulinier M. Is there a lamotrigine withdrawal syndrome? Acta Neurol Scand 2002;105(3):232-234. PubMed link
This case reports describes the development of psychomotor inhibition in a patient withdrawn from lamotrigine abruptly. The authors recommend withdrawal of lamotrigine over two weeks whenever possible to reduce risk.
90) Perlis R, Sachs G, Rosenbaum J, et al. Effect of abrupt change from standard to low serum levels of lithium: a reanalysis of double-blind lithium maintenance data. Am J Psychiatry 2002;159(7):1155-1159. PubMed link
Ninety-four patients were followed for up to 182 weeks after maintaining their dose of lithium or dose reduction. The authors conclude that the abrupt change in lithium serum levels is a stronger predictor of illness recurrence than dosage, and thus that abrupt dose changes should be avoided.
91) Cavanagh J, Smyth R, Goodwin G. Relapse into mania or depression following lithium discontinuation: a 7-year follow-up. Acta Psychiatr Scand 2004;109(2):91-95. PubMed link
Through long-term follow-up after discontinuation of lithium, the results of this study support that the risk of relapse is highest immediately after acute discontinuation. The authors did not find evidence of significantly increased risk of relapse after the increased period of risk immediately after discontinuation.