Stopping lithium and mood stabilisers

Rationale for stopping

Patients may request to stop lithium and other mood stabilisers because of the range of
adverse effects: in one cohort 54% of patients discontinued lithium, 62% of these because
of adverse effects: including diarrhoea (13%), tremor (11%), polyuria/polydipsia/diabetes
insipidus (9%), creatinine increase (9%) and weight gain (7%). 1 Alternatively, although
lithium and mood stabilisers can be useful in controlling acute symptoms and preventing
relapse, a clinician may judge that the balance of risks and benefits have shifted over time
(e.g. adverse physical effects accumulate, alternative coping mechanisms developed) so that
dose reduction or stopping may be beneficial. Other patients may be prescribed mood
stabilisers for conditions, such as personality disorders, for which there is a lack of evidence.
Stopping should be done in a manner that minimises the risk of withdrawal effects and
relapse.

Withdrawal effects from lithium and other mood stabilisers

Discontinuation of lithium can cause withdrawal effects, including both physical and
psychological symptoms (see Table 1). These withdrawal effects include mood episodes
(depression, but more commonly, mania) and are sometimes called ‘rebound’ effects.2,3
These conclusions have been drawn because the risk of relapse in the period following
abrupt cessation greatly exceeds the rate of relapse in the untreated disorder.2

For example, a review of 10 studies of lithium discontinuation in people with bipolar

disorder found that the untreated disorder had a mean cycle length (the average time
between episodes) of 11.6 months, whereas the time to a new episode following lithium
discontinuation was 1.7 months.2 This represents a 7-fold increase in the rate of relapse and
suggests that manic and depressive symptoms that occur following lithium withdrawal are
largely due to the physiological effects of lithium withdrawal rather than to the underlying
disorder.

These effects have been variously thought to be due to the development of dopaminergic
hypersensitivity, changes in neuronal membranes, cell transport function or other
neurotransmitter systems during lithium treatment.4 Other mood stabilisers have also been
associated with a withdrawal syndrome.5

Physical Psychological
▪ Tremor ▪ Anxiety
▪ Polyuria ▪ Nervousness
▪ Muscular weakness ▪ Irritability
▪ Polydipsia ▪ Alertness
▪ Dryness of mouth ▪ Sleep disturbances
▪ Elated mood/mania
▪ Depressed mood

Table 1 – Withdrawal effects from lithium3,6,7

Uncertainty in the evidence for long-term treatment

Most evidence for long-term treatment with lithium and other mood stabilisers is derived
from discontinuation studies where patients established on these medications are
randomised to either continue or cease these medications with relapse rates measured for
each group.8,9 In these studies lithium is often stopped abruptly. Abrupt stoppage of lithium
is likely to produce withdrawal effects, which can include precipitating mood episodes.2
Indeed, abruptly stopping lithium in patients with unipolar depression (and no bipolarity)
produced manic episodes in 13%.10 There is evidence that abrupt stoppage of other mood
stabilisers can also precipitate mood episodes.3 Patients who are ceased from these
medications rapidly demonstrate relapse rates that are far greater than the untreated
disorder, suggesting that withdrawal effects inflate the apparent rates of relapse.2 Few
maintenance studies extend beyond a two-year follow up period. These considerations
should temper justification for long-term lithium treatment.

Rationale for tapering

The observation that abrupt tapering can lead to withdrawal-related disturbances in mood
suggests that slower tapering may avoid some of these effects. 11 There is some evidence for
this: observational trials have found that in one year follow-up patients who tapered lithium
over greater than 15 days have a relapse rate one third that of those who rapidly
discontinue in less than 2 weeks.11 More strikingly, those who gradually stopped were
almost 5 times more likely to remain stable over the following three years (47% versus
10%).11

Hyperbolic tapering

Lithium, like all pharmacological agents, conforms to the law of mass action and therefore
demonstrates a hyperbolic pattern between dose and pharmacological effect.12 The mode
of action of lithium is unknown, however, it is known to affect GSK-3. The relationship
between dose of lithium and effect on this target is hyperbolic (Figure 1).13 Even if this is not
the target responsible for its mood stabilisation effects, all dose-receptor occupancy curves
are hyperbolic. As for other psychotropic agents this justifies a hyperbolically reducing dose
pattern (in order to produce linearly reducing effects on its target receptors (Figure1b)),
which may be clinically approximated by an exponential dose reduction (a reduction by the
same proportion each step, so that the size of the reduction becomes smaller and smaller as
the total dose gets lower).
A B

Figure 1 A) The relationship between lithium concentration and proportion of GSK- phosphorylation is
hyperbolic.13 Although the relationship between in vitro concentration and oral dosage is hard to establish it can
be seen that linear reductions of lithium cause increasingly large reductions in effect.
B) To produce linear reduction in effect (in this case GSK- phosphorylation) a hyperbolically decreasing regimen
of lithium is required. Although the relationship between in vitro concentration and oral dosage is hard to
establish it can be seen that final doses of lithium required before complete cessation will need to be very small
to avoid too large a decrease in effect.

Duration of tapering

There has been a lack of studies examining the optimal rate of tapering lithium. However,
the finding that 50% of relapses occur in the first 3 months after lithium is stopped quickly,
but then lessens over time2 suggests that this period might be required for underlying
adaptations to lithium to resolve and suggests that tapering over months may be beneficial.
One study which discontinued lithium over 2-5 months found higher relapse rates in those
patients compared to those who stayed on lithium14 – this might suggest that tapering
should be even slower than the 4 week to 3 month period suggested by NICE.15

Anti-epileptic drugs (AEDs) are tapered over between one month and four years in non-
psychiatric conditions, with relapse rates increased in the first 6 months, before converging
with patients continuing with AEDs.5

Practice guide to tapering

▪ Patients should be told that there is the possibility of withdrawal effects, and that
there may be an increased risk of affective relapse from stopping lithium or mood
stabilisers abruptly. These effects may be reduced if these medications are ceased in
a more gradual fashion.

▪ There is no clear evidence on how to taper, but following principles from other
psychotropic medications, an initial reduction of 10-25% of the current dose should
be offered, with withdrawal symptoms (Table 1) and affective symptoms monitored
for at least 4 weeks to ensure stability.
 ▪ Further reductions should be titrated to the tolerability of this reduction. Reductions
should probably be made according to an exponentially reducing pattern, whereby
each reduction is calculated on the most recent dose (becoming smaller and smaller
as the total dose becomes lower). An example reduction regimen for lithium is given
in Table 2. Reduction regimens for other mood stabilisers should follow a similar
pattern.

▪ The final dose before completely stopping should probably be very small, because
small doses still have large effects on target receptors. Similar to other psychotropic
classes, this may be as small as 1/50th therapeutic doses e.g. 20mg for lithium. To
achieve small doses pill cutters, liquid preparations or tapering strips may be
required.

▪ As the process of reducing lithium or mood stabilisers might be de-stabilising it may

be wise to pursue other means of coping during the tapering period.16 Ongoing
monitoring may be necessary for a number of months after complete cessation to
ensure stability.

▪ If withdrawal symptoms of symptoms of relapse emerge at any point pausing the

reduction, a small increase in dose or returning to a previously effective dose are all
possible responses. Difficulty reducing medication does not preclude a further
attempt at reduction, but might indicate the need for a more gradual reduction
regimen.

▪ If medication is reduced or ceased, other modalities can be effective for people with
bipolar disorder, including family therapy, interpersonal therapy, CBT,
psychoeducation, social rhythm therapy, as well as more individualised, idiosyncratic
coping strategies.17–19

Period* Lithium dose (mg) Period* Lithium dose (mg)

1 1000 11 110
2 800 12 85
3 640 13 70
4 510 14 55
5 410 15 45
6 330 16 35
7 260 17 30
8 210 18 25
9 170 19 20
10 135 20 0

Table 2 Suggested reduction regimen for lithium, based on 10% reductions per period,
based on the most recent dose, rounded to the nearest 5mg. Final dose before complete
cessation is an approximation. Currently available formulations may necessitate
approximations of this regimen. *Period – approximately four weeks, or as long as required
for stability to be assured.
References
1 Öhlund L, Ott M, Oja S, et al. Erratum: Reasons for lithium discontinuation in men and women with bipolar disorder: A
retrospective cohort study (BMC Psychiatry (2018) 18:37 DOI: 10.1186/s12888-018-1622-1). BMC Psychiatry 2018; 18: 1–10.
2 Suppes T, Baldessarini R, Faedda G, Tohen M. Risk of Recurrence Following Discontinuation of Lithium Treatment in Bipolar
Disorder. Arch Gen Psychiatry 1991; 48: 1082–8.
3 Franks MA, Macritchie KAN, Mahmood T, Young AH. Bouncing back: Is the bipolar rebound phenomenon peculiar to lithium? A
retrospective naturalistic study. J Psychopharmacol 2008; 22: 452–6.
4 Balon R, Yeragani VK, Pohl RB, Gershon S. Lithium discontinuation: Withdrawal or relapse? Compr Psychiatry 1988; 29: 330–4.
5 Vernachio K, Lovett AW, Williams J. A Review of Withdraw Strategies for Discontinuing Antiepileptic Therapy in Epilepsy and Pain
Management. Pharm Pharmacol Int J 2015; 3: 232–5.
6 Baastrup PC, Poulsen JC, Schou M, Thomsen K, Amdisen A. Prophylactic Lithium: Double Blind Discontinuation in Manic-
Depressive and Recurrent-Depressive Disorders. Lancet 1970; 296: 326–30.
7 Klein E, Mairaz R, Pascal M, Hefez A, Lavie P. Discontinuation of Lithium Treatment in Remitted Bipolar Patients. J Nerv Ment Dis
1991; 179: 499–501.
8 Geddes JR, Burgess S, Hawton K, Jamison K, Goodwin GM. Long-Term Lithium Therapy for Bipolar Disorder: Systematic Review
and Meta-Analysis of Randomized Controlled Trials. Am J Psychiatry 2004; 161: 217–22.
9 Moncrieff J. Evidence on the Action of Lithium and ‘Mood Stabilisers’. In: The Myth of the Chemical Cure. Palgrave Macmillan,
London., 2008. DOI:https://doi.org/10.1007/978-0-230-58944-5_12.
10 Faedda GL, Tondo L, Baldessarini RJ. Lithium Discontinuation: Uncovering Latent Bipolar Disorder? Am J Psychiatry 2001; 158:
1337-a-1338.
11 Baldessarini RJ, Tondo L, Viguera AC. Discontinuing lithium maintenance treatment in bipolar disorders: risks and implications.
Bipolar Disord 1999; 1: 17–24.
12 Holford N. Pharmacodynamic principles and the time course of delayed and cumulative drug effects. Transl Clin Pharmacol 2018;
26: 56.
13 Rao AS, Kremenevskaja N, Resch J, Brabant G. Lithium stimulates proliferation in cultured thyrocytes by activating Wnt/β-catenin
signalling. Eur J Endocrinol 2005; 153: 929–38.
14 Biel MG, Peselow E, Mulcare L, Case BG, Fieve R. Continuation versus discontinuation of lithium in recurrent bipolar illness: a
naturalistic study. Bipolar Disord 2007; 9: 435–42.
15 National Institute for Clinical Excellence (NICE). Bipolar disorder: assessment and management. 2014
https://www.nice.org.uk/guidance/cg185.
16 Guy A, Davies J, Rizq R. Guidance for Psychological Therapists: Enabling conversations with clients taking or withdrawing from
prescribed psychiatric drugs. London: APPG for Prescribed Drug Dependence, 2019.
17 Reinares M, Sánchez-Moreno J, Fountoulakis KN. Psychosocial interventions in bipolar disorder: What, for whom, and when. J
Affect Disord 2014; 156: 46–55.
18 Cappleman R, Smith I, Lobban F. Managing bipolar moods without medication: a qualitative investigation. J Affect Disord 2015;
174: 241–9.
19 Miklowitz DJ. Adjunctive psychotherapy for bipolar disorder: state of the evidence. Am J Psychiatry 2008; 165: 1408–19.