Clinical Practice Guidelines for Bipolar Disorder

MANAGEMENT OF PERSONS WITH PSYCHOSES
Rapid Cycling Annotations

MODULE G

View Flow Chart for Module G.

Go to Module F.

Go to Module H.

A. Ensure Normal Thyroid Function

Some studies have found an association between thyroid hypofunction and rapid cycling (Bauer 1994), and it may play a role in the pathogenesis of rapid cycling (Bauer & Whybrow 1991; Joffe et al. 1996). Thus, persons with rapid cycling warrant thyroid screening with determination of thyroid-stimulating hormone (TSH) level. If TSH is abnormal, L-thyroxine should be used in sufficient doses to bring TSH into the normal range around 1 microgram per pound of body weight. The goal of treatment is to correct thyroid abnormalities and not use high-dose thyroxine as a specific treatment for rapid cycling (Bauer & Whybrow 1991). Referral to endocrinology may be required depending on clinical status. If the person is benefiting from lithium therapy, discontinuation of lithium is not necessary, even in cases of lithium-induced thyroid hypofunction; thyroid replacement therapy should be sufficient.

B. Institute Prophylaxis and Consider Psychosocial Rehabilitation

See Bipolar/ Schizoaffective Disorder Core Module, annotation F.

C. Initiate/Maximize Current Mood Stabilizer

There is data supporting two strategies on initial mood stabilizer choice for persons with rapid cycling (see Bipolar/ Schizoaffective Disorder Core Module D, annotation F, Table of Evidence No. 2). One strategy assumes that no one mood stabilizer is preferred over another; the second asserts that valproate and carbamazepine are preferred, implying that persons who develop rapid cycling should be switched from lithium.

Several studies support the second strategy. Rapid cyclers have higher rates of nonresponse to lithium than nonrapid cyclers (Dunner & Fieven 1974; Kukopulos et al. 1980). Switching to valproate (Calabrese et al. 1992; Emerich et al. 1985) or carbamazepine (Post et al. 1990) in persons who had previously not responded completely to lithium resulted in clinical improvement. A recent poll of experts favored valproate as the first-line acute antimanic agent for persons with rapid cycling (Frances et al. 1996). However, there are no data on direct comparison of lithium versus valproate or carbamazepine for rapid cycling. Furthermore, data from a controlled trial indicate that higher doses of lithium (> 0.8 mEq/L) are more effective than lower trial doses in

persons with frequent episodes (Gelenberg et al. 1989). One longitudinal study indicates that rapid cyclers treated with lithium may enter remission if antidepressants are discontinued (Kukopulos et al. 1980). Finally, most studies of rapid cycling derive from tertiary care clinical centers that tend to select for nonresponse to conventional or conservative treatment. Thus, the clinical response in rapid cyclers in general primary care populations (as in most Veteran Administration Medical Centers, VAMCs) may be different than that in published studies, which may not generalize.

Thus, this algorithm recommends lithium as first-line use rather than valproate or carbamazepine for lithium-naive rapid cyclers. This is a conservative stance, taken in recognition that definitive studies of anticonvulsants in this disorder are yet to be done. However, if lithium is used, it should be given at moderate to high doses (0.8 to 1.2 mEq/L).

D. Is the Person Taking Antidepressants?

Antidepressants may induce or exacerbate rapid cycling (Wehr & Goodwin 1987; Bauer et al. 1994; Altshuler et al. 1995), although not all studies are positive (Coryell et al. 1991). Tricyclic antidepressants may induce rapid cycling more than nontricyclics (Altshuler et al. 1995).

E. Reduce or Stop Antidepressants

There are two strategies for treating rapid cyclers who are taking antidepressants. The first approach is to discontinue antidepressants, assuming they are making rapid cycling worse; however, prolonged depression may result. If this depression can be managed successfully, the person may subsequently experience remission (Kukopulos et al. 1980). The second approach is to continue antidepressants and add treatment(s) in an attempt to induce remission (e.g., Bauer & Whybrow 1991). This approach is based on the clinical experience that prolonged depression in rapid cyclers after antidepressant withdrawal is common and often has morbidity comparable to continued rapid cycling (Bauer 1994).

The present algorithm advocates discontinuation of antidepressants as the initial step (Step 7) followed by a trial of alternate or combination mood stabilizers if the person is still symptomatic (Step 9). Subsequently, nontricyclic antidepressants (Step 14) may be tried if needed. Antidepressants should be discontinued by slow taper not faster than two weeks, since abrupt discontinuation of other medications in bipolar persons may induce relapse (Faedda et al. 1993; Suppes et al. 1993).

F. Response?

See Current Manic/Hypomanic/Mixed Episode Module E, annotation E.

G. Combine Mood Stabilizers

As noted above (Step 4), several mood stabilizers may be beneficial in treatment of rapid cycling and in nonrapid cycling. However, no prospective controlled trials of rapid cycling have been reported to date. Clinical judgment supports switching or adding mood stabilizers if a person has failed to respond to the initial agent. Treatment for rapid cycling should not be abandoned prematurely. At least 6 to 8 weeks, or several mood cycles, at therapeutic levels for a given agent is recommended to determine efficacy or lack thereof (Bauer 1994; Sachs 1996). This may be difficult when persons swing from one extreme to the other; there may be pressure to change interventions in these situations. However, the greater danger may be of "chasing one’s tail," adding antidepressants for depression, discontinuing them for hypomania, adding another for depression. Once a treatment plan is decided upon it should be adhered to until its natural conclusion.

If a person has not responded well to a particular mood stabilizer, the question arises whether to add another mood stabilizer or switch mood stabilizers. Since no data are available in rapid cycling to address the "add or switch" question, this decision must be left to clinical judgment. Similarly, when switching mood stabilizer, the decision of whether to taper the one currently being administered before or during the addition of the second agent must be left to clinical judgment. In either event, data from Suppes et al. (1993) and Faedda et al. (1993) indicating that rapid discontinuation of lithium (< 2 weeks) results in a higher proportion of relapses than does slow discontinuation suggest that the first agent should be tapered over at least two weeks. Again, however, the paucity of data leaves final decisions to clinical judgment.

H. Trial of Alternative Therapies Until Mood Stabilizes

In addition to lithium, valproate, and carbamazepine, a number of other mood stabilizers have been studied, primarily in open trials. The following table of evidence summarizes these modalities, virtually all of which have a "C" level of evidence. Not surprisingly, all published studies indicate positive effects. (Negative studies of innovative treatments on small samples unfortunately are usually not published.) Any positive therapeutic effect in treatment-refractory rapid cycling persons may deserve consideration.

I. Prolonged Depression?

Persons with rapid cycling will on occasion stop cycling and develop long episodes of depression or subsyndromic (hypo)mania. No systematic data are available to indicate when a person is likely to persist in the mood state and need additional or different treatment or when a switch to euthymia or resumed rapid cycling may occur. Therefore, clinical judgment must be used in deciding when to observe and when to intervene.

J. Careful Trial of Non-TCA

While virtually all antidepressants have been reported to induce mania or rapid cycling, TCAs may be more likely to do so than nontricyclic antidepressants as described by Peet for mania; and Altshuler for mania, and/or cycling. Thus, nontricyclics are preferred, although tricyclics are not contraindicated. Specific dosing regimens are found in the Bipolar Depressive Episode, Module F.

L. Continued Rapid Cycling?

If rapid cycling continues, persons treated with antidepressants should have mood stabilizers changed or added. This is equivalent to the alternative strategy outlined in the note for Steps 6 through 8, although by this point in the algorithm the person has failed mood stabilizers without antidepressants. Therefore, antidepressants plus mood stabilizers (Step 9, revisited), and possibly antidepressants plus alternative therapeutics (Step 13), are tried.

View Flow Chart for Module G.

Go to Module F.

Go to Module H.