Sign up for our e-newsletter

Clinical Supplement

Print Friendly 

Treatment of Rapid-Cycling Bipolar Disorder

Moderator: Robert M. Post, MD; Discussants: Kiki D. Chang, MD; Trisha Suppes, MD, PhD; Section editor: David L. Ginsberg, MD


CNS Spectr. 2004;9(2):1-10

An Expert Panel Review of Clinical Challenges in Psychiatry and Neurology

Funding for this monograph supplement has been provided through an unresticted grant by AstraZeneca


Rapid cycling, traditionally defined as 4 episodes/year, occurs in ~40% of bipolar patients.

Many rapid cyclers continue to cycle frequently even after aggressive treatment.

Data increasingly indicates that complex combination therapy is the most effective means of treating rapid-cycling bipolar disorder, and may prevent certain side effects.

Early diagnosis and treatment of rapid-cycling bipolar disorder is an increasing challenge as the age of onset decreases.




Approximately 40% of bipolar patients experience rapid cycling, and half of these suffer from ultra-rapid or ultradian cycling. These patterns are also common in children. Rapid-cycling bipolar disorder is difficult to bring to remission and often requires treatment with four or more classes of psychotropic medications. Lithium, even in combination with anticonvulsants or antidepressants, is often associated with residual episodic depressions. Concerns with adjunctive antidepressant treatment include their low response and remission rates and their tendency to cause switch into mania. Atypical antipsychotics and selected agents within the anticonvulsant class are becoming increasingly important in the treatment of rapid cycling. In the absence of clear treatment guidelines, the use and sequencing of drugs in complex combination treatment remains exploratory, but should be individualized based on careful prospective mood charting by the patient. Use of several drugs below their side-effect thresholds may prevent certain side effects. In children, long-term safety considerations are particularly important in the absence of a strong controlled clinical trials database.




Funding for this monograph supplement has been provided through an unrestricted educational grant by AstraZeneca. Sponsorship of this review does not imply the sponsor’s agreement with the views expressed herein. Although every effort has been made to ensure that drug usage and other information are presented accurately in this publication, the ultimate responsibility rests with the prescribing physician. Neither the publisher, the sponsor, nor the participants, can be held responsible for errors or for any consequences arising from the use of information contained herein. Readers are strongly urged to consult any relevant primary literature. No claims or endorsements are made for any drug or compound at present under clinical investigation


Robert M. Post, MD—Moderator


Clinical Definition and Prevalence of Rapid-Cycling Bipolar Disorder


The traditional definition of rapid cycling is ≥4 episodes/year.1 Patients who have ≥4 episodes/month are considered ultra-rapid cyclers and those who switch within a day on 4 days/week are termed ultra-ultra rapid or ultradian cyclers.2 Using these diagnostic cut-offs, the former Stanley Foundation Bipolar Outpatient Network (Kupka, et al, unpublished data, 2003) found that 42% of the first 674 people reported having experienced rapid or faster cycling. Surprisingly, 26.8% reported having experienced ultra-rapid cycling and 19.7% said they had experienced ultradian cycling (Slide 1).


In a follow-up study, Kupka and colleagues (unpublished data, 2003) examined those patients who were rated prospectively on a daily basis for ≥1 year without any missing data (N=539). Complex combination therapy is the norm in bipolar illness outpatient treatment and an average of 4.6 medication classes were used in the rapid cyclers compared to 3.5 in the nonrapid cyclers. Using the most conservative Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition criteria,3 the researchers found that 38.2% of the patients continued to have rapid cycling in the first prospective year. This percentage was substantially higher when a slightly looser cut-off was used. Thus, disappointingly, most of the patients who had said that they were rapid cycling previously continued to have episodes and rapid cycling even after fairly aggressive treatment.


Of the patients who had rapid cycling in the first prospective year, 55% of them remained rapid cyclers in the second year, 49% in the third, 45% in the fourth, and 43% in the fifth. The number of episodes decreased, but many of the more frequent rapid cyclers still continued to show ≥4 episodes/year.


During the first year, the rapid cyclers averaged 7.1 episodes compared to 1.4 episodes in the nonrapid cyclers. The majority of the extra episodes in the rapid cyclers were manic or hypomanic in nature. The days of ultradian cycling were also surprisingly high: rapid cyclers had an average of 25.7 days of ultradian cycling, as opposed to just 3 days in the nonrapid cyclers.


An assessment of the medical histories showed that the rapid cyclers had a greater history of dysphoric mania at a slightly higher percentage use of antidepressants, ie, 67% as opposed to 57% in the nonrapid cyclers. In addition, they had a higher incidence of use of thyroid augmentation.


A logistic regression analysis looking at the risk factors for those who had prospective rapid cycling in the first year of follow-up, showed some pertinent positive and negative correlations. As expected, prior history of rapid cycling was a very strong predictor: those who had experienced >10 prior episodes or >20 episodes prior to entering the network had an increased likelihood of being a rapid cycler compared with those with fewer episodes. History of prior drug abuse and physical or sexual abuse as a child had a weaker but still significant correlation with rapid cycling. Regarding negative correlation, gender, bipolar II disorder, prior history of antidepressant exposure, and hypothyroidism were not significant correlates in this analysis. This was surprising, particularly with respect to gender, as females are disproportionately represented in the rapid cyclers in many studies.




All of the atypical antipsychotics have been shown to have acute antimanic effects compared to placebo. Thus, along with lithium, mood stabilizers, antidepressants, and some of the anticonvulsants, the atypicals have joined the armamentarium of drugs commonly prescribed for patients with rapid-cycling bipolar disorder (Slides 2–4). Despite the availability of Food and Drug Administration-approved treatments, most treatment for bipolar illness is off-label. With the influx of treatments available, the most critical question concerns how to utilize these drugs and in what sequence.


Complex Combination Treatment


A recent article4 indicated that the combination of three specific drugs was much more effective in the treatment of acquired immunodeficiency syndrome (AIDS) than other triple combinations and that this combination was as effective as some four-agent combinations. Similar kinds of data for rapid-cycling bipolar disorder will be necessary to ultimately help guide the clinician in treating the patient to remission.


While some may think that four or five medications is extreme, this is the norm in other chronic medical illnesses, such as AIDS, tuberculosis, congestive heart failure, arthritis, and cancer. There may even be some advantages to using multiple medications because they differentially target multiple neurotransmitter systems, residual symptoms, and comorbidities. A powerful rationale comes from considerable data indicating a high failure rate even when two of the most effective agents are used in combination. (Slide 5). For example, in a study by Calabrese and colleagues5 rapid-cycling bipolar patients were treated with combination lithium and valproate in preparation for randomization to monotherapy. Only 25% of the original cohort were stabilized with these two medicines. This indicates that rapid-cycling patients tend to need additional treatments, particularly when the goal is trying to treat patients to full remission.


Another rationale for using complex combination treatments is the possibility of actually avoiding side effects. When only lithium and neuroleptics were available, it was common practice to push one or two drugs to their side-effect threshold or beyond in an attempt to reach a good response. However, there is a much better chance of stabilizing patients without side effects by using multiple drugs each below their side-effect threshold.


Current Recommendations and Future Goals


In the absence of information and guidance from controlled clinical trials, complex combination treatment is difficult. Once clinical trials methodology for rapid-cycling bipolar disorder catches up to clinical practice there will be more systematic data to determine the best treatment options. The STEP-BD (Systematic Treatment Enhancement System for Bipolar Disorder) program will provide an increasing amount of data in that regard, especially in adults. In the meantime, patients should be encouraged to help with very detailed mood charting in order to find the best possible ratio of maximum treatment efficacy and minimal side effects. Daily mood charts are a good way to obtain information early on about which drugs are working, which ones are not, and which side effects may be occurring, so that medications and doses can be adjusted as quickly as possible.


There are many unsolved problems in the treatment of rapid-cycling patients. The bad news is that it is difficult to achieve remission in these patients, but the good news is that many patients can be brought out of their rapid cycling to response or remission if one utilizes the wide range of agents now available. There are many more options to target residual symptoms that continue to break through or comorbidities that continue to occur despite treatment with the basic paradigms.


Once a patient has started rapid cycling or having multiple episodes, there is a certain amount of illness momentum and a difficulty in stopping this process. Therefore, one of the fundamental problems that needs to be addressed is how to treat patients earlier before the illness progresses. Early diagnosis and treatment is an increasing challenge as the age of onset of illness decreases to adolescence and even young children. The notion of recognizing the illness earlier and treating it with sustained long-term prophylaxis to avoid developing rapid cycling has considerable merit.



1. Dunner DL, Fieve RR. Clinical factors in lithium carbonate prophylaxis failure. Arch Gen Psychiatry. 1974;30:229-233.


2. Leverich GS, Post RM. Life charting of affective disorders. CNS Spectr. 1998;3:21-37.


3. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, DC: American Psychiatric Association; 1994.


4. Robbins GK, De Gruttola V, Shafer RW, et al. Comparison of sequential three-drug regimens as initial therapy for HIV-1 infection. N Engl J Med. 2003;349:2293-2303.


5. Calabrese JR, Shelton M, Rapport D, et al. A 20-month, double-blind, maintenance study of lithium vs. divalproex monotherapy in bipolar I and II disorder accompanied by rapid cycling. Bipolar Disord. 2003;5(suppl 1):38.



Dr. Post is head of the Bipolar Collaborative Network in Chevy Chase, Maryland.


Disclosure: Dr. Post is on the advisory boards of Abbott, AstraZeneca, Bristol-Myers Squibb, GlaxoSmithKline, Johnson & Johnson, Novartis, Ortho-McNeil, Shire, and UCB Pharma.


Kiki D. Chang, MD


Rapid Cycling in Children


Presence of rapid cycling appears to be the rule rather that the exception in children with bipolar disorder, although there is not much data to support this clinical notion. Geller and colleagues1 studied 93 pediatric patients with bipolar disorder, mean age 10.9 years. They found that 87% had rapid cycling defined as ≥4 episodes/year and 77% had ultradian cycling, defined as >365 episodes/year. In addition, Findling and colleagues2 studied 90 bipolar patients with a mean age of 10.8 years and found that 50% had rapid cycling by veteran CV criteria. (The study did not look at ultra-rapid or ultradian cycling.) Finally, using clinical interviews to study children with affective illness, Schraufnagel and colleagues3 found that 41% experienced >365 cycles/year (Slide 6).


It is difficult to figure out the difference between a cycle and an episode when children are cycling so quickly. Tillman and Geller4 proposed that any kind of mood switching constitutes a cycle. An episode, they proposed, is defined as having 2 weeks of mood symptoms with either onset and offset of one mood state within that 2 weeks, or onset and offset of a period of ultra-rapid or ultradian cycling. It appears that with younger age comes more rapid cycling and more mixed states, which can be confused with ultra-rapid cycling. In turn, the older the child gets, the less rapid cycling occurs, and the more episodic the disorder becomes with more euphoria and less irritability. However, this tendency has not been well studied.


Attention-deficit/hyperactivity disorder (ADHD) is more common in younger cohorts, which again can make diagnosis difficult and make rapid cycling appear to be more common. When a patient is euthymic but experiencing ADHD symptoms during the euthymic period, ADHD can be confused with another mood episode. Also, patients with a younger age of onset typically have more family history of bipolar disorder in general, and more commonly of the rapid-cycling type. They also tend to have more family history of comorbid ADHD and substance abuse (Slide 7).


As there are few well-controlled studies of medication in pediatric bipolar disorder, the information on how to treat rapid cycling in children is scant. However, because many if not most children with bipolar disorder experience rapid cycling, the current literature could be extended to rapid-cycling populations, since most researchers have not specified if subjects were rapid cycling or not in the past studies.


Psychopharmacologic Treatment


Lithium and Anticonvulsants


The only placebo-controlled medication studied in monotherapy was with lithium, in a study conducted by Geller and colleagues.5 The study found lithium to be more effective than placebo in decreasing both substance use and manic symptoms in adolescents with bipolar disorder. More recently, there was one large open study6 of 100 patients treated with lithium in an in-patient unit; however, within that category, it was unclear if lithium was predominantly effective for rapidcycling.


While there are no placebo-controlled data for divalproex, there is some open data. A study by Findling and colleagues7 looked at the combination of lithium and divalproex in 90 children and adolescents with bipolar disorder. Of the 42 remitters (Young Mania Rating Scale score <12.5 and Children’s Depression Rating Scale score <40) 52% had a history of rapid cycling. Of the non-remitters, 67% had a history of rapid-cycling. Thus, having a history of rapid cycling did not seem to significantly affect response to treatment of lithium plus divalproex, ie, a large percentage of both remitters and non-remitters had a history of rapid cycling. Other open studies8-11 in divalproex were not segregated for rapid-cycling history.


There is one open sudy9 and some case series regarding carbamazepine in childhood bipolar disorder.12 In addition, the data with lamotrigine consists of only one open trial13 that included some adolescents, in which the history of rapid-cycling response was not known. DelBello and colleagues14 recently conducted a placebo-controlled study with topiramate. The incidence of rapid cycling and the responders versus nonresponders in this study is not yet known(Slide 8).

Atypical Antipyschotics

One prospective open study was conducted with olanzapine.15 A retrospective case analysis with risperidone was published16 and a prospective study is in press.17 Again, not much is known about the rapid-cycling status of the patients in those studies.


DelBello and colleagues18 conducted a placebo-controlled add-on study with quetiapine, where quetiapine or placebo was added to divalproex in a group of 30 manic adolescents. While the rapid-cycling history of those adolescents was not reported, 23 out of the 30 did have a mixed episode on admission. In children and adolescents it is often difficult to tell the difference between a mixed episode and ultradian cycling. Regardless, 87% of the divalproex plus quetiapine group responded well, compared to 53% for the divalproex plus placebo group.


Recommendations for Treatment


Despite the limited data for treatment of rapid cycling in children and adolescents, overall guidelines are needed because children commonly present with rapid cycling (Slide 9). Because of the limited data, most of the guidelines have not been based on empirical data from children but rather extrapolated from the adult data and combined with general clinical experience. It appears that most children with rapid cycling require combination therapy. Early data with combination therapy suggest that lithium plus divalproex would be effective acutely as well as divalproex plus an atypical antipsychotic.


Unpublished data (2003) from the Child and Adolescent Bipolar Foundation polled their constituents about how many medications their children, who had presumed diagnoses of bipolar disorder, were taking. The mean amount was more than three medications per child. Thus, it is pretty typical for bipolar children to be treated with combination pharmacotherapy.


Safety and Tolerability


Physicians should be aware of safety concerns of combination treatment in children because of their apparent necessity in children, and because there is a relative absence of well-studied treatments. It is important to “start low and go slow” since the long-term safety data in children are not as well known.


Baseline laboratories, including thyroid status, should be obtained for any agent before starting a medication trial. Physicians should also be aware of drug interactions including the effect of too many anticonvulsants in younger children. This includes effects of increasing the half-life of lamotrigine in patients taking valproate. This combination has led to a higher rash incidence in children than adults. Physicians must also be aware any comorbid medical conditions.


Administering antipsychotics in doses of twice a day or three times a day seems to make them more tolerable and more effective in children and adolescents. In addition, using longer-acting preparations may decrease side effects, simplify dosing, and increase adherence to medication regimens. Long-acting preparations exist for lithium, divalproex, and carbamazepine.


Finally, physicians should be very cautious when considering the use of selective serotonin reuptake inhibitors or stimulants in children or adolescents with bipolar disorder. The data is still forthcoming, but certainly clinically it has been reported that both these drug classes can exacerbate mania or rapid cycling and that children may be especially sensitive to this phenomenon. Therefore, the first thing we recommend before starting any medication treatment is to consider either tapering or even stopping any stimulant or antidepressant that the child is currently taking (Slide 10).




1. Geller B, Williams M, Zimerman B, et al. Prepubertal and early adolescent bipolarity differentiate from ADHD by manic symptoms, grandiose delusions, ultra-rapid or ultradian cycling. J Affect Disord. 1998;2:81-91.


2. Findling RL, Gracious BL, McNamara NK, et al. Rapid, continuous cycling and psychiatric co-morbidity in pediatric bipolar I disorder. Bipolar Disord. 2001;3:202-210.


3. Schraufnagel CD, Brumback RA, Harper CR, Weinberg WA. Affective illness in children and adolescents: patterns of presentation in relation to pubertal maturation and familyhistory. J Child Neurol. 2001;8:553-561.


4. Tillman R, Geller B. Definitions of rapid, ultrarapid, and ultradian cycling and of episode duration in pediatric and adult bipolar disorders: a proposal to distinguish episodes from cycles. J Child Adolesc Psychopharmacol. 2003;13:267-271.


5. Geller B, Cooper TB, Sun K, et al. Double-blind and placebo-controlled study of lithium for adolescent bipolar disorders with secondary substance dependency. J Am Acad Child Adolesc Psychiatry. 1998a;37:171-178.


6. Kafantaris V, Coletti DJ, Dicker R, Padula G, Kane JM. Lithium treatment of acute mania in adolescents: a large open trial. J Am Acad Child Adolesc Psychiatry. 2003;42:1038-1045.


7. Findling RL, McNamara NK, Gracious BL, et al. Combination lithium and divalproex sodium in pediatric bipolarity. J Am Acad Child Adolesc Psychiatr. 2003;42:895-901.


8. Kowatch RA, Sethuraman G, Hume JH, et al. Combination pharmacotherapy in children and adolescents with bipolar disorder. Biol Psychiatry. 2003;53:978-984.


9. Kowatch RA, Suppes T, Carmody TJ, et al. Effect size of lithium, divalproex sodium, and carbamazepine in children and adolescents with bipolar disorder. J Am Acad Child Adolesc Psychiatry. 2000;39:713-720.


10. Wagner KD, Weller EB, Carlson GA, et al. An open-label trial of divalproex in children and adolescents with bipolar disorder. J Am Acad Child Adolesc Psychiatry. 2002;41:1224-1230.


11. Papatheodorou G, Kutcher SP, Katic M, Szalai JP. The efficacy and safety of divalproex sodium in the treatment of acute mania in adolescents and young adults: an open clinical trial. J Clin Psychopharmacol. 1995;2:110-116.


12. Hsu LK. Lithium-resistant adolescent mania. J Am Acad Child Psychiatry. 1986;2:280-283.


13. Kusumakar V, Yatham LN. An open study of lamotrigine in refractory bipolar depression. Psychiatry Res. 1997;2:145-148.


14. DelBello MP, et al.Poster presented at: Annual Meeting of the American College of Neuropsychopharmacology; December 2003; San Juan, Puerto Rico.


15. Frazier JA, Biederman J, Tohen M, et al. A prospective open-label treatment trial of olanzapine monotherapy in children and adolescents with bipolar disorder. J Child Adolesc Psychopharmacol. 2001;11:239-250.


16. Frazier JA, Meyer MC, Biederman J, et al. Risperidone treatment for juvenile bipolar disorder: a retrospective chart review. J Am Acad Child Adolesc Psychiatry. 1999;38:960-965.


17. Biederman J. Open-label study of risperidone in children with bipolar disorder. Poster presented at: Annual Meeting of the European College of Neuropsychopharmacology; September 2003; Prague, Czech Republic.


18. Delbello MP, Schwiers ML, Rosenberg HL, Strakowski SM. A double-blind, randomized, placebo-controlled study of quetiapine as adjunctive treatment for adolescent mania. J Am Acad Child Adolesc Psychiatry. 2002;41:1216-1223.


Dr. Chang is assistant professor of psychiatry in the Department of Psychiatry at Stanford University in California.                 

Disclosure: Dr. Chang is a consultant to and on the speaker’s bureaus of Abbott, AstraZeneca, Eli Lilly, GlaxoSmithKline, Janssen, Ortho-McNeil, and Shire; and has received grant/research support from Abbott and GlaxoSmithKline.



Trisha Suppes, MD, PhD


The Spectrum of Rapid Cycling


While rapid cycling was first identified by Dunner and Fieve in 1974,1 there is quite a good historical background on the concept. In the 1800s Falret and colleagues2 identified what they called a “circular course distinct from alternating.” What they meant by “alternating” was a period of remission following an episode. Currently, all types and ranges of cycling are lumped under one rubric; the spectrum ranges from having >4 cycling episodes, the minimum required for rapid cycling, through ultradian cycling, which is far more severe.


While some patients manifest chronic depression with periodic hypermania, which is known as bipolar II, others have ultradian cycles, which are often seen as pure affective instability. We do not know whether these types represent different stages of a disorder that exists on a spectrum or whether they are actually distinct categories of the disorder: one being patients who experience mania, depression, and then remission; the second being those with more continuous cycling that includes ultradian cycling. Greater understanding of the biological underpinnings of the illness will help to determine whether these conditions exist on a spectrum.


Rapid cycling may become less prevalent with age. In the Depression Collaborative Study, Coryell and colleagues3 found a diminishment of rapid cycling over the course of 13 years. Studies from the Stanley Bipolar Foundation group4 suggest that the cycling may accelerate in some, but not others. Over a long observation period in ~100 patients, Koukopoulos and colleagues5 found that once a pattern of rapid cycling is established in a patient, this pattern may become a hallmark of their illness. Rapid cycling is also sometimes correlated with an earlier onset age of illness and some have found that is correlated with increased risk of suicide as well (Slide 11).




Although the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV),6 recognizes that antidepressants in and of themselves can cause a switch into mania, the debate continues as to whether they truly are a causal factor. Whether or not an antidepressant-induced “switch” signals the future development of biopolar disorder, it does indicate a vulnerability. A second concern is whether antidepressants can worsen course of illness during its use and even once it is discontinued. When to discontinue an antidepressant is another concern. These issues are pertinent because of the burden of depression that most patients with rapid cycling experience.


In terms of discontinuation of antidepressants, Altshuler and colleagues7 conducted a study of ~80 patients in which 50% discontinued their antidepressant after remission of symptoms and 50% did not. This was not a randomized study, but there were prospective ratings. They found that those who did not discontinue their antidepressants over the following 6–8 months had a lower burden of illness, including decreased hypomania and overall burden itself, in terms of depression. This is important as a hypothesis-generating finding because all of the treatment guidelines, including those of the American Psychiatric Association, international groups, and the Texas algorithm recommend terminating antidepressant use as soon as remission of symptoms is achieved, which is usually within 2–4 months (Slide 12).

Ghaemi and colleagues8 conducted an open prospective trial and followed 30% of the overall group of rapid cyclers for 1 year. Patients who had remitted on antidepressants and were continuing their antimanic agents were randomized to either long-term maintenance, or short-term, in which case the antidepressant was discontinued. When they adjusted for confounds including the presence of rapid cycling, gender, and antidepressant attitude, the overall response and survival rate for long-term versus short-term continuation of antidepressants was about the same. Thus, the risk of relapse was not increased or decreased by continuing the antidepressant. Rapid cycling itself was the strongest predictor of poor overall response. In addition, for those nonrapid-cycling patients who continued the antidepressant, the overall magnitude of morbidity, which was a combined score of mania and depression, was somewhat worse. Overall, whether to continue or discontinue an antidepressant after response, is still an open question (Slide 13).9,10



In the 1970s it was recognized that lithium appeared to be somewhat less effective for rapid-cycling patients. Calabrese and colleagues11 evaluated monotherapy in bipolar patients with rapid cycling in an extended open phase of patients who had stabilized on a combination of lithium and divalproex. All subjects had rapid-cycling bipolar 1 or II. Of those who entered the initial open phase, only one in four stabilized (~50/200) and could be randomized to monotherapy with lithium or divalproex. Of those 50, only 10% were full completers of the study, which lasted many months. Thus, out of 200 rapid-cycling patients only 5 showed extended stability on monotherapy. This indicates that more combination treatment trials are warranted to find more effective treatments. In addition, the majority of the noncompleters both in the open phase and double-blind phase were drop-outs due to depression. This indicates that rapid cycling in and of itself is a prediction of poorer outcome and may be considered a more severe variant of the illness. Furthermore, the study found equal efficacy between lithium and divalproex in the randomized group of patients who completed the double-blind phase. This contradicts other studies indicating that anticonvulsants may be more effective than lithium, particularly in monotherapy.




Anticonvulsants that have shown some efficacy include divalproex, carbamazepine, oxcarbazepine, topiramate, and lamotrigine. Calabrese and colleagues12 conducted a 6-month study of lamotrigine in rapid cyclers, which included a 10-week open-phase period. Patients who stabilized on lamotrigine monotherapy for 2 weeks were randomized to double-blind treatment with lamotrigine versus placebo. The study found a significant advantage for patients with lamotrigine over placebo. In a long-term controlled study, Denicoff and colleagues13 found a significant advantage for lithium plus carbamazepine than either medication alone in rapid cyclers.A number of small, open case series14-16 support the use of the newer anticonvulsants, particularly in combination treatment, for patients inadequately responsive to monotherapy.Thus, in terms of the emerging studies, virtually all of the newer anticonvulsants show some promise. While none of the drugs appear to be as helpful in monotherapy, they may be useful adjuncts (Slide 14).

Atypical Antipsychotics


In several open add-on case series, all of the newer atypicals, including olanzapine, quetiapine, risperidone, and clozapine showed some degree of efficacy. The acute controlled trials with the atypical antipsychotics also show good response. One study17 involved randomization between clozapine add-on versus treatment as usual and followed the patients for 1 year. Of those who received clozapine, ~50% were rapid cycling versus nonrapid cycling according to DSM-IV criteria. The researchers expected that rapid-cycling patients would be more responsive to clozapine than nonrapid cyclers. However, an analysis of the data showed that the nonrapid cyclers not only did better initially, they actually did better throughout the whole study.


In a secondary analysis, Tohen and colleagues18 evaluated a cohort of patients who received either divalproex or olanzapine for 47 weeks. In this blinded study, the nonrapid cyclers were more responsive than the rapid cyclers in both treatment groups. Initially the rapid cyclers in the divalproex group did a bit better than the nonrapid cyclers but this effect was no longer seen after week 10.


In a placebo-controlled study of olanzapine, Sanger and colleagues19 looked specifically at response in rapid cyclers. A secondary analysis out of their acute mania study showed that those with a history of rapid cycling responded to olanzapine; however, the study did not compare the nonrapid cyclers with the rapid cyclers.


A recent controlled study revealed preliminary results for quetiapine in depression. The short-term study included rapid cyclers, but the data is not yet available.


Thus, in terms of the emerging studies, virtually all of the atypical antipsychotics show some promise and may be useful adjuncts. However, none appear to be very helpful in monotherapy (Slide 14).

Predictors of Treatment Response


Factors associated with less severe illness generally predict better response. These include later onset of illness (eg, early adult onset versus childhood onset), presence of less substance abuse or neurological impact (eg, head injury), and not having mania to depression sequence. These factors predict response in rapid-cycling patients in the same way they do those without rapid cycling. Thus, we do not yet know how patients with rapid-cycling bipolar disorder differ from those with bipolar disorder generally—whether they are biologically different or whether they just possess a more severe variant of illness that predicts decreased response.


Analyses of results of a controlled lamotrigine study by Calabrese and colleagues12 indicated that being bipolar II or being female predicted an improved response. In addition, early stabilization versus later stabilization during the 10-week open phase of the trial also predicted response.20 Other studies21 have shown an increased burden toward depression among those who are rapid cycling. One study22 reviewing the treatment data for bipolar II showed more of a burden of depression, and in some cases a burden of rapid cycling.


Treatment Consideration and Guidelines

Although lithium is commonly used in bipolar patients, as early as the 1970s it was recognized that the antimanic effect of lithium would often lead to patients stabilizing a bit below baseline; this is something observed with several antimanic medications. Antimanic agents do not always fully normalize mood and whether resulting from that or from the natural course of the disorder, the patient does not necessarily become truly euthymic after receiving treatment for a manic episode, although they may not meet criteria for a full depressive episode. Quality of life for the patient is very much affected by low-grade ongoing depressive symptoms as well as potential risk for suicide. There is no single medication that is the answer to the problem; there are no magic bullets with lithium, the newer anticonvulsants, or the atypical antipsychotics. Therefore, combination antimanic and antidepressant drugs are often necessary.           


There is no real consensus nor body of evidence to really inform the physician on treatment of rapid-cycling patients. Overall, the general principles are to use combinations of treatments, and the atypicals look promising. There is still a relative role for lithium, anticonvulsants, and lamotrigine. In addition, physicians are cautioned to use antidepressants sparingly although there are several cases where they may be appropriate even with all the treatment options (Slide 15).

Experimental Treatments


There are some experimental and unusual clinical approaches to treating rapid-cycling bipolar disorder. First, is calcium channel blockers, particularly nimodipine, and high-dose levothyroxine. There are also various case reports on the use of electroconvulsive therapy (ECT) in rapid cycling. One of the problems with ECT is that while one can accomplish acute stabilization, particularly if there is a lot of depression, the cognitive impact may be cumulative in maintenance treatment. Because of the frequency of cycles in these patients, one cannot prevent cycling without risking significant cognitive impact.


A study with a very new experimental approach called vagus-nerve stimulation (VNS) is now being conducted. VNS has shown some efficacy in treatment-resistant depression.23,24 The mechanical device, which functions like a pacemaker, is approved by the Food and Drug Administration for use in treatment-resistant epilepsy. It is inserted in the upper chest and the wire runs up to the vagus nerve and stimulates it with a very low dose of electricity. However, the treatment resistent depression studies have found a cumulative effect with VNS.25 Thus, VNS should be reserved for very ill people in whom other treatments have failed.


We are now studying VNS in rapid-cycling bipolar disorder in collaboration with Marangell in Houston, Texas. In the small sample we have right now, we have seen an impressive degree of improvement in patients receiving VNS. Some of these experimental approaches may prove to be quite important for the more severely ill (Slide 16).     



1. Dunner DL, Fieve RR. Clinical factors in lithium carbonate prophylaxis failure. Arch Gen Psychiatry 1974;30:229-233.


2. Falret JP. Marche de lafolie (suite). Gaz Hop. 1851;24(18-19).


3. Coryell W, Soloman D, Turvey C, et al. The long-term course of rapid-cycling bipolar disorder. Arch Gen Psychiatry. 2003;60:914-920.


4. Suppes T, Leverich GS, Keck PE Jr, et al. The Stanley Foundation bipolar outcome network II. Demographics and illness characteristics of the first 261 patients. J Affect Disord. 2001;67:45-59.


5. Koukopoulos A, Sani G, Koukopoulos AE, et al. Duration and stability of the rapid-cycling course: a long-term personal follow-up of 109 patients. J Affect Disord. 2003;73:75-85.


6. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, DC: American Psychiatric Association; 1994.


7. Altshuler LL, Suppes T, Black D, et al. Impact of antidepressant discontinuation after acute bipolar depression remission on rate of depressive relapse at 1-year follow-up. Am J Psychiatry. 2003;160:1252-1262.


8. Ghaemi S, El-Mallakh R, Baldassano CF, et al. Antidepressant treatment of bipolar I depression: long-term outcome. Poster presented at: Annual Meeting of the American College of Neuropsychopharmacology; December 2003; San Juan, Puerto Rico.


9. Ghaemi SN, Hsu DJ, Soldani F, Goodwin FK. Antidepressants in bipolar disorder: the case for caution. Bipolar Disord. 2003;5:421-433.


10. Post RM, Leverich GS, Nolen WA, et al. A re-evaluation of the role of antidepressants in the treatment of bipolar depression: data from the Stanley Foundation Bipolar Network. Bipolar Disord. 2003;5:396-406.


11. Calabrese JR, Shelton MD, Rapport DJ, et al. Is rapid cycling a predictor of non-response to lithium? Poster presented at: Annual Meeting of the American College of Neuropsychopharmacology; December 2002; San Juan, Puerto Rico.


12. Calabrese JR, Suppes T, Bowden C, et al. A double blind, placebo-controlled, prophylaxis study of lamotrigine in a rapid-cycling bipolar disorder. J Clin Psychiatry. 2000;61:841-850.


13. Denicoff KD, Smith-Jackson EE, Disney ER, et al. Comparative prophylactic efficacy of lithium, carbamazepine, and the combination in bipolar disorder. J Clin Psychiatry. 1997;11:470-478.


14. Suppes T. Review of the use of topiramate for treatment of bipolar disordrers. J Clin Psychopharmacol. 2002;6:599-609.


15. Pies R. Combining lithium and anticonvulsants in bipolar disorder: a review. Ann Clin Psychiatry. 2002;14:4.


16. Calabrese JR, Shelton MD, Rapport DJ, et al. Current research on rapid cycling bipolar disorder and its treatment. J Affect Disord. 2001;67:241-255.


17. Suppes T, Webb A, Paub B, et al. Clinical outcomes in a randomized one-year trial of clozapine versus treatment as usual for patients with treatment-resistance illness and a history of mania. Am J Psychiatry. 1999;8:1164-1169.


18. Tohen M, Ketter TA, Zarate CA, et al. Olanzapine versus divalproex sodium for the treatment of acute mania and maintenance of remission: a 47-week study. Am J Psychiatry. 2003;160:1263-1271.


19. Sanger TM, Tohen M, Vieta E, et al. Olanzapine in the acute treatment of bipolar I disorder with a history of rapid cycling. J Affect Disord. 2003;73:155-161.


20. Suppes T, Calabrese J, Bowden C, et al. Lamotrigine in rapid-cycling bipolar disorder: predictors of response. W J Biol Psychiatry. 2001;3(suppl 1):S347.


21. Calabrese JR, Shelton MD, Bowden CL, et al. Bipolar rapid cycling: focus on depression as its hallmark. J Clin Psychiatry. 2001;62(suppl 14):34-41.


22. Suppes T, Dennehy EB. Evidence-based long-term treatment of bipolar II disorder. J Clin Psychiatry. 2002;63:29-33.


23. Rush JA, George MS, Sackeim JA, et al. Vagus nerve stimulation (VNS) for treatment-resistant depression: a multicenter study. Biol Psychiatry. 2000;47:276-286.


24. Sackeim HA, Rush JA, George MS, et al. Vagus nerve stimulation (VNS) for treatment-resistant depression: efficacy, side effects, and predictors of outcome. Neuropsychopharmacology. 2001;5:713-728.


25. Marangell LB, Rush JA, George MS, et al. Vagus nerve stimulation (VNS) for major depressive episodes: one year outcomes. Biol Psychiatry. 2002;4:280-287.

Dr. Suppes is associate professor of psychiatry in the Department of Psychiatry and director of the Bipolar Disorders Clinic and Research Program at UT Southwestern Medical Center in Dallas, Texas.

Disclosure: Dr. Suppes is a consultant to or on the advisory boards of Abbott, AstraZeneca, Bristol-Myers Squibb, Eli Lilly, GlaxoSmithKline, Janssen, Johnson & Johnson, Novartis, Ortho-McNeil, Pharmaceutical Research Institute, Pfizer, and UCB Pharma; and has received grants from Abbott, AstraZeneca, Bristol-Myers Squibb, GlaxoSmithKline, Janssen, Novartis, Robert Wood Johnson Pharmaceutical Research Institute, and the Stanley Medical Research Institute.


Question & Answer Forum


Q: How often do rapid-cycling patients switch mood states and is there a difference in how one would treat those patients with more frequent cycling?


Dr. Post: When Kupka and colleagues1 looked at the frequency of cycling, they found that it was really on a continuum—that there was no clear break at four episodes/year. Thus, while cycling is a handy way of characterizing the fast subgroup, in fact there is little empirical evidence for a dichotomous cut-off at four episodes. When Kupka and colleagues1 correlated episode frequency with all the other variables, they again found that they were continuously and not dichotomously related. Therefore, when we refer to rapid cycling, we are really talking about an illness that can occur at almost any cycle frequency. This is important in relation to both pathophysiology and treatment.


Dr. Suppes: In our lamotrigine predictors study,2 we saw no difference in treatment response among patients with more frequent episodes of rapid cycling, for example, >6 episodes/year versus those with 4–6 episodes/year.


Dr. Post: That seems to be the case in many studies. But how this affects our diagnostic approaches or therapeutics is not entirely clear other than recognizing that we are dealing with this continuum. It seems to be even more of problem in the young children with bipolar illness.


Q: Is super high-dose thyroid treatment an effective and safe method of reducing rapid cycling?


Dr. Suppes: Some recent studies have found that it can be very helpful in some patients. The issues are the need for close medical monitoring and potential risk for medical complications over time.


Dr. Post: Bauer and colleagues3,4 indicate that there are more than 100 patients in the literature exposed to levothyroxine (T4) in doses of 200–500 µg/day, and that patients either respond pretty clearly or not at all. They have not found much in the way of medical complications, although I have heard about tachycardia and other complications from others. This type of thyroid treatment is something to consider in patients who are not well stabilized despite a series of medications. It may be worth a careful try in highly treatment-refractory rapid cyclers.


Dr. Chang: As far as thyroid augmentation, there are no studies in children, unfortunately. However, clinically and anecdotally, if we do have a child who is unstable and has even a borderline thyroid-stimulating hormone (TSH) level of 5 or 6, we will often augment with thyroid to see if that helps, especially if they are on lithium. Gracious and colleagues5 found that ~35% of children will have significant elevations of TSH after initiation of lithium, so thyroid abnormalities are present in ~the same percentage of children as adults and should be looked at very carefully especially in patients on lithium.


Dr. Post: Are you referring to regular and not supraphysiological doses?


Dr. Chang: Yes. We use regular dose levothyroxine either T3 or T4. We have not tried high-dose suppression of TSH, although I routinely try to get the TSH level under 1.0.


Dr. Post: We have used a lot of low-dose thyroid with some success, so my approach is to actually try ordinary T3 augmentation (25.0–37.5 µg/day) for refractory cyclic or depressed patients, and see if that makes a difference before considering the more problematic very high-dose T4 regimens.


Dr. Suppes: There used to be this idea that rapid cyclers as a group had more thyroid abnormalities. However, a review by Leibenluft6 looking at women, and data by Kupka and colleagues,7 did not find systematic hypothyroidism or thyroid abnormalities in these patients.


Dr. Post: Several studies7-9 have not found increased thyroid abnormalities in rapid cyclers. It looks like there may be an increased sensitivity to the antithyroid effects of lithium in rapid cyclers, but hypothyroidism or thyroid antibodies do not seem to be associated with rapid cycling per se.


Q: What is the role of omega-3 fatty acids in bipolar patients?


Dr. Post: Omega-3 fatty acids are particularly of interest in children as they seem to be pretty benign in terms of side-effect profile.


Dr. Chang: Yes. If it does turn out to be useful it may be more prophylactic rather than effective in any kind of acute treatment of full-blown mania or depression in these children. We do not seem to see an acute effect with the omega-3 fatty acids, but I am particularly interested in looking at high-risk populations and seeing if it can delay or prevent the onset of bipolar disorder in high-risk children. The Stanley Foundation Bipolar Network recently found some negative findings with the omega-3 fatty acids (Keck P, unpublished data, 2003).


Dr. Suppes: They actually had a positive finding too.


Dr. Post: Yes, these data have been submitted for review. It was a double-blind 4-month study of EPA, the supposed active ingredient of the omega-3 fatty acid, 6 g/day versus placebo for 4 months in ~120 patients randomized either for an acute depressive episode or for cycling. There did not seem to be any evidence of omega-3 fatty acids being more effective than placebo overall. In a secondary analysis, it looked like in the cycling group, older patients dropped out much faster on active drug and vice versa in younger patients. Thus, there is still a possibility that some doses or preparations of omega-3 fatty acids may be helpful in younger patients.


Whether we used too high a dose (EPA 6 g) has been called into question, because there are now several positive studies with lower doses, 1–2 g EPA. While EPA looked more promising in the younger patients, the entire study population was >18 years of age. Thus, the omega-3 fatty acids clearly deserve further study in children, especially because of the high safety profile.


Q: What are your preferred treatment methods for patients with difficult-to-treat or ultra-rapid cycling?


Dr. Suppes: First, I try to remove any agent that might be contributing to the rapid cycling. This would involve decreasing the dose or discontinuing stimulants and antidepressants. Sometimes we are unable to fully discontinue those medications because the depression symptoms become too severe. Then I try to increase the agents typically used for mania. I have found the atypical antipsychotics to be helpful with this. Therefore, it was surprising to see that non-rapid cyclers were more responsive than the rapid cyclers on clozapine10 and olanzapine,11 but I think that is just indicative of the nature of rapid cycling, which is in and of itself a predictor of decreased response. In my own clinical experience I have had patients, particularly with the more rapid kinds of cycling and mood instability, respond very well to the atypicals, such as quetiapine or olanzapine, perhaps in combination with an anticonvulsant and an antidepressant. Due to side effects, clozapine is lower down on my treatment algorithm, but when a patient has a lot of rapid cycling or mood instability, it is definitely one of the options I consider.


Dr. Post: What kind of dose of clozapine do you tend to use and what anticonvulsant do you like to combine with it?


Dr. Suppes: In our clozapine study,12 we found that those patients who were rapid cycling had a lower mean peak dose than the non-rapid cycling patients. This may indicate that one reason we are seeing less response is the dose is too low; however, I have found that many of these patients are often sensitive to side effects. And the kinds of doses we use with clozapine and rapid-cycling bipolar disorder is quite low, at 50–200 mg/day.


Dr. Post: Yes. Since we are all using an average of at least three drugs in these more complex presentations of the illness, it is really important to titrate doses to keep them below the side-effect threshold and achieve good tolerability for the whole regimen. If one really achieves good side-effect tolerability, not only can patients improve, but they are much more likely to stay on long-term maintenance treatment. That becomes a key issue in this illness.


Dr. Chang: My treatment algorithm follows Dr. Suppes’ pretty closely. However, children are very often on stimulants because most of them have comorbidity of attention-deficit/hyperactivity disorder so they often present on high stimulant regimens or even combinations of different stimulants. We then try to decrease the stimulants, although it is very difficult because sometimes they need these medications to achieve in school. We are experimenting with replacing the stimulants without atomoxetine to see if there is less cycling, but the early anecdotal data is still pending.


We follow a very similar approach in children as in adults as far as using atypicals in conjunction with lithium and divalproex. Typically, when we use combination therapies we try to use one from each category, so very often we will have children on lithium plus divalproex plus an atypical. However, very often there are tolerability or side-effect problems, such as weight gain and cognitive effects. These problems typically occur more often in children than adults. 


Q: How do you manage combination treatment in children, who tend to be particularly sensitive to side effects such as weight gain?


Dr. Chang: We try to institute diet and exercise, but unfortunately this is very difficult to do with some children and their families. We often add topiramate, which has a ~30% success rate in decreasing appetite and weight gain. We may also try to remove the agent that is most responsible for the weight gain and replace it with something else that is not associated with as much weight gain.Unfortunately, it seems like the agents with the most efficacy data also have potential for the greatest weight gain. We also try to use the medications short term when possible, and then slowly withdraw the weight-inducing agent after the stabilization period is over. However, we are hopeful that some of the newer agents, which appear to have less weight gain, will show more efficacy data so that we can start using them instead.


Dr. Post: In adults, Leverich and colleagues13 found that zonisamide showed good antimanic effect and to a lesser extent, antidepressant effects as an open add-on. Zonisamide did have parallel degrees of weight loss to topiramate, ~one third of a pound/week. Zonisamide may be a potential agent to consider for weight loss as well, especially for people who cannot tolerate topiramate.


Dr. Chang: Absolutely. However, there are no data yet on zonisamide use in children.



1. Kupka RW, Luckenbaugh DA, Post RM, et al. Rapid and non-rapid cycling bipolar disorder: a comparative study using daily prospective mood ratings in 539 outpatients. Bipolar Disord. 2003;5:62.


2. Suppes T, Calabrese J, Bowden C, et al. Lamotrigine in rapid-cycling bipolar disorder: predictors of response. W J Biol Psychiatry. 2001;3(suppl 1):S347.


3. Bauer M, Priebe S, Berghofer A, Bschor T, Kiesslinger U, Whybrow PC. Subjective response to and tolerability of long-term supraphysiological doses of levothyroxine in refractory mood disorders. J Affect Disord. 2001;64:35-42.


4. Bauer M, Berghofer A, Bschor T, et al. Supraphysiological doses of L-thyroxine in the maintenance treatment of prophylaxis-resistant affective disorders. Neuropsychopharmacol. 2002;27:620-628.


5. Gracious BL, Findling RL, Seman C, et al. Elevated thyrotropin in bipolar youths prescribed both lithium and divalproex sodium. J Am Acad Child Adolesc Psychiatry. 2004;2:215-220.


6. Leibenluft E. Women with bipolar illness: clinical and research issues. Am J Psychiatry. 1996;153:163-173.


7. Kupka RW, Nolen WA, Post RM, et al. High rate of autoimmune thyroiditis in bipolar disorder: lack of association with lithium exposure. Biol Psychiatry. 2002;51:305-311.


8. Joffe RT, Kutcher S, MacDonald C. Thyroid function and bipolar affective disorder. Psychiatry Res. 1988;25:117-121.


9. Post RM, Kramlinger KG, Joffe RT, et al. Rapid cycling bipolar affective disorder: lack of relation to hypothyroidism. Psychiatry Res. 1997;72:1-7.


10. Post RM, Leverich GS, Nolen WA, et al. A re-evaluation of the role of antidepressants in the treatment of bipolar depression: data from the Stanley Foundation Bipolar Network. Bipolar Disord. 2003;5:396-406.


11. Tohen M, Ketter TA, Zarate CA, et al. Olanzapine versus divalproex sodium for the treatment of acute mania and maintenance of remission: a 47-week study. Am J Psychiatry. 2003;160:1263-1271.


12. Suppes T, Ozcan ME, Carmody T. Response to clozapine of rapid cycling vs. noncycling patients with a history of mania. Bipolar Disord. 2004. In press.


13. Leverich GS, McElroy SL, Altshuler LL, et al. The anticonvulsant zonisamide in bipolar illness: clinical response and weight loss. Bipolar Disord. 2004. In press.



RSS Link  Current Issue
RSS Link  CME Articles
RSS Link  Non-CME Articles

View more issues