Self-reported history of manic/hypomanic switch associated with antidepressant use: data from the systematic treatment enhancement program for bipolar disorder (step-bd)

Self-Reported History of Manic/Hypomanic Switch
Associated With Antidepressant Use: Data From the
Systematic Treatment Enhancement Program for
Bipolar Disorder (STEP-BD)
Christine J. Truman, M.D.; Joseph F. Goldberg, M.D.; S. Nassir Ghaemi, M.D., M.P.H.;
Claudia F. Baldassano, M.D.; Stephen R. Wisniewski, Ph.D.; Ellen B. Dennehy, Ph.D.;
Michael E. Thase, M.D.; and Gary S. Sachs, M.D.
Received Jan. 3, 2007; accepted March 26, 2007. From Finney Psychotherapy Associates, Norfolk, Va. (Dr. Truman); the Affective Objective: Antidepressant safety and efficacy
Disorders Program, Silver Hill Hospital, New Canaan, Conn., and the remain controversial for the treatment of bipolar Department of Psychiatry, Mount Sinai School of Medicine, New YorkN.Y. (Dr. Goldberg); the Department of Psychiatry and Behavioral depression. The present study utilized data from Sciences, Bipolar Disorder Research Program, Emory University, the National Institute of Mental Health System- Atlanta, Ga. (Dr. Ghaemi); the Department of Psychiatry, the University atic Treatment Enhancement Program for Bipolar of Pennsylvania, Philadelphia (Drs. Baldassano and Thase); University Disorder (STEP-BD) to examine the prevalence of Pittsburgh Medical Center and the Western Psychiatric Institute and and clinical correlates of self-reported switch Clinic, Pittsburgh, Pa. (Drs. Wisniewski and Thase); the Department into mania/hypomania during antidepressant of Psychological Sciences, Purdue University, West Lafayette, Ind.
(Dr. Dennehy); and the Department of Psychiatry, Partners Bipolar Treatment Center, Massachusetts General Hospital, Harvard Medical Method: Antidepressant treatment histories
School, Boston (Dr. Sachs). were examined from intake assessments for This project has been funded in whole or in part by the National the first 500 subjects enrolled into the STEP-BD Institute of Mental Health (NIMH), National Institutes of Health, under between November 1999 and November 2000.
Affective switch was defined as a report of Presented in part at the 156th annual meeting of the American Psychiatric Association, San Francisco, Calif., May 17–21, 2003. mania, hypomania, or mixed episodes within the Acknowledgments appear at the end of the article. first 12 weeks of having started an antidepressant.
Any opinions, findings, and conclusions or recommendations Demographic and clinical characteristics were expressed in this publication are those of the authors and do not compared for subjects with or without a history necessarily reflect the views of the NIMH. of acute switch during antidepressant treatment.
Financial disclosure appears at the end of the article.
Corresponding author and reprints: Joseph F. Goldberg, Results: Among the 338 subjects with prior
M.D., Silver Hill Hospital, 208 Valley Rd., New Canaan, CT 06840 antidepressant treatment and complete data on switch event outcomes, 44% reported at least 1such occurrence. Patients with a shorter durationof illness (odds ratio [OR] = 1.02, 95% CI = 1.01 epression remains the most common and disabling to 1.04) and a history of multiple antidepressant trials (OR = 1.73, 95% CI = 1.38 to 2.16) were phase of illness associated with bipolar disorder.1,2 more likely to report a history of switch than Traditional antidepressant agents are widely used for the other patients. A significantly increased risk for treatment of bipolar depression despite a limited database affective polarity switch was identified in patients to support their efficacy and safety.3 Notably, the phe- who had ever switched to mania/hypomania while nomenon of abnormal mood elevation during antide- taking tricyclic antidepressants (OR = 7.80, 95%CI = 1.56 to 28.9), serotonin reuptake inhibitors pressant treatment has been recognized to occur in about (OR = 3.73, 95% CI = 1.98 to 7.05), or bupropion 15% to 20% of individuals with bipolar disorder,4 al- (OR = 4.28, 95% CI = 1.72 to 10.6). Switch was though limited information exists to identify risk factors less common during treatment with electrocon- for such outcomes. The present study examined the his- vulsive therapy or monoamine oxidase inhibitors torical presence and illness correlates of antidepressant- than other antidepressants.
Conclusions: Antidepressants are associated
associated mania or hypomania among the first 500 sub- with the potential risk for treatment-emergent jects enrolling in the National Institute of Mental Health mania or hypomania, particularly in bipolar (NIMH) Systematic Treatment Enhancement Program for patients with short illness duration, multiple Bipolar Disorder (STEP-BD).
past antidepressant trials, and past experience Previous reports suggest that treatment-emergent af- of switch with at least one antidepressant.
(J Clin Psychiatry 2007;68:1472–1479) fective switch may be more common in bipolar patientswith a previous history of antidepressant-induced mania,bipolar I (versus II) disorder,5 comorbid substance J Clin Psychiatry 68:10, October 2007 Self-Reported Bipolar Mood Switch and Antidepressants ◆ Antidepressants are associated with the potential risk for treatment-emergent mania or hypomania in patients with bipolar disorder.
◆ Bipolar patients with a short duration of illness, multiple antidepressant trials, and past experience of switch with at least 1 antidepressant may be at particularly high risk.
◆ Clinicians should make every effort to clarify patient-specific vulnerabilities when considering risk-benefit ratios for the treatment of bipolar depression.
abuse,6,7 recent mania/hypomania,8 hyperthymic tem- have not attempted to control for such factors in multi- perament,9 younger age or earlier illness onset,6,10,11 or a variate models or else have involved sample sizes that genetic risk based on a polymorphism of the serotonin lacked adequate power to identify potential confounding transporter gene.12 Prior literature further suggests an in- herently greater risk for treatment-emergent mania with The present investigation utilized data from a large, the use of certain antidepressants, such as venlafaxine13 or nationwide effectiveness study of bipolar disorder in tricyclic antidepressants,14 although such studies have not order to (a) compare risks for affective polarity switch considered patient-specific baseline factors potentially among various antidepressant classes in a large, well- contributing to an intrinsic vulnerability for affective po- characterized bipolar cohort and (b) identify clinical fea- larity switch after antidepressant exposure.
tures associated with manic/hypomanic switch utilizing A recent meta-analysis concluded that antidepressants multiple regression analyses to control for potential con- pose no greater risk than placebo for inducing mania in founding factors.
bipolar depression15; however, inclusion of a dispropor-tionate number of subjects in that meta-analysis from a single randomized trial, in which olanzapine added tofluoxetine was construed as "placebo" added to fluoxe- The study cohort included the first 500 patients en- tine, limits the extent to which generalizations can be rolled in STEP-BD, a multi-site, NIMH-funded project drawn about the potential safety of antidepressants. Two that includes a large prospective, naturalistic study and a observational studies of 1-year antidepressant outcomes series of randomized, controlled trials. All participants after acute treatment for bipolar depression found high received the same standardized assessments at baseline rates of depression relapse with antidepressant cessation and during subsequent treatment visits, as well as quar- but rare switches to mania during long-term antidepres- terly outcome assessments. Full study procedures, out- sant continuation.16,17 While these findings would seem come measures, and sample characteristics have been provocative, causal inferences about the effects of antide- described in detail elsewhere.18,19 The protocol was ap- pressant continuation or cessation cannot be drawn from proved by the institutional review board at each site, and these studies because of their noncontrolled designs— all participants provided oral and written consent.
that is, subjects may have relapsed following antide-pressant cessation or simply may have stopped taking Study Participants and Procedures antidepressants after relapsing. Moreover, antidepressant STEP-BD subjects in the present study were enrolled continuation was feasible for only a small minority of in the protocol between November 1999 and November acutely depressed bipolar patients ( 15%) in the outcome 2000. Subjects were required to be at least 15 years of study by Altshuler and colleagues,16 which also excluded age and to be able to give informed consent. For partici- patients with DSM-IV rapid cycling.
pants aged 15 to 17 years, written assent was obtained While randomized, controlled trials represent an ideal along with written consent from a parent or legal guard- strategy to resolve uncertainties about the safety of anti- ian. All bipolar subtypes (bipolar I, bipolar II, bipolar not depressants, traditional efficacy-based studies often ex- otherwise specified, or cyclothymia) were included, as clude "real-world" patients for whom factors such as co- well as patients with schizoaffective bipolar disorder.
morbid substance abuse6,7 could influence switch rates Diagnoses were made from screening interview with antidepressants. Because observational (or effective- batteries conducted by experienced clinical investigators ness) studies impose fewer exclusionary criteria for sub- who underwent certification training in the adminis- ject enrollment, they provide greater opportunity to iden- tration of study diagnostic instruments. These included tify moderators and mediators of outcome that may be (1) the Mini-International Neuropsychiatric Interview altogether absent in samples recruited for traditional effi- (MINI)20 and (2) a standardized Affective Disorder cacy trials. Most existing observational studies either Evaluation (ADE),18 which integrates mood and psy- J Clin Psychiatry 68:10, October 2007 Truman et al.
Table 1. Clinical Characteristics of Subjects With or Without Antidepressant-Associated Manic/Hypomanic Switch
Full Antidepressant- Age, mean ± SD, y Age at onset of illness, mean ± SD, y Duration of illness, mean ± SD, y Marital status, % (N)a,b Race/ethnicity, % (N)c aBecause of incomplete data availability, percentages are based on N = 304 (full antidepressant-exposed sample), N = 171 (no history of switch), and N = 133 (history of switch).
bPercentage total (full antidepressant-exposed sample) is less than 100 because of rounding.
cBecause of incomplete data availability, percentages are based on N = 333 (full antidepressant-exposed sample), N = 186 (no history of switch), and N = 147 (history of switch).
Abbreviation: OR = odds ratio.
chosis modules from the Structured Clinical Interview were conducted in exploratory fashion to screen candidate for DSM-IV diagnosis (SCID).21 The ADE also obtained variables for entry into a subsequent multivariate regres- information about illness onset, prior episodes, past treatment and response, childhood psychopathology, co- Predictors of polarity switch were examined using SAS morbid medical and psychiatric conditions, psychoactive software (v. 8.2; SAS Institute, Cary, N.C.) with stepwise substance use, and family history.
logistic regression analyses using odds ratios (ORs) and The ADE treatment history module records yes/no re- 95% CIs. All models were adjusted for duration of illness, sponses to direct queries regarding exposure to more including all interim models in the stepwise selection than 30 standard treatment options for the treatment of process. In order to assess the hazards associated with bipolar disorders, including mood stabilizing agents, individual antidepressants, we used generalized estimat- electroconvulsive therapy (ECT), antidepressant and ing equations. This approach permitted within-subject antipsychotic medications, as well as light treatment and correlations to be controlled, as many subjects had been psychotherapy. The ADE records prestudy treatment ex- treated with multiple antidepressants. In assessing manic posures using a yes/no format; therefore, multiple trials switch for different antidepressant classes, selective se- of the same antidepressant medication are recorded as rotonin reuptake inhibitors (SSRIs), heterocyclics, and a single entry. As a result, each antidepressant trial de- monoamine oxidase inhibitors (MAOIs) were grouped scribed in this analysis represents at least one trial of a given treatment. STEP-BD defined manic/hypomanicswitch as a report of mania, hypomania, or a mixed epi- sode within the first 12 weeks of a given treatment. Theretrospective assessment design did not permit finer dis- Of the first 500 entrants into STEP-BD, 395 reported at tinctions between antidepressant-associated manias and least one prior trial of antidepressant medication prior to hypomanias by DSM-IV criteria, and either outcome was study entry. Complete data regarding a history of manic counted as a switch event. Histories of manic/hypomanic switch were available for 338 subjects. The 57 subjects switches were ascertained by direct subject report at the with incomplete data were excluded from the present baseline interview and recorded as a yes/no response for each treatment queried.
Within the final cohort of 338 STEP-BD participants with at least one antidepressant trial and complete data on Statistical Analysis switch history, 150 (44.4%) reported histories of at least For baseline characteristics, t tests were used to test one manic switch. A comparison of clinical and demo- for differences in the means of all continuous variables graphic characteristics of those subjects with and without such as age, length of illness, and age at onset of illness.
a history of polarity switch is presented in Tables 1 and 2.
For categorical variables, such as gender, race, and em- Those with a history of antidepressant-associated manic ployment, either χ2 or Fisher exact probability tests were switch were significantly younger, had a history of life- used, as appropriate. All statistical tests were 2-tailed time rapid cycling, and were more likely to report expo- with an α level of .05. Alpha levels were not corrected sure to multiple antidepressant medications than those for multiple comparisons, because univariate analyses without such a history. Gender, family history of bipolar J Clin Psychiatry 68:10, October 2007 Self-Reported Bipolar Mood Switch and Antidepressants Table 2. Clinical Features of Bipolar Patients With and Without Antidepressant-Associated Manic/Hypomanic Switch
1.05 (0.68 to 1.64) 0.71 (0.43 to 1.17) History of past-year rapid cycling 1.76 (1.03 to 3.01) No history of past-year rapid cycling Family history of bipolar illness 0.91 (0.43 to 1.92) No family history of bipolar illness History of substance abuse or dependence 1.19 (0.76 to 1.86) No history of substance abuse or dependence Prior manic phases, No.
Prior treatment with mood stabilizer 1.31 (0.69 to 2.51) No prior treatment with mood stabilizer Antidepressant trials, No.
Patients reporting a trial of 2 antidepressants 3–4 antidepressants ≥ 5 antidepressants aVariability in column numbers reflects missing data for variable being examined.
Abbreviations: FET = Fisher exact test, OR = odds ratio.
Symbol: … = odds ratios not calculable for continuous variables.
illness, bipolar subtype, history of substance abuse or de- Reports of manic/hypomanic switch with different anti- pendence, and number of prior manic episodes were not depressants are summarized in Table 3. Subjects were half significant correlates of antidepressant-associated switch as likely to report manic switches during MAOI treatment to mania/hypomania.
(OR = 0.16, 95% CI = 0.08 to 0.36) as compared to all The mean ± SD number of exposures to antidepres- other antidepressant classes. These observations remained sants was significantly higher in patients with a history significant after controlling for duration of illness and a of polarity switch than those without such a history history of rapid cycling. No other single treatment or class (3.64 ± 2.1 vs. 2.63 ± 1.8 trials, p = .001). Among the 150 (ECT, SSRIs, mirtazapine, bupropion, venlafaxine, nefa- subjects reporting a history of manic switch, 54 (36.0%) zodone, and heterocyclics) was found to have a significant reported affective switches with 2 or more different anti- association with manic/hypomanic switch when compared depressant medications, and 24 (15.0%) reported affec- to all other treatments.
tive switches with 3 or more different antidepressant As shown in Table 4, if patients had switched with any other antidepressant medications, they were also more In a stepwise logistic regression model, 2 predictors likely to switch with tricyclic antidepressants, SSRIs (with of manic/hypomanic switch emerged as significant pre- the exception of fluvoxamine), or bupropion. Those pa- dictors of antidepressant-associated switch: shorter dura- tients who switched to mania/hypomania during treatment tion of illness (OR = 1.02, 95% CI = 1.01 to 1.04) and a with one of those agents on a separate occasion had a 2- to history of multiple antidepressant trials (OR = 1.73, 95% 5-fold increased risk for a subsequent mania or hypomania CI = 1.38 to 2.16). While lifetime history of rapid cycling switch with any antidepressant. However, among indi- was a significant correlate of antidepressant-associated vidual SSRIs, fluoxetine was associated with an approxi- switch in univariate analyses, this variable was no longer mately 2.7-fold increased risk for switch to mania/hypo- significant when controlling for other parameters in the mania, while nefazodone had a significantly lower rate of polarity switches as compared to all other treatments.
J Clin Psychiatry 68:10, October 2007 Truman et al.
Table 3. Study Subjects Reporting At Least 1 Antidepressant Trial and Presence or Absence of
Antidepressant-Associated Manic/Hypomanic Switch
Patients Reporting Trials Without Patients Reporting Trials With Manic/Hypomanic Switch, Manic/Hypomanic Switch, 1 or more SSRI trialsb aOR reflects probability of a switch event's occurring with a given agent relative to all others.
bAnalysis for SSRIs as a class compared to other antidepressants did not include the separate specific SSRI trials. In contrast, analysis for each specific SSRI included the separate specific SSRI trials added to all other antidepressants.
Abbreviations: ECT = electroconvulsive therapy, MAOI = monoamine oxidase inhibitor, OR = odds ratio, SSRI = selective serotonin reuptake inhibitor.
Table 4. Probability of Manic Switch Based on History of Manic/Hypomanic Switch With Another
Antidepressanta
Manic/Hypomanic Switch Manic Switch Associated With Another Antidepressant 7.80 (1.56 to 28.9) 1.09 (0.16 to 7.59) 2.53 (0.47 to 13.6) 3.73 (1.98 to 7.05) 17.5 (1.90 to 161.4) 3.63 (1.85 to 7.14) 3.11 (1.33 to 7.27) 8.39 (2.69 to 26.2) 2.20 (0.17 to 28.1) 4.28 (1.72 to 10.6) 1.75 (0.13 to 23.7) < 3.20 (0.32 to 31.5) 1.54 (0.51 to 4.67) aVarying Ns reflect number of subjects who received each treatment.
Abbreviations: ECT = electroconvulsive therapy, MAOI = monoamine oxidase inhibitor, SSRI = selective serotonin
reuptake inhibitor, TCA = tricyclic antidepressant.
and Ghaemi et al.22 (49%) but higher than those reportedby Altshuler et al.16 (35%), Boerlin et al.11 (28%), Henry The present retrospective study represents the largest et al.9 (27%), or Joffe et al.17 (16%). Methodological investigation of affective polarity switch during treatment variations across studies may account for differences in with contemporary antidepressants for bipolar disorder.
reported prevalence rates of antidepressant-induced ma- The observed lifetime prevalence of 44% is comparable to nia, including factors such as small sample sizes, retro- similar rates reported by Goldberg and Whiteside6 (42%) spective recall, diversity of specific antidepressants and J Clin Psychiatry 68:10, October 2007 Self-Reported Bipolar Mood Switch and Antidepressants mood stabilizers, inconsistencies of medication dosing The lower reported manic/hypomanic switch rate with and adherence, failure to control for potential confound- MAOIs seen in the present study contrasts with findings ing factors,7 and variation in the operational definition of from a previous randomized trial by Himmelhoch and manic switch. While the present findings are limited by colleagues,25 which found no difference in switches be- the retrospective and nonrandomized study design, the tween tranylcypromine or imipramine (N = 56); how- use of regression modeling to control for potential con- ever, patients in that latter study were not routinely tak- founding factors provides an advantage relative to most ing adjunctive mood stabilizers. Another nonrandomized existing reports in this area.
study11 found comparable switch rates between tricyclic A key aspect of the current findings, particularly in antidepressants and MAOIs, both of which were higher relation to prior reports, is the observation that past than seen with fluoxetine, although the small sample size antidepressant-associated manic switches pose a risk for (N = 29) and lack of control for potential confounding additional switch events with other antidepressants. This factors limit generalizability of the findings. In addition, observation is consistent with observations of a patient- Bottlender and colleagues26 reported lower rates of specific vulnerability to antidepressant-induced mania switches to mania during treatment for bipolar depres- identified in several prior reports6,7,12,23 but contrasts with sion with an MAOI (regardless of the presence of a con- findings from the Stanley Bipolar Network,24 where such comitant mood stabilizer) relative to switch rates seen an association was not seen during prospective 10-week with tricyclic antidepressants.
acute treatment with bupropion, sertraline, or venla- It was also notable that all SSRIs were not equal faxine added to mood stabilizers. The extent to which in their associations with treatment-emergent mania/ antidepressant-associated mania reflects an adverse hypomania, since the overall class risk was elevated, iatrogenic event (as conceptualized within DSM-IV) driven predominantly by fluoxetine. These results are or an evoked response within a subgroup of vulnerable consistent with the only randomized study of fluoxetine individuals with bipolar disorder (as formulated in alone in bipolar I disorder,27 in which a 16% manic DSM-III-R) remains a critical, unresolved issue. It is switch rate was noted during crossover treatment with possible that antidepressant-associated switches arise fluoxetine, a rate similar to that of imipramine. Manic as the confluence of patient-specific and medication- switch risk may be lower with coadministration of olan- specific factors in a diathesis-stress model for a subgroup zapine28 or in bipolar II disorder.29 These data suggest of individuals with bipolar disorder. It is conceivable that that, in real-world practice, fluoxetine may carry higher the relative risk for switch events with specific antide- risk than other SSRIs.
pressants also may differ in patients once they have been The main limitations of the present study, similar sensitized to antidepressants, or had prior switches. Stud- to other observational studies, involve its noncontrolled, ies that report relatively high rates of lifetime manic nonrandomized design and retrospective assessment of switch (e.g., Ghaemi et al.22), including the present one, historical switch events. Information about past antide- may also reflect an enriched sample of tertiary care pa- pressant doses, concomitant mood stabilizer treatments, tients from academic settings for whom this baseline dia- pharmacotherapy adherence, and fulfillment of DSM-IV thesis may be especially high. However, the STEP-BD criteria for past mania or hypomania were not systemati- purposefully employed broad inclusion criteria in an ef- cally available. The retrospective design also may have fort to enroll patients who were representative of indi- limited the accuracy with which the historical count of viduals with bipolar disorder from across the United lifetime antidepressant trials could be recorded. The ob- States. Patients were not selected for treatment resistance served relationship between antidepressant-induced ma- or atypical demographic features.
nia and short duration of illness could also have been Significant predictors of manic switch for the current influenced by recall bias or possibly more extensive study group included younger age and greater past expo- treatment with mood stabilizing agents later in the course sure to antidepressants. These findings were independent of illness. Cell sizes for some specific antidepressant tri- of age at onset of illness and may suggest a vulnerable als in the current study group were relatively low (e.g., period early in the course of bipolar illness. Among the mirtazapine), limiting the statistical power with which to more commonly used antidepressants, rates of manic/ detect possible differences. Hence, outcomes with spe- hypomanic switch were relatively similar across SSRIs, cific antidepressant agents must be interpreted with cau- bupropion, and venlafaxine. MAOIs were associated tion and may serve more to generate rather than test with lower reported rates of manic/hypomanic switch hypotheses for affirmation in controlled trials.
than other classes. Patients were more likely to report In summary, antidepressant-associated mania appears a history of manic/hypomanic switch with bupropion, evident across a range of antidepressant classes and may citalopram, fluoxetine, sertraline, paroxetine, or tricyclic be influenced by patient-specific vulnerability factors.
antidepressants if they also had a history of switch with at Identifying and controlling for such potential factors and least 1 other antidepressant agent.
their interplay with specific antidepressants and adjunc- J Clin Psychiatry 68:10, October 2007 Truman et al.
tive mood stabilizers represents a critical next step for Drs. Truman, Baldassano, and Dennehy have no personal affiliations clarifying risk-benefit ratios when using antidepressants or financial relationships with any proprietary entity producing healthcare goods or services consumed by, or used on, patients to disclose for bipolar depression.
relative to the article.
Drug names: bupropion (Wellbutrin and others), citalopram (Celexa and others), fluoxetine (Prozac and others), imipramine (Tofranil and others), mirtazapine (Remeron and others), olanzapine(Zyprexa), paroxetine (Paxil, Pexeva, and others), sertraline (Zoloft 1. Judd LL, Akiskal HS, Schettler PJ, et al. The long-term natural history of and others), venlafaxine (Effexor and others).
the weekly symptomatic status of bipolar I disorder. Arch Gen Psychiatry2002;59:530–537 Disclosure of off-label usage: The authors have determined that, 2. Judd LL, Akiskal HS, Schettler PJ, et al. A prospective investigation to the best of their knowledge, bupropion, citalopram, fluoxetine, of the natural history of the long-term weekly symptomatic status of imipramine, mirtazapine, olanzapine, paroxetine, sertraline, venlafax- bipolar II disorder. Arch Gen Psychiatry 2003;60:261–269 ine, fluvoxamine, and nefazodone are not approved by the U.S. Food 3. Ghaemi SN, Lenox MS, Baldessarini RJ. Effectiveness and safety and Drug Administration for the treatment of bipolar depression.
of long-term antidepressant treatment in bipolar disorder.
If you have questions, contact the medical affairs department of J Clin Psychiatry 2001;62:565–569 the manufacturer for the most recent prescribing information.
4. Goldberg JF, Truman CJ. Antidepressant-induced mania: an overview of current controversies. Bipolar Disord 2003;5:407–420 Acknowledgments: This article was approved by the publication 5. Altshuler LL, Suppes T, Black DO, et al. Lower switch rate in depressed committee of STEP-BD. Investigators for STEP-BD are: patients with bipolar II than bipolar I disorder treated adjunctively with STEP-BD Contract: Gary S. Sachs, M.D. (Principal Investigator), second-generation antidepressants. Am J Psychiatry 2006;163:313–315 Michael E. Thase M.D. (Co-Principal Investigator), Mark S. Bauer, 6. Goldberg JF, Whiteside JE. The association between substance abuse M.D. (Co-Principal Investigator).
and antidepressant-induced mania in bipolar disorder: a preliminary STEP-BD Sites and Principal Investigators: Baylor College study. J Clin Psychiatry 2002;63:791–795 of Medicine (Lauren B. Marangell, M.D.); Case University (Joseph 7. Manwani SG, Pardo TB, Albanese MJ, et al. Substance use disorder and R. Calabrese, M.D.); Massachusetts General Hospital and Harvard other predictors of antidepressant-induced mania: a retrospective chart Medical School (Andrew A. Nierenberg, M.D.); Portland VA Medical review. J Clin Psychiatry 2006;67:1341–1345 Center (Peter Hauser, M.D.); Stanford University School of Medicine 8. MacQueen GM, Young LT, Marriott M, et al. Previous mood state (Terence A. Ketter, M.D.); University of Colorado Health Sciences predicts response and switch rates in patients with bipolar depression.
Center (Marshall Thomas, M.D.); University of Massachusetts Medi- Acta Psychiatr Scand 2002;105:414–418 cal Center (Jayendra Patel, M.D.); University of Oklahoma College 9. Henry C, Sorbara F, Lacoste J, et al. Antidepressant-induced mania of Medicine (Mark D. Fossey, M.D.); University of Pennsylvania in bipolar patients: identification of risk factors. J Clin Psychiatry 2001; Medical Center (Laszlo Gyulai, M.D.); University of Pittsburgh Western Psychiatric Institute and Clinic (Michael E. Thase, M.D.); 10. Nasrallah HA, Lyskowski J, Schroder D. TCA-induced mania: differ- University of Texas Health Science Center at San Antonio ences between switchers and nonswitchers. Biol Psychiatry 1982;17: (Charles L. Bowden, M.D.).
Additional details on STEP-BD can be located at http:// 11. Boerlin HL, Gitlin MJ, Zoellner LA, et al. Bipolar depression and antide- pressant-induced mania: a naturalistic study. J Clin Psychiatry 1998;59:374–379 Financial disclosure: In the spirit of full disclosure and in compli- 12. Mundo E, Walker M, Cate T, et al. The role of the serotonin transporter ance with all ACCME Essential Areas and Policies, the faculty for protein gene in antidepressant-induced mania in bipolar disorder.
this CME article were asked to complete a statement regarding all Arch Gen Psychiatry 2001;58:539–544 relevant financial relationships between themselves or their spouse/ 13. Post RM, Altshuler LL, Leverich GS, et al. Mood switch in bipolar partner and any commercial interest (i.e., a proprietary entity produc- depression: comparison of adjunctive venlafaxine, bupropion and ing health care goods or services) occurring within at least 12 months sertraline. Br J Psychiatry 2006;189:124–131 prior to joining this activity. The CME Institute has resolved any con- 14. Peet M. Induction of mania with selective serotonin re-uptake inhibitors flicts of interest that were identified. The disclosures are as follows: and tricyclic antidepressants. Br J Psychiatry 1994;164:549–550 Dr. Goldberg has served as a consultant for Cephalon, Eli Lilly, and 15. Gijsman HJ, Geddes JR, Rendell JM, et al. Antidepressants for bipolar GlaxoSmithKline and has served on speakers/advisory boards for depression: a systematic review of randomized, controlled trials.
Abbott, AstraZeneca, Eli Lilly, GlaxoSmithKline, and Pfizer.
Am J Psychiatry 2004;161:1537–1547 Dr. Ghaemi currently receives grant/research support from 16. Altshuler LL, Post RM, Leverich GS, et al. Antidepressant-induced GlaxoSmithKline and Pfizer; currently serves on the speakers mania and cycle acceleration: a controversy revisited. Am J Psychiatry bureaus for GlaxoSmithKline, AstraZeneca, Pfizer, and Abbott; and has served on the advisory boards for GlaxoSmithKline, Janssen, 17. Joffe RT, MacQueen GM, Marriot M, et al. Induction of mania and cycle Pfizer, Shire, and Abbott. Dr. Wisniewski has served as a consultant acceleration in bipolar disorder: effect of different classes of antidepres- for Cyberonics, ImaRx Therapeutics, and Bristol-Myers Squibb.
sant. Acta Psychiatr Scand 2002;105:427–430 Dr. Thase has served in an advisory/consultant capacity for 18. Sachs GS, Thase ME, Otto MW, et al. Rationale, design, and methods AstraZeneca, Bristol-Myers Squibb, Cephalon, Cyberonics, of the Systematic Treatment Enhancement Program for Bipolar Disorder Eli Lilly, GlaxoSmithKline, Janssen, MedAvante, Neuronetics, (STEP-BD). Biol Psychiatry 2003;53:1028–1042 Novartis, Organon, Sepracor, Shire, and Wyeth; has served on the 19. Perlis RH, Ostacher MJ, Patel JK, et al. Predictors of recurrence in speakers bureaus for AstraZeneca, Bristol-Myers Squibb, Cyberonics, bipolar disorder: primary outcomes from the Systematic Treatment Eli Lilly, GlaxoSmithKline, Organon, Sanofi Aventis, and Wyeth; has Enhancement Program for Bipolar Disorder (STEP-BD).
equity holdings in MedAvante; and receives royalty/patent income Am J Psychiatry 2006;163:217–224 from American Psychiatric Publishing, Guilford Publications, and 20. Sheehan DV, Lecrubier Y, Sheehan KH, et al. The Mini-International Herald House; his spouse/partner is Senior Medical Director of Neuropsychiatric Interview (M.I.N.I.): the development and validation Advogent (formerly Cardinal Health). Dr. Sachs has served as a of a structured diagnostic psychiatric interview for DSM-IV and ICD-10.
consultant for Abbott, AstraZeneca, GlaxoSmithKline, Eli Lilly, J Clin Psychiatry 1998;59(suppl 20):22–33; quiz 34–57 Merck, Pfizer, and Wyeth; has received grant/research support from 21. First MB, Spitzer RL, Gibbon M, et al. Structured Clinical Interview for Abbott, AstraZencea, Novartis, and Pfizer; and has served on the DSM-IV Axis I Disorders (SCID). New York, NY: Biometrics Research, speakers/advisory boards for Abbott, AstraZeneca, Bristol-Myers New York State Psychiatric Institute; 1996 Squibb, GlaxoSmithKline, Pfizer, Eli Lilly, and Sanofi; and his 22. Ghaemi SN, Rosenquist KJ, Ko JY, et al. Antidepressant treatment in spouse/partner is a stock shareholder of Concordant Rater Systems.
bipolar versus unipolar depression. Am J Psychiatry 2004;161:163–165 J Clin Psychiatry 68:10, October 2007 Self-Reported Bipolar Mood Switch and Antidepressants 23. Fogelson DL, Bystritsky A, Pasnau R. Bupropion in the treatment of developing antidepressant-induced maniform states in acute treatment of bipolar disorders: the same old story? J Clin Psychiatry 1992;53: of bipolar I depressed patients. J Affect Disord 2001;63:79–83 27. Cohn JB, Collins G, Ashbrook E. A comparison of fluoxetine, 24. Leverich GS, Altshuler LL, Frye MA, et al. Risk of switch in mood imipramine, and placebo in patients with bipolar depressive disorder.
polarity to hypomania or mania in patients with bipolar depression Int Clin Psychopharmacol 1989;4:313–322 during acute and continuation trials of venlafaxine, sertraline, and bupro- 28. Tohen M, Vieta E, Calabrese J, et al. Efficacy of olanzapine and olanza- pion as adjuncts to mood stabilizers. Am J Psychiatry 2006;163:232–239 pine-fluoxetine combination in the treatment of bipolar I depression.
25. Himmelhoch JM, Thase ME, Mallinger AG, et al. Tranylcypromine Arch Gen Psychiatry 2003;60:1079–1088 versus imipramine in anergic bipolar depression.
29. Amsterdam JD, Shults J. Fluoxetine monotherapy of bipolar type II Am J Psychiatry 1991;148:910–916 and bipolar NOS major depression: a double-blind, placebo-substitution, 26. Bottlender R, Rudolf D, Strauß A, et al. Mood-stabilisers reduce the risk continuation study. Int Clin Psychopharmacol 2005;20:257–264 For the CME Posttest for this article, see pages 1625–1627.
J Clin Psychiatry 68:10, October 2007

Table of Contents

Source: http://www.jbrf.org/wp-content/uploads/2012/04/Truman_JCP_2007_TEM_StepBD.pdf