Drug Treatments for Maintenance

Mary Butler; Snezana Urosevic; Priyanka Desai; Scott R. Sponheim; Jonah Popp; Victoria A. Nelson; Viengneesee Thao; Benjamin Sunderlin

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Butler M, Urosevic S, Desai P, et al. Treatment for Bipolar Disorder in Adults: A Systematic Review [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2018 Aug. (Comparative Effectiveness Review, No. 208.)

Cover of Treatment for Bipolar Disorder in Adults: A Systematic Review

Treatment for Bipolar Disorder in Adults: A Systematic Review [Internet].

Show details

Contents

< Prev Next >

Appendix IDrug Treatments for Maintenance

Section 1. Lithium Monotherapy vs. Inactive Controls

Appendix Table I1Characteristics of eligible studies: lithium monotherapy for maintenance

Study, Year Design Location Funder Risk of Bias PMID	# Randomized Age (mean) Sex (% Female) Race (% White) Diagnosis (% BP I, II, NOS) Setting	Inclusions Key Exclusions	Intervention Dosage	Comparison Dosage	Followup Duration	Outcomes Reported Withdrawal (%) at endpoint
Bowden, 2000¹ RCT Multisite US Industry RoB High 10807488 12784116²	N=372 Mean Age 39 Female 51% White 94% BP I 100% Outpatient	No Episode at Randomization; Scores of MRS ≤ 11, DSS ≤ 13, GAS > 60; At least one other manic episode in past three years. Substance Abuse; Other Mental Health Conditions; Taking Other Medications; Pregnant/Nursing	Divalproex 71–125 mcg/mL (Mean 84.8 mcg/mL) N= 187	C1: Placebo n= 92 C2: Lithium 0.8–1.2 mEq/L (Mean 0.9 mEq/L) n= 90	52 week	Time to recurrence, any Time to recurrence, mania (MRS≥16) Time to recurrence, depression (DSS≥25) Withdrawal 69% (Time to recurrence outcomes only included) Nonrelapse Withdrawal 40%
Amsterdam, 2010³ RCT Single-site US Gov’t+nonprofit RoB Moderate 20360317	N =81 Mean Age 38 Female 52% White NR BP II 100% Outpatient	Depression and HAM-D≥16 initialy; then HAM-D≤8 after 12 weeks of initial Fluoxetine therapy at 20–80mg/day) Substance abuse Neurological Disorders Taking other medications Pregnant/NursingLabs/Other Conditions	Fluoxetine (n=28) 10–40 mg/day (mean 34.3 mg/day) N=28	C1: Placebo (n=27) C2: Lithium (n=26) 300–1200 mg/day; 0.5–1.5 mmol/L (mean 1027 mg/day; 0.69 mmol/L)	50 weeks	YMRS Relapse (HAM-D ≥14) Adverse Events Lab Values Withdrawal 72% (Time to recurrence outcomes only included) Nonrelapse Withdrawal 25%
Bowden, 2003⁴ RCT of responders Multisite 2 continents Industry RoB Low 12695317	N=175 Mean Age 41 Female 47% Race NR BP I 100% Outpatient	Lamotrigine responders (CGI-S≤3 for at least 4 continuous weeks) For open label period: Manic; DSM-IV Criteria for Mania or Hypomania currently or within past 60 days with previous episodes in past 3 years Other Mental Health Conditions	Lamotrigine 100–400 mg/day (Mean NR) N=59	C1: Placebo N=70 C2: Lithium 0.8–1.1 mEq/L Mean NR N=59	18 Months	Time to intervention for mania, hypomania, mixed, depression, and any mood episodes Time to early discontinuation Adverse Events l Withdrawal 80% (Time to recurrence outcomes only included) Nonrelapse Withdrawal 20%
Calabrese, 2003⁵ RCT of responders Multisite 4 Continents Industry RoB Moderate 14628976	N=410 Mean Age 43 Female 56% Race NR BP I 100% Outpatient	Lamotrigine responders (CGI-S<3 for at least 4 continuous weeks) For open label period: depression; DSM-IV criteria for depression currently or within past 60 days with previous depression and mania episodes in past 3 years. Other Mental Health Conditions	Lamotrigine 50, 200, or 400 mg/day (in analysis 50mg group was censored, 200 and 400 mg/day groups were combined) N=221	C1: Placebo N=121 C2: Lithium 0.8–1.1 mEq/L (Mean 0.8 mEq/L) N=121	18 Months	Time to intervention for mania, hypomania, depression, any mood episode Time to early discontinuation Adverse Events Withdrawal 84.9% (Time to recurrence outcomes only included) Nonrelapse Withdrawal 34%
Prien, 1973⁶ RCT Multisite US Government RoB High 4569674	N=205 Median Age 44 Sex NR Race NR BP I 100% Outpatient	No Episode at Randomization; Neurological Disorders; Abnormal Lab Results	Lithium 0.5–1.4 mEq/L (Mean 0.7 mEq/L)	Placebo	2 years	Relapse, type/proportion (manic, mixed, schizoaffective, depressive event requiring hospitalization ‘severe’ or supplementary drugs ‘moderate’) Time to early discontinuation GAS IMPS (Inpatient Multidimensional Psychiatric Scale) Self-Report Mood Scale KAS (Katz Adjustment Scale) Adverse Events Withdrawal 42%
Weisler, 2011⁷ RCT Multisite 5 continents Industry RoB Moderate 22054050	N=1226 Mean Age 40 Female 53% White 63% BP 1 100% Outpatient	Meeting stability criteria of YMRS ≤ 12 and MADRS ≤ 12 after last episode of depression/mania/mixed episode at study entry or within past two years Substance Abuse Other Mental Health Conditions Pregnant/Nursing Labs/Other Conditions	Quetiapine 300–800 mg/day (543 mg/day mean) N=404	C1: Placebo N=404 C2:Lithium 600–1800 mg/day 0.6–1.2 mEq/L (0.63 mEq/L mean) N=364	104 weeks	Time to recurrence any mood (algorithm) Time to manic event Time to depressive event Time to all-cause discontinuation SDS MOS-Cog CGI-BP PANNS-P WPAI TMT Withdrawal 55% (Time to recurrence outcomes only included) Nonrelapse withdrawal 22%

: Abbreviations: AIMS=Abnormal Involuntary Movement Scale; BARS=Barnes Akathisia Scale; BIS-11=Barratt Impulsiveness Scale; BMI=Body Mass Index; BP=bipolar disorder; C=Comparison; CGI-BP=Clinical Global Impressions Scale for Bipolar Disorder; CGI-BP-M=Clinical Global Impressions Scale-Bipolar-modified (for long-term follow-up); CGI-BP-S=Clinical Global Impressions, Bipolar, Severity Scale; CGI-EI=Clinical Global Impressions-Efficacy Index; CGI-I=Clinical Global Impressions Scale, Improvement; CGI-S=Clinical Global Impressions, Severity Scale; DSM=Diagnostic and Statistical Manual of Mental Disorders; DSS=Depressive Syndrome Scale; EPS=extrapyramidal symptoms; ESRS=Extrapyramidal Symptom Rating Scale; GAF=General Assessment of Functioning Scale; GAS=Global Assessment Scale; HAM-D=Hamilton Scale for Depression; IMPS=Inpatient Multidimensional Psychiatric Scale; KAS=Katz Adjustment Scale; LIFE-RIFT= Longitudinal Interval Follow-up Evaluation-Range of Impaired Functioning Tool; MADRS=Montgomery-Asberg Depression Rating Scale; MOS-Cog=Medical Outcomes Study Cognitive Scale; MRS=Mania Rating Scale; NOS=not otherwise specified; NR=not reported; PANSS=Positive and Negative Syndrome Scale; PGWB=Psychological General Well-being Scale; PMID=PubMed Identification Number; PSQI=Pittsburgh Sleep Quality Index; QIDS-SR=Quick Inventory of Depressive Symptomatology (Self-reported); RCT=randomized controlled trial; ROB=risk of bias; SAS=Simpson Angus Scale; SDS=Sheehan Disability Scale; TMT=Trail Making Test; WPAI=Work Productivity and Activity Impairment Questionnaire; YMRS =Young Mania

Appendix Table I2Summary risk of bias assessments: Lithium monotherapy for maintenance

Drug	Study Funding Source PMID	Overall Risk of Bias Assessment	Rationale
Lamotrigine vs. Lithium vs. placebo	Bowden, 2003⁴ Industry 12695317	Moderate (High for log rank test)	Randomization and blinding procedures not described. Balanced traits among groups at baseline. Efficacy data may be biased by dropout as endpoints are LOCF and differential nonrelapse dropout rates that range from 16–25% of patients for each arm.
Lamotrigine vs. Lithium vs. placebo	Calabrese, 2003⁵ Industry 14628976	Moderate (High for log rank test)	Randomization and allocation concealment not described. Balanced traits among groups at baseline. Efficacy data may be biased by dropout as endpoints are LOCF. 34% withdrawal.
Fluoxetine vs. lithium vs. placebo	Amsterdam, 2010³ Government/Nonprofit 20360317	Moderate	Randomization may not have been successful as it relates to a person’s likelihood of relapse; randomization, allocation, and blinding procedures are underdescribed. 23% dropout.
Divalproex vs. lithium vs. placebo	Bowden, 2000¹ Industry 10807488 12784116²	High	Appears to be unblinded. Randomization not described. Nonrelapse dropout of 40%.
Lithium vs. placebo	Prien, 1973⁶ Government 4569674	High	Differential dropout rate is significant. Study didn’t demonstrate allocation concealed.
Quetiapine vs. lithium vs. placebo + lithium	Weisler, 2011⁷ Industry 22054050	Moderate	Generally a well conducted and reported study, however some sources of bias present related to dropout rates. Overall 21% withdraw due to reasons unrelated to recurrence; however by week 16 more than 50% of the placebo group has dropped for all causes, as well as 40% of the lithium group and 25% of the quetiapine group)

: Abbreviations: BP=bipolar; LOCF=Last observation carried forward; PMID=PubMed Identification Number; RCT=randomized controlled trial

Appendix Table I3Outcomes summary table: lithium monotherapy versus placebo for maintenance

Drug	Study Comparison PMID	Responder/Remitter	Symptom	Function	Other	AE
Lithium vs. placebo	Bowden, 2000¹ 10807488 12784116²	Time to any recurrence 12 months NS Time to mania recurrence 12 months NS Time to depression recurrence 12 months NS	NA (Exclude for attrition = 69%)	NR	NR	SAE NR Tremor Lithium 42% (38/94) Placebo 13% (12/94) Favors Placebo p<0.001 Akathisia Lithium 4% (4/94) Placebo 1% (1/94) NS
	Amsterdam, 2010³ 20360317	Time to Relapse to Depression 50 week Log rank NS	NA (Exclude for attrition = 72%)	NR	NR	SAE Reported no events EPS Reported no events
	Bowden, 2003⁴ 1269531	Time to any recurrence 18 months Log rank Favors Lithium p=0.001 Time to mania recurrence 18 months Log rank Favors Lithium p=0.006 Time to depression recurrence 18 months Log rank NS	NA (Exclude for attrition = 80%)	NA	NA	SAE No events reported Suicidality per HAM-D NS between groups Tremor Lithium 42% (38/94) Placebo 13% (12/94) Favors Placebo P<0.001 Akathisia Lithium 4% (4/94) Placebo 1% (1/94) NS
	Calabrese, 2003⁵ 14628976	Time to any recurrence 18 months Log rank Favors Lithium p=0.03 Time to mania recurrence 18 months Log rank NS Time to depression recurrence 18 months Log rank NS	NA (Exclude for attrition = 85%)	NA	NA	SAE 1 death in lamotrigine groups Suicidality per HAM-D NS between groups Tremor Lithium 17% (20/120) Placebo 5% (6/121) Favors Placebo P<0.05
	Prien, 1973⁶ High 4569674	Relapse, any episode type (manic, mixed, schizoaffective, depressive event requiring hospitalization or supplementary drugs) Lithium 43% (43/101) Placebo 80% (84/104) Favors Lithium p<0.001	NR	NR	Suicide Placebo 1/104 Overall Withdrawal Lithium 27% (27/101) Placebo 57% (59/104) Favors lithium p<0.001 Withdrawal due to ‘poor clinical response’ (further participation would seriously jeopardize the patients physical or mental health) Lithium 11% (11/101) Placebo 40% (41/104)	SAE Lithium 24% (24/101) Placebo 0% (assumed - none reported) Lithium: 3/101 – Toxicity (1) Hypothyroidism with goiter, polyuria, polydipsia (1) Hypothyroidism w/o goiter (3) Leukocytosis (1) Death Lithium 1/101 Placebo 1/104
	Weisler, 2011⁷ 22054050	Time to recurrence of any mood event Favors Lithium HR 0.46 (95% CI 0.36, 0.59), p<0.0001 Time to recurrence of manic event Favors Lithium HR 0.37 (95% CI 0.27, 0.53), p<0.0001 Time to recurrence of depression symptoms Favors Lithium HR 0.59 (95% CI 0.42, 0.84), p<0.004	NA (Exclude for attrition 55%)	NR	NR	SAE 10 lithium 11 placebo EPS 38 lithium 18 placebo No deaths

: Abbreviations: AE=Adverse Events; CGI-BP-S=Clinical Global Impressions, Bipolar, Severity Scale; CI=Confidence Interval; EPS=extrapyramidal symptoms; HAM-D=Hamilton Scale for Depression; HR=Hazard Ratio; LSM=least-squares means; MADRS=Montgomery-Asberg Depression Rating Scale; NA=Not applicable; NR=not reported; NS=not significant; OR=Odds Ratio; PMID=PubMed Identification Number; SAE=Serious Adverse Events; YMRS = Young Mania Rating Scale

Appendix Table I4Strength of evidence assessment: lithium monotherapy versus placebo for maintenance

Comparison	Outcome	# Studies/Design (n analyzed)	Finding or Summary Statistic Outcome Timing	Study Limitations	Consistency	Directness	Precision	Overall Grade/Conclusion
Lithium vs. placebo	Time to overall relapse 1–2 yrs	6 RCT (n=1579)	Favors Lithium 1 to 2 years	Moderate	Consistent	Direct	Imprecise	Low (weighted by moderate study and hazard ratio)
Lithium vs. placebo	Time to manic or depressive relapse 1–2 yrs	6 RCT (n=1579)	See table above	Moderate	Inconsistent	Direct	Imprecise	Insufficient

: Abbreviations: MADRS=Montgomery-Asberg Depression Rating Scale; RCT=randomized controlled trial; YMRS = Young Mania Rating Scale
: Notes:
1: Publication bias for antipsychotics, antidepressants, and behavioral interventions for depressive disorders is suspected.
2: Data were generally imprecise due to missing data from high attrition rates, which was commonly dealt with by Last Observation Carried Forward (LOCF). LOCF requires an assumption that the health status of patients who dropped out of the trial would not have changed had future observations been recorded, a strong assumption in the context of bipolar disorder research.

Section 2. Other Monotherapy

Appendix Table I5Characteristics of eligible studies: other monotherapy for maintenance by year then first author

Study, Year Design Location Funder Risk of Bias PMID	# Randomized Age (mean) Sex (% Female) Race (% White) Diagnosis (% BP I, II, NOS) Setting	Inclusions Key Exclusions	Intervention Dosage	Comparison Dosage	Followup Duration	Outcomes Reported Withdrawal (%) at endpoint
Calabrese, 2017⁸ RCT Multisite 4 Continents Industry RoB Moderate 28146613	N=266 Mean Age 41 Female 58% Race White 54% Black/African American 28% BP I 100% Inpatient/Outpatient	≥1 previous manic or mixed episode severity requiring hospitalization or treatment with mood stabilizer or treatment with antipsychotic agent. Current episode YMRS ≥20 but then met YMRS≤12, MADRS≤12, no active suicidality Rapid Cycling Refractory BP First Manic Episode Substance Abuse Other Mental Health Labs/Other Conditions	Aripiprazole once-monthly injections 400 mg	Placebo	52 week	Time to recurrence (hospitalization or YMRS>15; MADRS>15; CGI-BP>4; further medication; suicidality) Proportion meeting recurrence YMRS MADRS Withdrawal 61.7% (nonrelapse withdrawal 23%)
Keck, 2006⁹ RCT Multisite 2 Continents Industry Rob Moderate 16669728	N=161 Mean Age 40; Female 67% Race White 56% Hispanic/Latino 23% BP I 100% Outpatient	Symptom stability: YMRS≤10 and MADRS≤13 for 4 consecutive visits over 6 weeks Substance Abuse Other Mental Health Labs/Other Conditions Pregnant/Nursing Unresponsive to Clozapine ECT in last 2 years	Aripiprazaole 15 or 30 mg/day based on investigator discretion	Placebo	26 weeks	Time to relapse (defined as lack of efficacy) YMRS MADRS PANSS CGI-BP Withdrawal 58% (nonrelapse withdrawal 24%)
Hartong, 2003¹⁰ RCT Multisite Netherlands Industry/Government RoB Low 12633122	N = 98 Mean Age 42 Female 54% Race NR BP I 77% BP-II 23% Outpatient	Remission from any Episode Type; According to Bech Rafaelsen Mania or Melancholia Scales First Manic Episode	Carbamazepine 6–10 mg/L (6.8 mg/L average)	Lithium 0.6–1 mmol/L (0.75 mmol/L average)	2 years	Remission (YMRS <=12) Response (50% decrease in YMRS) Efficacy YMRS CGI-BP-S MADRS Adverse events Extrapyramidal symptoms SAS AIMS BARS Withdrawal 31%
Greil, 1997¹¹ RCT Multisite Germany Government RoB High 9165384 also 9864077¹² 10529070¹³ 10529071¹⁴ 11093063¹⁵	N = 171 Mean Age 40 Female 57% Race NR BP I 58% BP-NOS 33% Outpatient	Remission from any Episode Type; GAS > 70 First Manic Episode Substance Abuse Other Mental Health Neurological Disorders	Carbamazepine 635 mg/day	Lithium 26.8 mmol/day	2 years (130 weeks?)	Remission (YMRS <=12) Response (50% decrease in YMRS) Efficacy YMRS CGI-BP-S MADRS Adverse events Extrapyramidal symptoms SAS AIMS BARS Withdrawal 23%
Calabrese, 2005¹⁶ RCT of responders US Government ROB Moderate 16263857	N=60 Mean Age 37 Female 52% White NR BP I 60% BP-II 40% Outpatient	Responders to both drugs Rapid cycling; mood episode in previous 3 months Substance Use Other Mental Health Conditions Pregnant/Nursing Lab/other conditions Intolerant of lithium	Divalproex/valproate mean divalproex dose 1571 mg/day; mean valproate dose 77	Lithium mean lithium dose 1359 mg/day	20 month	Time to relapse (HAM-D ≥20 or YMRS ≥20 for 8 weeks) Withdrawal 88% (Time to recurrence outcomes only included) Nonrelapse Withdrawal 25%
Bowden, 2000¹ RCT Multisite US Industry RoB High 10807488 12784116²	N=372 Mean Age 39 Female 51% White 94% BP I 100% Outpatient	No Episode at Randomization; Scores of MRS ≤ 11, DSS ≤ 13, GAS > 60; At least one other manic episode in past three years. Substance Abuse; Other Mental Health Conditions; Taking Other Medications; Pregnant/Nursing	Divalproex 71–125 mcg/mL (Mean 84.8 mcg/mL) N= 187	C1: Placebo n= 92 C2: Lithium 0.8–1.2 mEq/L (Mean 0.9 mEq/L) n= 90	52 week	Time to recurrence, any Time to recurrence, mania (MRS≥16) Time to recurrence, depression (DSS≥25) Withdrawal 69% (Time to recurrence outcomes only included) Nonrelapse Withdrawal 40%
Amsterdam, 2010³ RCT Single-site US Gov’t+nonprofit RoB Moderate 20360317	N =81 Mean Age 38 Female 52% White NR BP II 100% Outpatient	Depression and HAM-D≥16 initialy; then HAM-D≤8 after 12 weeks of initial Fluoxetine therapy at 20–80mg/day) Substance abuse Neurological Disorders Taking other medications Pregnant/NursingLabs/Other Conditions	Fluoxetine (n=28) 10–40 mg/day (mean 34.3 mg/day) N=28	C1: Placebo (n=27) C2: Lithium (n=26) 300–1200 mg/day; 0.5–1.5 mmol/L (mean 1027 mg/day; 0.69 mmol/L)	50 weeks	YMRS Relapse (HAM-D ≥14) Adverse Events Lab Values Withdrawal 72% (Time to recurrence outcomes only included) Nonrelapse Withdrawal 25%
Bowden, 2003⁴ RCT of responders Multisite 2 continents Industry RoB Low 12695317	N=175 Mean Age 41 Female 47% Race NR BP I 100% Outpatient	Lamotrigine responders (CGI-S≤3 for at least 4 continuous weeks) For open label period: Manic; DSM-IV Criteria for Mania or Hypomania currently or within past 60 days with previous episodes in past 3 years Other Mental Health Conditions	Lamotrigine 100–400 mg/day (Mean NR) N=59	C1: Placebo N=70 C2: Lithium 0.8–1.1 mEq/L Mean NR N=59	18 Months	Time to intervention for mania, hypomania, mixed, depression, and any mood episodes Time to early discontinuation Adverse Events l Withdrawal 80% (Time to recurrence outcomes only included) Nonrelapse Withdrawal 20%
Calabrese, 2003⁵ RCT of responders Multisite 4 Continents Industry RoB Moderate 14628976	N=410 Mean Age 43 Female 56% Race NR BP I 100% Outpatient	Lamotrigine responders (CGI-S<3 for at least 4 continuous weeks) For open label period: depression; DSM-IV criteria for depression currently or within past 60 days with previous depression and mania episodes in past 3 years. Other Mental Health Conditions	Lamotrigine 50, 200, or 400 mg/day (in analysis 50mg group was censored, 200 and 400 mg/day groups were combined) N=221	C1: Placebo N=121 C2: Lithium 0.8–1.1 mEq/L (Mean 0.8 mEq/L) N=121	18 Months	Time to intervention for mania, hypomania, depression, any mood episode Time to early discontinuation Adverse Events Withdrawal 84.9% (Time to recurrence outcomes only included) Nonrelapse Withdrawal 34%
Calabrese, 2000¹⁷ RCT of responders Multisite US, Canada Industry ROB High 11105737	N=182 Mean Age 38 Female 58% Race NR BP I 70% BP-II 30% Outpatient	Rapid cyclers, stabilized on lamotrigine (no mood episodes requiring other drugs or ECT) Other Mental Health Conditions Labs/Other conditions	Lamotrigine 100–500 mg/day N=93	Placebo N=89	26 weeks	Time to addition drug treatment Time to overall withdrawa Withdrawal 67% (Time to recurrence outcomes only included) Nonrelapse withdrawal 15%
Balance Investigators, 2010¹⁸ RCT Multisite US and Europe Industry RoB Moderate 20092882	N=330 Mean Age 43 Female 49% Race NR BP I 100% Outpatient	Not having acute episode; Not defined Pregnant/Nursing	Lithium 0.4–1.0 mmol/L mean NR + Valproate 750–1250 or highest tolerated mean NR n=110	C1: Lithium 0.4–1.0 mmol/L mean NR N=110 C2: Valproate 750–1250 or highest tolerated mean NR n=110	24 months	GAF EuroQol (EQ-5D) (quality of life) Relapse Withdrawal 20%
Tohen, 2006¹⁹ RCT Multisite 2 Continents Industry RoB Moderate 16449478	N= 361 Mean Age 41 Female 39% White 87% BP I 100% Outpatient	Remission; YMRS ≤ 12 HAM-D ≤ 8 First Manic Episode	Olanzapine 5–20 mg/day	Placebo	48 weeks	Time to Relapse Symptom Severity YMRS HAM-D Adverse Events Withdrawal 84% (Time to recurrence outcomes only included) Nonrelapse Withdrawal 25%
Tohen, 2003²⁰ RCT Multisite US Industry ROB High 12832240 Extension of Tohen, 2002b²¹ 12042191	N=251 Mean Age 40 Female 57% White 82% BP-I 100% Outpatient	YMRS >19 Substance Use Pregnant/Nursing Labs/other conditions	Olanzapine 5–20 mg/day	Divalproex 500–2500 mg/day	42 weeks	Time to relapse Withdrawal 84% (Time to relapse only included) Nonrelapse withdrawal unclear
Quiroz, 2010²² RCT of Responders Multisite 3 Continents Industry ROB Moderate 20227682	N=303 Mean Age 39 Female 49% White 80% BP-I 100% Outpatient	Responders to Phase III: stable at CGI-BP-S <3 Substand abuse Taking other meds Pregnant/nursing Rapid cycling Other mental health Labs/other conditions	Risperidone long-acting injectable 25, 37.5, or 50 mg; 77% received 25mg/2 weeks N=154	Placebo N=149	24 months	Time to recurrence of mood episode Time to: elevated mood depressive early discontinue due to medications Efficacy YMRS MADRS CGI-S Withdrawal 66% (Time to recurrence outcomes only) Nonrelapse withdrawal 25%
Vieta, 2012²³ RCT of Responders Multisite 3 Continents Industry RoB High 22503488	N = 398 Mean Age 36 Female 52% White 45% BP 1 100% Outpatient	No recurrence event (Responders from Phase II): Not hypomanic, manic, mixed, or depressive episode; treatment with a mood stabilizer, antidepressant or prohibited antipsychotic and benzodiazepine usage; hospitalization for a mood episode; or CGI-S≥4 with either YMRS>12 or MADRS>12 First Manic Episode Schizoaffective Other Mental Health Taking Other Meds Pregnant/Nursing	Risperidone long-acting injectable 1–6mg/day (fixed dose distribution provided; 25mg/2 weeks 64%, 37.5mg/2weeks 32%; 50mg/2 weeks 4%) N=131	C1: Placebo N=133 C2: Olanzapine 10mg/day N=130	18 months	Response (YMRS ≤ 19) Time to first recurrence of mood symptoms Efficacy YMRS CGI-S MADRS Adverse events Extrapyramidal symptoms ESRS Withdrawal 58% (Time to recurrence outcomes only included) Nonrelapse withdrawal 29%
Tohen, 2005²⁴ RCT Multisite 5 Continents Industry RoB Moderate 15994710	N= 431 Mean Age 42 Female 53% White 99% BP NR Outpatient	Met remission criteria: including YMRS ≤ 15 and HAM-D ≤ 8 After open-label period: Manic or Mixed Episode YMRS ≥ 20 Substance Abuse Other Mental Health Neurological Disorders Taking Other Meds Labs/Other Conditions	Olanzapine 15 mg/day flexible dosing 5–20 mg/day	Lithium 600 mg/day flexible dosing for blood level 0.6–1.2 meq/liter	52 weeks	Time to Episode Adverse Events Extrapyramidal Symptoms Withdrawal 60% (Time to recurrence outcomes only included) Nonrelapse Withdrawal 30%
Berwaerts, 2012²⁵ RCT Multisite 5 Continents Industry RoB High 22377512	N = 383 Mean Age 40 Female 53% White 60% BP I 100% Outpatient	Remission; YMRS and MADRS ≤12 for last three weeks of acute and continuation treatment study phases First manic episode Schizoaffective Substance abuse Other mental health Neurological disorders Labs/other conditions	Paliperidone EX 3–12 mg/day (n=152)	C1: Placebo (n=148) C2: Olanzapine 5–20 mg/day – few usable outcomes N=83	Up to 3 years (until recurrence)	YMRS Relapse (HAM-D ≥14) Adverse Events Lab Values Withdrawal 38%
Amsterdam, 2015²⁶ RCT extension of responders Single Site US Government RoB High 26143402	N = 55 Mean Age 42 Female 54% White 17% BP II 100% Outpatient	Responders to RCT phase: >50% reduction in baseline HAM-D + CGI-BP-S <3 Substance abuse Neurological disorders Taking other meds Pregnant/nursing Labs/other conditions	Venlafaxine 75–375 mg/day (mean 253.9 mg/day)	Lithium 300–1200 mg/day (Serum level of 0.8–1.5 mEq/L)	6 months	Relapse (HAM-D >14+CGI>3 for at least 14 days) Relapse hazard Time to relapse YMRS Withdrawal 18% (after 43% attrition from acute treatment)
Newport, 2008²⁷ Observational Single Site 1 Continent Government/Nonprofit High 18402631	N=26 Mean Age NR Female 100% White 91% BP I 73% BP-II 23% BP-NOS 4% Outpatient	Euthymic at conception Labs/Other Conditions	Lamotrigine Average 252 mg/day	Mood Stabilizer Discontinuation Initial doses Divalproex: 1200 mg/day average Lithium: 825 mg/day average Lamotrigine: 190 mg/day average	40 weeks	Survival Time Withdrawal NR
Weisler, 2011⁷ RCT Multisite 5 continents Industry RoB Moderate 22054050	N=1226 Mean Age 40 Female 53% White 63% BP 1 100% Outpatient	Meeting stability criteria of YMRS ≤ 12 and MADRS ≤ 12 after last episode of depression/mania/mixed episode at study entry or within past two years Substance Abuse Other Mental Health Conditions Pregnant/Nursing Labs/Other Conditions	Quetiapine 300–800 mg/day (543 mg/day mean) N=404	C1: Placebo N=404 C2:Lithium 600–1800 mg/day 0.6–1.2 mEq/L (0.63 mEq/L mean) N=364	104 weeks	Time to recurrence any mood (algorithm) Time to manic event Time to depressive event Time to all-cause discontinuation SDS MOS-Cog CGI-BP PANNS-P WPAI TMT Withdrawal 55% (Time to recurrence outcomes only included) Nonrelapse withdrawal 22%

: Abbreviations: AIMS=Abnormal Involuntary Movement Scale; BARS=Barnes Akathisia Scale; BIS-11=Barratt Impulsiveness Scale; BMI=Body Mass Index; BP=bipolar disorder; C=Comparison; CGI-BP=Clinical Global Impressions Scale for Bipolar Disorder; CGI-BP-M=Clinical Global Impressions Scale-Bipolar-modified (for long-term follow-up); CGI-BP-S=Clinical Global Impressions, Bipolar, Severity Scale; CGI-EI=Clinical Global Impressions-Efficacy Index; CGI-I=Clinical Global Impressions Scale, Improvement; CGI-S=Clinical Global Impressions, Severity Scale; DSM=Diagnostic and Statistical Manual of Mental Disorders; DSS=Depressive Syndrome Scale; EPS=extrapyramidal symptoms; ESRS=Extrapyramidal Symptom Rating Scale; GAF=General Assessment of Functioning Scale; GAS=Global Assessment Scale; HAM-D=Hamilton Scale for Depression; IMPS=Inpatient Multidimensional Psychiatric Scale; KAS=Katz Adjustment Scale; LIFE-RIFT= Longitudinal Interval Follow-up Evaluation-Range of Impaired Functioning Tool; MADRS=Montgomery-Asberg Depression Rating Scale; MOS-Cog=Medical Outcomes Study Cognitive Scale; MRS=Mania Rating Scale; NOS=not otherwise specified; NR=not reported; PANSS=Positive and Negative Syndrome Scale; PGWB=Psychological General Well-being Scale; PMID=PubMed Identification Number; PSQI=Pittsburgh Sleep Quality Index; QIDS-SR=Quick Inventory of Depressive Symptomatology (Self-reported); RCT=randomized controlled trial; ROB=risk of bias; SAS=Simpson Angus Scale; SDS=Sheehan Disability Scale; TMT=Trail Making Test; WPAI=Work Productivity and Activity Impairment Questionnaire; YMRS =Young Mania Rating Scale

Appendix Table I6Summary risk of bias assessments: other monotherapy

Drug	Study Funding Source PMID	Overall Risk of Bias Assessment	Rationale
Long-activing injectable Aripiprazole vs. placebo	Calabrese, 2017⁸ Industry 28146613	Moderate	Allocation concealment unclear. Differential attrition between arms. 61.7% attrition rate, 29% not due to relapse. Sensitivity testing for informative withdrawal
Aripiprazole vs. placebo	Keck, 2006⁹ Industry 16669728	Moderate	Randomization not described. Allocation concealment not described. Blinding of patients, providers, outcome assessors not described. Attrition 58% and differential drop-out.
Olanzapine vs. placebo	Tohen, 2006¹⁹ Industry 16449478	Moderate	Differential withdrawal rates (32% olan, 13% plac) and high dropout of olanzapine group may bias results.
Olanzapine vs. placebo vs. risperidone	Vieta, 2012²³ Industry 22503488	High	High - blinding and randomization procedures not well described. Period II results are biased by the drug assignment being open label. Period three efficacy scores are likely to be biased by the large attrition rate.
Risperidone vs. placebo	Quiroz, 2010²² Industry 20227682	Moderate (High for log-ramk test)	Randomization and blinding well described. Nonrelapse withdrawal 26%.
Lamotrigine vs. Lithium vs. placebo	Bowden, 2003⁴ Industry 12695317	Moderate (High for log rank test)	Randomization and blinding procedures not described. Balanced traits among groups at baseline. Efficacy data may be biased by dropout as endpoints are LOCF and differential nonrelapse dropout rates that range from 16–25% of patients for each arm.
Lamotrigine vs. Lithium vs. placebo	Calabrese, 2003⁵ Industry 14628976	Moderate (High for log rank test)	Randomization and allocation concealment not described. Balanced traits among groups at baseline. Efficacy data may be biased by dropout as endpoints are LOCF. 34% withdrawal.
Lamotrigine vs. placebo (rapid cyclers)	Calabrese, 2000¹⁷ Industry 11105737	Moderate (High for log rank test)	Randomization and allocation concealment not described. Balanced traits among groups at baseline. ITT using LOCF. Log rank test. 67% attrition, 15% nont related to relapse.
Fluoxetine vs. lithium vs. placebo	Amsterdam, 2010³ Government/Nonprofit 20360317	Moderate	Randomization may not have been successful as it relates to a person’s likelihood of relapse; randomization, allocation, and blinding procedures are underdescribed. 23% dropout.
Divalproex vs. lithium (rapid cyclers)	Calabrese, 2005¹⁶ Government 16263857	Moderate (High for log rank test)	Randomization and allocation not described. Balanced traits among groups at baseline. ITT. Log rank test. 25% nonrelapse attrition
Divalproex vs. lithium vs. placebo	Bowden, 2000¹ Industry 10807488 12784116²	High	Appears to be unblinded. Randomization not described. Nonrelapse dropout of 40%.
Carbamazepine vs. lithium	Hartong, 2003¹⁰ Industry/Government 12633122	Low	No sources of bias identified.
Carbamazepine vs. lithium	Greil, 1997¹¹ Government 9165384 9864077¹² 10529070¹³ 10529071¹⁴ 11093063¹⁵	High	Study is unblinded, there may be underlying differences between the groups in their likelihood to recur because of differences baseline disease history. Dropout is inconsistent between groups and no explanation is provided for why patients dropped.
Lithium + valproate vs. valproate vs. lithium	Balance Investigators, 2010¹⁸ Industry 20092882	Moderate	Open label. Intention To Treat used, but handling of dropouts/missing data not described.
Olanzapine vs. divalproex	Tohen, 2003²⁰ Industry 12832240 Extension of Tohen, 2002b²¹ 12042191	High	Randomization and blinding procedures not described. Log rank test. Attrition 84%, unclear nonrelapse withdrawal.
Olanzapine vs. lithium	Tohen, 2005²⁴ Industry 15994710	High	Well-constructed and reported study with high attrition rate (61%). Time to recurrence only includable outcome (log rank test).
Lamotrigine vs. discontinue mood stabilizers	Newport, 2008²⁷ Gov’t + nonprofit 18402631	High	Patients chose treatment assignment. Initial differences at baseline are noted.
Venlafaxine vs. placebo	Amsterdam, 2015²⁶ Government 26143402	High (RCT extension of responders)	RCT extension. Did report baseline at maintenance phase; appeared balanced on measured variables. 43% loss from initial randomization, further 18% at maintenance
Paliperidone vs. placebo vs. olanzapine	Berwaerts, 2012²⁵ Industry 22377512	High (RCT Extension)	Large dropout rate among all study arms, across all time periods Did report baseline at maintenance phase.
Quetiapine vs. lithium vs. placebo + lithium	Weisler, 2011⁷ Industry 22054050	Moderate	Generally a well conducted and reported study, however some sources of bias present related to dropout rates. Overall 21% withdraw due to reasons unrelated to recurrence; however by week 16 more than 50% of the placebo group has dropped for all causes, as well as 40% of the lithium group and 25% of the quetiapine group)

: Abbreviations: BP=bipolar; LOCF=Last observation carried forward; PMID=PubMed Identification Number; RCT=randomized controlled trial

Appendix Table I7Outcomes summary: other monotherapy versus placebo for maintenance

Drug	Study Comparison PMID	Responder/Remitter	Symptom	Function	Other	AE
Long-acting injectable Aripiprazole vs. placebo	Calabrese, 2017⁸	Time to any recurrence 52 weeks HR 0.45 (95% CI 0.30, 0.68) Favors Aripiprazole Any relapse 52 weeks Aripiprazole 35/132 Placebo 68/133 Favors Aripiprazole p<0.0001 Manic relapse 52 weeks Favors Aripiprazole p<0.0001 Depression relapse 52 weeks NS	NA (Exclude for attrition = 62%)	NR	Weight gain > 7% Aripiprazole 23/132 Placebo 17/133	SAE > 1 patient Aripiprazole 0.8% Placebo 2.3% 1 death reported EPS Aripiprazole 36/132 Placebo 22/133
Aripiprazole vs. placebo	Keck, 2006⁹ 16669728	Time to relapse 26 weeks HR 0.52 (95% CI 0.30, 0.91) Favors Aripiprazole Time to manic relapse 26 weeks HR 0.31 (95% CI 0.12, 0.77) Favors Aripiprazole Time to depression relapse 26 weeks NS	NA (Exclude for attrition = 58%	NA (Exclude for attrition = 58%	NR	1 placebo group patient attempted suicide EPS “more frequently in the aripiprazole group” 7% Weight gain Aripiprazole 13% Placebo 0%
Divalproex vs. placebo	Bowden, 2000¹ 10807488 12784116²	Time to any recurrence 12 months NS Time to mania recurrence 12 months NS Time to depression recurrence 12 months NS	NA (Exclude for attrition = 69%)	NR	NR	SAE NR Tremor Divalproex 41% (77/187) Placebo 13% (12/94) Favors Placebo p<0.001 Akathisia Divalproex <1% (1/187) Placebo 1% (1/94) NS
Fluoxetine vs. placebo	Amsterdam, 2010³ 20360317	Time to Relapse to Depression 50 week Log rank Favors Fluoxetine p=0.03	NA (Exclude for attrition = 72%)	NR	NR	SAE No Events Akathisia Fluoxetine 1 event/28 patients
Lamotrigine vs. placebo	Bowden, 2003⁴ 12695317	Time to any recurrence 18 months Log rank Favors Lamotrigine p=0.02 Time to mania recurrence 18 months Log rank NS Time to depression recurrence 18 months Log rank Favors Lamotrigine p=0.002	NA (Exclude for attrition = 80%)	NA	NA	SAE 1 lamotrigine patient hospitzalized for rash 1 lamotrigine patient attempted suicide Suicidality per HAM-D NS between groups
Lamotrigine vs. placebo	Calabrese, 2003⁵ 14628976	Time to any recurrence 18 months Log rank Favors Lamotrigine p=0.03 Time to mania recurrence 18 months Log rank NS Time to depression recurrence 18 months Log rank Favors Lamotrigine p=0.047	NA (Exclude for attrition = 85%)	NA	NA	SAE 1 death in lamotrigine groups Suicidality per HAM-D NS between groups Tremor Lamotrigine 5% (9/169) Placebo 5% (6/121) Lamotrigine vs. Placebo NS
Lamotrigine vs. placebo (rapid cyclers)	Calabrese, 2000¹⁷ Industry 11105737	Time to new drug 26 weeks Log rank No difference by group, or stratified by bipolar type	NA (Exclude for attrition = 68%)	NA	Time to overall withdrawal No difference by group. BPII Lamotrigine group more like remain in study p=0.015	SAE Lamotrigine: 1 tachycardia Placebo: 1 basal cell carcinoma, 1benign skull tumor None reporte related to treatment
Lamotrigine vs. discontinued mood stabilizers	Newport, 2008²⁷ 18402631	Relapse 40 weeks Lamotrigine 3/10 Discontinued 16/16 Favors Lamotrigine	NR	NR	NR	NR
Olanzapine vs. placebo	Tohen, 2006¹⁹ Moderate 16449478	Time to Relapse, Any Mood Episode (Median) 48 weeks Favors Olanzapine HR 2.67 (95% CI 2.03, 3.50), p<0.001 Time to relapse, Mania (25th percentile) Favors Olanzapine HR 3.90 (95% CI 2.40, 6.33), p<0.001 Time to relapse, depression (25th percentile) Favors Olanzapine HR 2.10, (95% CI 1.46, 3.02), p<0.001	NA (Exclude for attrition = 84%	NA	NA	SAE Olanzapine 3% (7/225) Placebo 7% (10/136) NS EPS Parkinsonism Olanzapine 2% (5/206) Placebo 0% (0/118) Favors Placebo Absolute Risk Reduction 0.02 (95% CI 0.003, 0.045) Akathisia Olanzapine 5% (9/194) Placebo 1% (1/119) NS Dyskinesia Olanzapine 0% (0/216) Placebo 1% (1/133) NS
	Vieta, 2012²³ 22503488	Time to first recurrence of mood symptoms 18 months Log-rank (by region) Favors Olanzapine p<0.001	NA (Due to attrition =58%)	NA (Due to attrition =58%)	NR	Reported no SAE in period III (maintenance phase)
	Berwaerts, 2012²⁵ 22377512	Time to Relapse to Any Mood Episode Favors Olanzapine Post hoc p<0.001 (YMRS≥15, CGI-BP-S for mania≥4; OR YMRS<15, MADRS≥16 and CGI-BP-S for depression≥4; hospitalization; therapeutic intervention or other clinically relevant indicators) Paliperidone Mean 558 days; Placebo Mean 283 days Time to Relapse to Manic Episode Favors Olanzapine Post hoc p<0.001	NR	NR	Death none Withdrawal for AE Paliperidone 5 (3%) Placebo 4 (3%) Withdrawal for Nonresponse Paliperidone 1 (1%) Placebo 2 (1%)	SAE up to 3 years Olanzapine 10% Placebo 22% EPS up to 3 years Olanzapine 1% Placebo 1%
Paliparidone vs. placebo	Berwaerts, 2012²⁵ 22377512	Time to Relapse to Any Mood Episode Favors Paliperidone HR 1.43 (95% CI 1.03,1.98) p=0.017 (YMRS≥15, CGI-BP-S for mania≥4; OR YMRS<15, MADRS≥16 and CGI-BP-S for depression≥4; hospitalization; therapeutic intervention or other clinically relevant indicators) Paliperidone Mean 558 days; Placebo Mean 283 days Time to Relapse to Manic Episode Favors Paliperidone HR 2.06 (95% CI 1.32,3.22) p<0.001 Paliperidone Mean 558 days; Placebo Mean 283 days	YMRS Change up to 3 years Favors Paliperidone LSM difference −4.5 (95% CI −6.92, −1.98) MADRS Change up to 3 years NS LSM difference 0.3 (95% CI −1.87,2.55)	NR	Death up to 3 years Paliperidone 2 (pneumonia, overdose) Placebo 0 Withdrawal for AE Olanzapine 7/83 (8%) Placebo 4/148 (3%) Withdrawal for Nonresponse Olanzapine 0/83 (1%) Placebo 2 /148(1%)	SAE up to 3 years Paliperidone 11% Placebo 22% EPS up to 3 years Paliperidone 1% Placebo 1%
Quetiapine vs. placebo	Weisler, 2011⁷ 22054050	Time to recurrence of any mood event Favors Quetiapine HR 0.29 (95% CI 0.23, 0.38), p<0.0001 Time to recurrence of manic event Favors Quetiapine HR 0.29 (95% CI 0.21, 0.40), p<0.0001 Time to recurrence of depression symptoms Favors Quetiapine HR 0.30 (95% CI 0.20, 0.44), p<0.0001	NA (Exclude for attrition 55%)	NR	NR	SAE 5 quetiapine 11 placebo EPS 16 quetiapine 18 placebo No deaths
Risperidone vs. placebo	Quiroz, 2010²² 20227682	Time to first recurrence of mood symptoms 24 months Log-rank test Favors Risperidone HR 0.40 (95% CI 0.27, 0.59) p=0.001	NA (Exclude for attrition =64%)	NA	Time to withdrawal for any reason 24 months Log-rank test Favors Risperidone HR 0.49 (95% CI 0.36, 0.67) p=0.001	SAE 1 diabetes mellitus in Risperidone group EPS 3%both groups
Risperidone vs. placebo	Vieta, 2012²³ 22503488	Time to first recurrence of mood symptoms 18 months Log-rank (by region) Favors Risperidone p=0.03	NA (Exclude for attrition =58%)	NA	NR	Reported no SAE in period III (maintenance phase)

: Abbreviations: AE=Adverse Events; CGI-BP-S=Clinical Global Impressions, Bipolar, Severity Scale; CI=Confidence Interval; EPS=extrapyramidal symptoms; HAM-D=Hamilton Scale for Depression; HR=Hazard Ratio; LSM=least-squares means; MADRS=Montgomery-Asberg Depression Rating Scale; NA=Not applicable; NR=not reported; NS=not significant; OR=Odds Ratio; PMID=PubMed Identification Number; SAE=Serious Adverse Events; YMRS = Young Mania Rating Scale

Appendix Table I8Strength of evidence assessment: other monotherapy versus placebo for maintenance

Comparison	Outcome	# Studies/Design (n analyzed)	Finding or Summary Statistic Outcome Timing	Study Limitations	Consistency	Directness	Precision	Overall Grade/Conclusion
Long-acting injectable Aripiprazole vs. placebo	Time to relapse 52 wks Relapse 52 wks	1 RCT (n=266)	See table above	Moderate	Unknown	Direct	Imprecise	Insufficient
Aripiprazole vs. placebo	Time to Relapse 26 wks	1 RCT (n=161)	See table above	Moderate	Unknown	Direct	Imprecise	Insufficient
Divalproex vs. placebo	Time to relapse 52 wks	1 RCT (n=281)	See table above	High	Unknown	Direct	Imprecise	Insufficient
Fluoxetine vs. placebo	Time to relapse to depression 50 wks	1 RCT (n=55)	See table above	Moderate	Unknown	Direct	Imprecise	Insufficient
Lamotrigine vs. placebo	Time to relapse 18 mths	2 RCT (n=471)	See table above	High (log rank tests)	Consistent	Direct	Imprecise	Insufficient
Lamotrigine vs. placebo (rapid cyclers)	Time to new drug treatment	1 RCT (n=182)	See table above	High	Unknown	Direct	Imprecise	Insufficient
Lamotrigine vs. discontinue mood stabilizers	Relapse 40 wks	1 Observational (n=26)	See table above	High	Unknown	Direct	Imprecise	Insufficient
Olanzapine vs. placebo	Time to relapse 18 mths to 3 yrs	3 RCT (n=855)	18 months to 3 years	High	Consistent	Direct	Imprecise	Insufficient
Paliperidone vs. placebo	Time to Relapse 3 yrs YMRS 3 yrs MADRS	1 RCT extension (n=300)	See table above	High	Unknown	Direct	Imprecise	Insufficient
Quetiapine vs. placebo	Time to relapse 104 wks	1 RCT (n=808)	See table above	Moderate	Unknown	Direct	Imprecise	Insufficient
Risperidone vs. placebo	Time to relapse 52 wks	2 RCT (n=353)	See table above	High (log rank test)	Consistent	Direct	Imprecise	Insufficient

: Abbreviations: MADRS=Montgomery-Asberg Depression Rating Scale; RCT=randomized controlled trial; YMRS = Young Mania Rating Scale
: Notes:
1: Publication bias for antipsychotics, antidepressants, and behavioral interventions for depressive disorders is suspected.
2: Data were generally imprecise due to missing data from high attrition rates, which was commonly dealt with by Last Observation Carried Forward (LOCF). LOCF requires an assumption that the health status of patients who dropped out of the trial would not have changed had future observations been recorded, a strong assumption in the context of bipolar disorder research.

Appendix Table I9Outcomes summary: other monotherapy versus active control for maintenance

Drug	Study Comparison PMID	Responder/Remitter	Symptom	Function	Other	AE
Carbamazepine vs. lithium	Hartong, 2003¹⁰ 12633122	Time to relapse Proportional hazard assumption did not hold	NR	NR	NR	SAE NR
Carbamazepine vs. lithium	Greil, 1999 10529070¹³ 10529071¹⁴	Time to clinical or subclinical recurrence BP-I 2.5 years Favors lithium p=0.034 n=114 Time to clinical or subclinical recurrence BP-II or NOS 2.5 years NS n=57	NR	NR	NR	1 suicide, 1 attempted suicide in carbazamepine group
Divalproex vs. lithium (rapid cyclers)	Calabrese, 2005¹⁶ Government 16263857	Time to treatment for mood episode, depression treatment, elevated mood treatment 26 weeks No differences between groups	NA (attrition 88%)	NA	Time to overall withdrawal 26 weeks No difference between groups	SAE NR Tremors/polyuria/polydipsia “more common in those assigned to lithium”
Paliparidone vs. olanzapine	Berwaerts, 2012²⁵ 22377512	No usable outcomes
Fluoxetine vs. Lithium	Amsterdam, 2010³ 20360317	Time to Relapse to Depression 50 week Favors Fluoxetine p=0.03 Relapse 50 week Favors Fluoxetine HR=0.04 (95%CI 0.2,0.9)	YMRS Change 50 week Fluoxetine −6.3 (95%CI −47.5, 34.9) Lithium 7.2 (95%CI −33.3,53.8)	NR	Withdrawal for AE Fluoxetine 1 Lithium 1	SAE No Events
Lamotrigine vs. Lithium	Bowden, 2003⁴ 12695317	Time to any recurrence 18 months Log rank NS Time to mania recurrence 18 months Log rank NS Time to depression recurrence 18 months Log rank NS	NA (Exclude for attrition = 80%)	NA	NA	SAE 1 lamotrigine patient hospitzalized for rash 1 lamotrigine patient attempted suicide Suicidality per HAM-D NS between groups
Lamotrigine vs. Lithium	Calabrese, 2003⁵ 14628976	Time to any recurrence 18 months Log rank NS Time to mania recurrence 18 months Log rank NS Time to depression recurrence 18 months Log rank NS	NA (Exclude for attrition = 85%)	NA	NA	SAE 1 death in lamotrigine groups Suicidality per HAM-D NS between groups
Lithium vs. valproate	Balance Investigators, 2010¹⁸ 20092882	Time to new intervention for emerging mood episode 24 months Hazard ratio Favors Lithium HR 0.71 (0.51,1.00) p=0.047	NR	GAF NS	EuroQol (EQ-5D) (quality of life) NS Overall Withdrawal Lithium: 23/110 Valproate: 23/110 Withdrawal lack of efficacy NR Withdrawal adverse events Lithium: 6/110 Valproate: 4/110	SAE NS Valproate: 7 SAE including 3 deaths Lithium: 5 SAE including 2 deaths
Lithium vs. divalproex	Bowden, 2000¹ 10807488 12784116²	Time to any recurrence 12 months NS Time to mania recurrence 12 months NS Time to depression recurrence 12 months NS	NA (Exclude for attrition = 69%)	NR	NR	SAE NR Tremor Divalproex 41% (77/187) Lithium 42% (38/94) Divalproex vs. Lithium NS Akathisia Divalproex <1% (1/187) Lithium 4% (4/94) Divalproex vs. Lithium Favors divalproex p=0.04
Olanzapine vs. divalproex	Tohen, 2003²⁰ Industry 12832240 Extension of Tohen, 2002b²¹ 12042191	Time to relapse 42 weeks Log rank NS	NA (attrition 84%)	NA	NR	SAE not reported
Olanzapine vs. Lithium	Tohen, 2005²⁴ High 15994710	Time to Relapse (YMRS and/or HAM-D>15) 52 weeks Log rank NS p=0.07 Time to hospitalization 52 weeks Log rank Favors Olanzapine p=0.01	NA (Exclude for attrition = 61%)	NR	Overall Withdrawal Olanzapine: 116/217 Lithium: 144/214	SAE 2 deaths, lithium, 1 suicie, 1 accident EPS NS
Quetiapine vs. Lithium	Weisler, 2011⁷ 22054050	Time to recurrence of any mood event Favors Quetiapine HR 0.66 (95% CI 0.49, 0.88), p=0.005 Time to recurrence of manic event NS Time to recurrence of depression symptoms Favors Quetiapine HR 0.54 (95% CI 0.35, 0.84), p=0.006	NA (Exclude for attrition 55%)	NR	NR	SAE 5 quetiapine 10 lithium EPS 16 quetiapine 38 lithium No deaths
Venlafaxine vs. Lithium	Amsterdam, 2015²⁶ 26143402	Time to depression relapse 6 months Log rank NS	YMRS NS	NR	Withdrawal for AE Venlaxafine 1 Lithium 0	NR
Risperidone vs. olanzapine	Vieta, 2012²³ 22503488	Time to first recurrence of mood symptoms 18 months Post-hoc Log-rank Favors Olanzapine p=0.001	NA (Due to attrition =58%)	NA (Due to attrition =58%)	NR	Reported no SAE in period III (maintenance phase)

: Abbreviations: AE=Adverse Events; BP=bipolar disorder; CI=Confidence Interval; EPS=extrapyramidal symptoms; GAF=General Assessment of Functioning Scale; HAM-D=Hamilton Scale for Depression; HR=Hazard Ratio; NA= Not Applicable; NOS=Not Otherwise Specified; NR=not reported; NS=not significant; PMID=PubMed Identification Number; SAE=Serious Adverse Events; YMRS = Young Mania Rating Scale

Appendix Table I10Strength of evidence assessment: other monotherapy versus active control for maintenance

Comparison	Outcome	# Studies/Design (n analyzed)	Finding or Summary Statistic Outcome Timing	Study Limitations	Consistency	Directness	Precision	Overall Grade/Conclusion
Carbamazepine vs. Lithium	Time to recurrence 2.5 yrs	1 RCT (Greil 1999) (n=171)	See table above	Moderate	Unknown	Direct	Imprecise	Insufficient
Divalproex vs. lithium (rapid cyclers)	Time to treatment for mood episode 26 weeks	1 RCT (n=60)	See table above	High	Unknown	Direct	Imprecise	Insufficient
Fluoxetine vs. Lithium	Time to relapse to depression 50 wks	1 RCT (n=54)	See table above	Moderate	Unknown	Direct	Imprecise	Insufficient
Lamotrigine vs. Lithium	Time to recurrence 18 mths	2 RCTs (n=390)	See table above	High	Consistent	Direct	Imprecise	Insufficient
Lithium vs. Valproate^*	Time to new intervention for emerging mood episode 24 mths EuroQoL 24 mths	1 RCT (n=220)	See table above	Moderate	Unknown	Direct	Imprecise	Insufficient
Lithium vs. divalproex^*	Time to recurrence 12 mths	1 RCT (n=372)	See table above	Moderate	Unknown	Direct	Imprecise	Insufficient
Olanzapine vs. divalproex	Time to recurrence 47 weeks	1 RCT (n=251)	See table above	High	Unknown	Direct	Imprecise	Insufficient
Olanzapine vs. Lithium	Time to relapse 104 wks time to hospitalization 104 wks	1 RCT (n=431)	See table above	High	Consistent	Direct	Imprecise	Insufficient
Quetiapine vs. Lithium	Time to relapse 104 wks	1 RCT (n=768)	See table above	Moderate	Unknown	Direct	Imprecise	Insufficient
Paliperidone vs. Olanzapine	No usable outcomes 3 yrs	1 RCT extension (n=235)	See table above	High	Unknown	Direct	Imprecise	Insufficient
Venlafaxine vs. Lithium	Time to Relapse 6 mths YMRS 6 mths	1 RCT extension (n=55)	See table above	High	Unknown	Direct	Imprecise	Insufficient
Risperidone vs. Olanzapine	Time to relapse 18 mths	1 RCT (n=263)	See table above	High	Unknown	Direct	Imprecise	Insufficient

*: If aggregating across lithium versus divalproex or valproate, strength of evidence remains insufficient due to inconsistency between study findings.
: Abbreviations: RCT=randomized controlled trial; YMRS = Young Mania Rating Scale
: Notes:
1: Publication bias for antipsychotics, antidepressants, and behavioral interventions for depressive disorders is suspected.
2: Data were generally imprecise due to missing data from high attrition rates, which was commonly dealt with by Last Observation Carried Forward (LOCF). LOCF requires an assumption that the health status of patients who dropped out of the trial would not have changed had future observations been recorded, a strong assumption in the context of bipolar disorder research.

Section 3. Combination Therapy

Appendix Table I11Characteristics of eligible studies: combination therapy for maintenance

Study, Year Design Location Funder Risk of Bias PMID	# Randomized Age (mean) Sex (% Female) Race (% White) Diagnosis (% BP I, II, NOS) Setting	Inclusions Key Exclusions	Intervention Dosage	Comparison Dosage	Followup Duration	Outcomes Reported Withdrawal (%) at endpoint
Balance Investigators, 2010¹⁸ RCT Multisite US and Europe Industry RoB Moderate 20092882	N=330 Mean Age 43 Female 49% Race NR BP I 100% Outpatient	Not having acute episode; Not defined Pregnant/Nursing	Lithium 0.4–1.0 mmol/L mean NR + Valproate 750–1250 or highest tolerated mean NR n=110	C1: Lithium 0.4–1.0 mmol/L mean NR N=110 C2: Valproate 750–1250 or highest tolerated mean NR n=110	24 months	GAF EuroQol (EQ-5D) (quality of life) Relapse Withdrawal 20%
Marcus, 2011²⁸ RCT Multisite/Not Disclosed (8 countries) Industry RoB High 21443567	N = 337 Mean Age 39 Female 55% White 68% BP I 100% Outpatient	Manic/Mixed; YMRS≥16 at study entry; Current episode <2 years; YMRS≤12, MADRS≤12 at randomization after 12 weeks of stabilization treatment First manic episode Schizoaffective Substance abuse Other mental health Neurological disorders Taking other meds Labs/other conditions	Aripiprazole 10–30 mg/day (15.8–16.9 mg/day) + Lithium or valproic acid	Placebo + Lithium or valproic acid	52 weeks	YMRS MADRS CGI-BP-S Relapse Adverse Events EPS Withdrawal 43%
Carlson, 2012²⁹ RCT of responders Multisite US Industry RoB High 22329471	N=351 Mean Age 39 Female 65% White 90% BP I 100% Outpatient	Randomization after stabilization; 8 weeks at YMRS≤12, MADRS≤12. Study entry manic or mixed YMRS≥16 in previous 3 months with or without rapid cycling (4 to 7 mood episodes per year) Substance abuse Other Mental Health Conditions Pregnant/Nursing Labs/other conditions First manic episode Treatment refractory mania/mixed mania	Aripiprazole target 15 mg/day (range 10–30 mg/day) +Lamotrigine target 200 mg/day (range 100–200 mg/day)	Placebo +Lamotrigine target 200 mg/day (range 100–200 mg/day)	52 weeks	Time to relapse (hospitalization; SAE or lack of effect, including YMRS>14 and MADRS≤16 manic or YMRS≤14 and MADRS >16) YMRS MADRS CGI-BP Withdrawal 66% (Time to recurrence outcomes only included) Nonrelapse withdrawal ~40%
Woo, 2011³⁰ RCT Multisite/Asia Industry RoB High 22134973	N = 83 Mean Age 38 Female 68% White NR Japanese (32%) Korean/Chinese (43%) Other (25%) BP I 100% Outpatient	Manic/Mixed initially with YMRS≥20 at study entry; then YMRS≤12, MADRS≤13 at randomization after 6 weeks of stabilization treatment Schizoaffective Substance abuse Neurological disorders Taking other meds Pregnant/nursing Labs/other conditions	Aripiprazole 10–30 mg/day (17.9 mg/day) + Lithium or valproic acid	Placebo + Lithium or valproic acid	24 weeks	YMRS MADRS CGI-BP-S Relapse Adverse Events EPS Withdrawal 42%
Kemp, 2009³¹ RCT of responders Single site US Gov’t RoB High 19192457	N=31 Mean Age 36 Female 36% White 82% BP I 75% BP II 25% Outpatient	Stable responders (HAM-D score ≤ 20, YMRS score ≤ 12.5) Rapid cycling, substance use disorder as ascertained by structured interview; mood episode in previous 3 months Labs/other conditions Pregnant/nursing	Divalproate 250 mg/day (target blood level 50mg/day+ Lithium	Placebo + Lithium	26 weeks	Time to treat mood episode Efficacy YMRS HAM-D GAS Withdrawal 74% (Time to recurrence outcomes only included) Nonrelapse withdrawal 19%
Tohen, 2004³² RCT of Responders Multisite US Industry ROB High 15056579 extension of Tohen, 2002³³ 11779284	N = 99 Mean Age 41 Female 48% White 85% BP-I 100% Outpatient	Responders to olanzapine + lithium or valproate mania and depression no worse than mild; First Manic Episode Labs/Other Conditions	Olanzapine 10 mg/day with flexible dosing from 5–20 mg/day Adjunctive to ongoing open-label valproate or lithium n=51	Placebo Adjunctive to ongoing open-label valproate or lithium n=48	18 months	Time to any mood episosde; Withdrawal 78% (Time to recurrence outcomes only included) Unclear nonrelapse withdrawal
Suppes, 2009³⁴ RCT Multisite US/Canada Industry RoB High 19289454	N = 623 Mean Age 40 Female 53% White 82% BP I 100% Inpatient	Mania at entry; Stable at randomization after Lithium or Valproate; YMRS and MADRS ≤ 12 AND at least 1 mood episode of any type in past 2 years and another 6 months prior to randomization First Manic Episode Substance Abuse Other Mental Health Conditions Pregnant/Nursing	Quetiapine 400–800 mg/day (519 mg/day mean) + Valproate 50–125 mcg/mL target OR Lithium 0.5–1.2 mEq/L target	Placebo + Valproate 50–125 mcg/mL target OR Lithium 0.5–1.2 mEq/L target	104 weeks (only time to occurrence and withdrawals used due to attrition)	Recurrance Adverse Events Withdrawal 71% (Time to recurrence outcomes only included) Nonrelapse withdrawal 35%
Vieta, 2008³⁵ RCT Multisite 4 Continents Industry RoB 18579216	N = 706 Mean Age 42 Female 55% White 97% BP I 100% Inpatient (1 week) Outpatient (2–6 weeks, subject to inspector discretion)	Mania, Depression, Mixed; latest episode of any type within 26 weeks, achived clinical stability (YMRS and MADRS ≤ 12) prior to randomization, subject to specified time periods Substance Abuse; Other Mental Health Conditions; Taking Other Meds; Pregnant/Nursing	Quetiapine 400–800 mg/day (497 mg/day mean) + Valproate 50–125 mcg/mL target OR Lithium 0.5–1.2 mEq/L target	Placebo + Valproate 50–125 mcg/mL target OR Lithium 0.5–1.2 mEq/L target	104 weeks (only time to occurrence and withdrawals used due to attrition)	Recurrance CGI-BP PANSS-P SDS PGWB SAS BARS AIMS Withdrawal 51% (Time to recurrence outcomes only included) Nonrelapse withdrawal 16%
Bobo, 2011³⁶ RCT Single-site US Industry RoB High 22104634	N = 50 Mean Age 40 Female 67% White 67% BP I 73% BP II 27% Outpatient	Any Phase Schizoaffective; Other Mental Health Conditions; Pregnant/Nursing	Risperidone long-acting injectable 27 ± 10.4 mg every 2 weeks + Treatment as Usual	No Placebo + Treatment as Usual	52 weeks	AIMS BARS CGI-S MADRS Quick Inventory of depressive symptoms self-report (QIDS-SR) SAS YMRS Withdrawal 25%
Macfadden, 2009³⁷ RCT Multisite 2 Continents Industry RoB High 19922552	N = 124 Mean Age 39 Female 28% White 10% BP I 100% Inpatient and outpatient	Any Phase including euthymic; 4 or more mood episodes in past year Substance Abuse; Other Mental Health Conditions; Taking other Medications; Abnormal Lab Results	Risperidone long-acting injectable 25–30 mg every 2 weeks +Treatment as Usual	Placebo + Treatment As Usual	52 weeks	CGI-BP-C CGI-BP-S MADRS Relapse – Time to (DSM diagnosis for acute mood episode + other complicated criteria) YMRS Withdrawal 48%
Bowden, 2010³⁸ RCT of responders Multisite 3 Continents Industry RoB High 20122373 (also 22999893)	N = 240 Mean Age 39 Female 54% White 62% BP I 100% Outpatient	Mania; Initial inclusion: YMRS ≥ 14 with score ≥ 2 on at least four items at screening and admission. Extension inclusion: stabilized (CGI-I ≤ 3 at least 2 consecutive weeks Substance Abuse; Other Mental Health Condition; Pregnant/Nursing; Labs/Other Conditions	Ziprasidone (80–160 mg/day) + Lithium (0.6–1.2 mEq/L) or Valproate (50–125 mcg/mL)	Placebo+ Lithium (0.6–1.2 mEq/L) or Valproate (50–125 mcg/mL)	26 weeks	BMI or Weight MADRS Mania Rating Scale (MRS) Relapse – Relative Risk of Relapse - Time to intervention for mood episode Withdrawal 42%
Vieta, 2008³⁹ RCT Multisite Spain Industry RoB Moderate 18346292	N = 55 Mean Age 44 Female 65% White NR BP I 76% BP II 24% Outpatient	Euthymic; YMRS≤12; MADRS≤20; 2+ episodes in past year; ≥6 months in remission; Being treated with Lithium (≥0.6 meq/L) Substance Abuse; Other Mental Health Conditions; Taking Other Meds; Pregnant/Nursing	Oxcarbazepine 1200 mg/day+ Lithium 300–1200 mg/d (mean NR)	Placebo + Lithium 300–1200 mg/d (mean NR)	52 weeks	Relapse (DSM-IV criteria for manic, hypomanic, mixed or depressive episode; OR YMRS≥12; OR MADRS≥20) CGI-BP-M GAF BIS-11 Withdrawal 36%
Vieta, 2006⁴⁰ RCT Multisite Spain Industry RoB High 16649836	N = 25 Mean Age 49 Female 72% White NR BP I 76% BP II 24% Outpatient	Euthymic; CGI-BP-M ≥4; HAMD≤8 YMRS≤4; Treated with lithium for 6 weeks; Last episode within 6 mos; Substance abuse Pregnant/nursing Labs/other conditions	Gabapentin 300–800 mg/tid (400mg/tid) + Lithium and/or Valproate and/or Carbamazepine NR	Placebo + Lithium and/or Valproate and/or Carbamazepine NR	52 week	CGI-BP-M YMRS HAM-D PQSI Time to Relapse Withdrawal 48%
Zarate, 2004⁴¹ Single-Site RCT US Gov’t+nonprofit RoB High 14702269	N = 37 Mean Age 34 Female 78% White NR BP I 100% Setting NR	Manic /Mixed at study entry; DSM-IV criteria (Structured Clinical Interview) Euthymic by randomization (week 10); YMRS≤10; HAM-D≤10 Schizoaffective Substance abuse Other mental health Labs/other conditions	Perphenazine 4–64 mg/day + (Mood Stabilizers Lithium 0.6–1.2 meq/L And/Or Carbamazepine 4–12 mg/L And/Or Valproate 50–125 mg/L)	Placebo+ (Mood Stabilizers Lithium 0.6–1.2 meq/L And/Or Carbamazepine 4–12 mg/L And/Or Valproate 50–125 mg/L)	6 months	Relapse (Not Defined) HAM-D Withdrawal 35%
Nierenberg, 2016⁴² RCT Multisite US Government RoB High 26845264	N=482 Mean Age 39 Female 59% White 72% BP I 68% BP II, NOS NR Outpatient	Any status Pregnant/Nursing; Labs/Other Conditions	Quetiapine 150–900 mg/day (345 mg/day) + Adjunctive personalized treatment Texas Medication Algorithm	Lithium 0.6–1.2 mEq/L (0.6 mEg/L) + Adjunctive personalized treatment Texas Medication Algorithm	24 weeks	CGI-EI MADRS YMRS Columbia-Suicide Severity Rating Scale Withdrawal 25%
Vieta, 2010⁴³ Observational (Partial responders of earlier RCT) Multisite Not Disclosed Industry RoB High 20429835 (Continuation of 18381903)	N = 283 Mean Age 43 Female 53% White 93% BP I 100% Outpatient	Mania; Mania Rating Scale (Spitzer, 1978) ≥ 14 with score ≥ 2 on four items at screening and admission Other Mental Health Conditions; Substance Abuse	Valproate 500–2500 mg/day (1174.3 mg/day average - last 4 weeks) + Aripiprazole 15–30 mg/day (17.1–18.5 mg/day average)	Lithium 500–1500 mg/day (1105.5 mg/day average) + Aripiprazole 15–30 mg/day (16.9–18.4 mg/day average)	46 weeks	BMI or Weight LIFE-RIFT MADRS Relapse - Emergent Depression Incidence (MADRS total score >= 18 and >= 4 point increase in two consecutive assessments or last observation) Relapse (YMRS total score <=12 and MADRS total <=8 of patients who achieved remission at end of week 6) YMRS Withdraw 48%

: Abbreviations: AIMS=Abnormal Involuntary Movement Scale; BARS=Barnes Akathisia Scale; BIS-11=Barratt Impulsiveness Scale; BMI=Body Mass Index; BP=bipolar disorder; C=Comparison; CGI-BP=Clinical Global Impressions Scale for Bipolar Disorder; CGI-BP-M=Clinical Global Impressions Scale-Bipolar-modified (for long-term follow-up); CGI-BP-S=Clinical Global Impressions, Bipolar, Severity Scale; CGI-EI=Clinical Global Impressions-Efficacy Index; CGI-I=Clinical Global Impressions Scale, Improvement; CGI-S=Clinical Global Impressions, Severity Scale; DSM=Diagnostic and Statistical Manual of Mental Disorders; DSS=Depressive Syndrome Scale; EPS=extrapyramidal symptoms; ESRS=Extrapyramidal Symptom Rating Scale; GAF=General Assessment of Functioning Scale; GAS=Global Assessment Scale; HAM-D=Hamilton Scale for Depression; IMPS=Inpatient Multidimensional Psychiatric Scale; KAS=Katz Adjustment Scale; LIFE-RIFT= Longitudinal Interval Follow-up Evaluation-Range of Impaired Functioning Tool; MADRS=Montgomery-Asberg Depression Rating Scale; MOS-Cog=Medical Outcomes Study Cognitive Scale; MRS=Mania Rating Scale; NOS=not otherwise specified; NR=not reported; PANSS=Positive and Negative Syndrome Scale; PGWB=Psychological General Well-being Scale; PMID=PubMed Identification Number; PSQI=Pittsburgh Sleep Quality Index; QIDS-SR=Quick Inventory of Depressive Symptomatology (Self-reported); RCT=randomized controlled trial; ROB=risk of bias; SAS=Simpson Angus Scale; SDS=Sheehan Disability Scale; TMT=Trail Making Test; WPAI=Work Productivity and Activity Impairment Questionnaire; YMRS =Young Mania Rating Scale

Appendix Table I12Summary risk of bias assessments: oombination therapy for maintenance

Drug	Study Funding Source PMID	Overall Risk of Bias Assessment	Rationale
Lithium + valproate vs. valproate vs. lithium	Balance Investigators, 2010¹⁸ Industry 20092882	Moderate	Open label. Intention To Treat used, but handling of dropouts/missing data not described.
Aripiprazole + mood stabilizer vs. placebo + mood stabilizer	Marcus, 2011²⁸ Industry 21443567	High	High Withdrawal rate (43%) - maintenance study; blinding procedures not disclosed.
	Carlson, 2012²⁹ Industry 22329471	High	High nonrelapse withdrawal rate (40%). Overall attrition 66%. Randomization, concealment, and blinding not described.
	Woo, 2011³⁰ Industry 22134973	High	High Withdrawal rate (42%) – maintenance study; Randomization and blinding procedures not disclosed.
Divalproex + lithium vs. placebo + lithium alone	Kemp, 2009³¹ Government 19192457	High	Randomization and allocation not reported. Overall 19% withdraw due other than relapse.
Quetiapine + mood stabilizers vs. placebo + mood stabilizers	Suppes, 2009³⁴ Industry 19289454	High	Blinding not described; differential dropout rates. High drop-out rates overall.
	Vieta, 2008³⁵ Industry 18579216	High	Generally well reported with minor concerns related to blinding, subjective definition of recurrence. High drop-out rates overall.
Risperidone long-acting injectable+ treatment as usual vs. placebo + treatment as usual	Macfadden, 2009³⁷ Industry 19922552	High	BPII patients enrolled, but removed from analysis. 48% dropout overall. Large differential dropout with 58% placebo and 40% of treatment groups dropping.
Risperidone long-acting injectable+ treatment as usual vs. treatment as usual	Bobo, 2011³⁶ Industry 22104634	High	No blinding. Treatment As Usual not well controlled. Treatment regimes of the two groups prior to study were not tested for similarity and appear as though they may differ statistically. Results may be due to differences in TAU. 25% dropout.
Olanzapine + mood stabilizer vs. placebo + mood stabilizer	Tohen, 2004³² Industry 15056579 extension of Tohen, 2002³³ 11779284	Moderate (High for log rank)	Allocation concealment not described. Log rank test. Unclear nonrelapse withdrawal.
Oxcarbazepine lithium vs. placebo + lithium	Vieta, 2008³⁹ Industry 18346292	Moderate	Patients, staff, and raters may not be blinded; procedures not described
Gabapentin + mood stabilizers vs. placebo + mood stabilizers	Vieta, 2006⁴⁰ Industry 16649836	High	Distribution of BP I and BP II patients differs between treatment arms, creating a residual confounder.
Perphenazine + mood stabilizers placebo + mood stabilizers	Zarate, 2004⁴¹ Gov’t+nonprofit 14702269	High	Randomization and blinding procedures not disclosed; numerical results of several measured outcomes not presented; Relapse not defined
Ziprasidone + mood stabilizers vs. placebo + mood stabilizers	Bowden, 2010³⁸ Industry 20122373 (also 22999893)	High	Randomization and blinding procedures not described. 40%+ dropout.
Quetiapine + personalize treatment vs. Lithium + personalized treatment	Nierenberg, 2016⁴² Government 26845264	High	Does not report on adjunctive treatments received in results. Includes those who have ‘off-procedure’ treatment deviations in analysis, who are people that have taken antipsychotics. Also included are the roughly 30% of people in both treatment arms who have no adjunctive treatment. None of these is accounted for in analysis as a possible confounding influence on the underlying comparison of Quetiapine and Lithium.
Valproic acid + Aripiprazole vs. Lithium + Aripiprazole	Vieta, 2010⁴³ Industry 20429835 (Continuation of 18381903)	High	Non-Randomized continuation study of partial responders, no blinding, initial baseline measures of this group may not be similar, appears to be underpowered for the subgroup analysis that is presented. 48% dropout.

: Abbreviations: BP=bipolar; LOCF=Last observation carried forward; PMID=PubMed Identification Number; RCT=randomized controlled trial

Appendix Table I13Outcomes summary: combination therapy versus placebo for maintenance

Drug	Study RoB PMID	Responder/Remitter	Symptom	Function	Other	AE
Aripiprazole + mood stabilizer vs. placebo + mood stabilizer	Marcus,²⁸ 2011 High 21443567	Time to Relapse 52 weeks Hazard Ratio 0.54 (0.33, 0.89) Favors Aripiprazole	YMRS 52 weeks Mean change Aripiprazole: −0.1 Placebo: 2.9 p<0.001 Favors Aripiprazole MADRS 52 weeks Mean change Aripiprazole: 1.5 Placebo: 2.5 p=0.02 Favors Aripiprazole	CGI-BP-S 52 weeks Ari n=162 Plc n=164 Mean change difference −0.3 (−0.62, −0.07) Favors Aripiprazole Less than MID=1	Weight gain >7% NS Overall Withdrawal Aripiprazole 65/168 (38.7%) Placebo 80/169 (47.3%) NS Withdrawal lack of efficacy Aripiprazole 6/168 (8.3%) Placebo 31/169 (18.3%) p=0.007 Withdrawal for AE Aripiprazole 19/168 11.3% Placebo 15/169 (8.9%) NS	2 deaths, 1 in each arm; 1 suicide day 83 deemed not due to treatment 1 tardive dyskinesia (placebo group) SAE Aripiprazole: 11 (6.6%) Placebo: 8 (4.8%) NS
	Woo, 2011³⁰ High 22134973	Time to Relapse 6 months NS	YMRS 6 months NS MADRS 6 months NS	CGI-CP-S 6 months NS	Weight gain >7% NS Overall Withdrawal Aripiprazole 17/40 (38.7%) Placebo 18/43 (41.9%) NS Withdrawal lack of efficacy Aripiprazole 6/40 (15%) Placebo 8/43 (18.6%) NS Withdrawal for AE Aripiprazole 0% Placebo 9.3% p=0.049	SAE Aripiprazole: 5% Placebo: 11% Included 1 suicide EPS No discontinuation in either group
	Carlson, 2012²⁹ High 22329471	Time to Relapse 52 weeks Hazard Ratio 0.67 (0.45, 1.00) NS	NA (Attrition 66%)	NA (Attrition 66%)	Withdraw for AE Aripiprazole 14/176 (8%) Placebo 12/165 (7.3%)	SAE Aripiprazole 5/176 (2.8%) Placebo 9/165 (5.5%) No deaths or suicides At least 1 EPS AE Aripiprazole 28/176 (15.9%) Placebo 15/165 (9.1%) Weight gain >7% Aripiprazole 11% Placebo 3.5% p=0.007
Divalproex + lithium vs. placebo + lithium alone	Kemp, 2009³¹ High 19192457	Time to Relapse 52 weeks Hazard Ratio 0.72 (0.32, 1.65) NS	NA (Attrition 74%)	NA (Attrition 74%)	NA	No SAEs mentioned. EPS Tremors NS
Olanzapine + mood stabilizer vs. placebo + mood stabilizer	Tohen, 2004³² Industry 15056579 extension of Tohen, 2002³³ 11779284	Time to relapse 18 months Log rank NS	NA (Attrition 78%)	NA	Time to overall withdrawal 18 month Favors Olanzapine Log rank p=0.049	SAE not reported EPS No difference between groups
Oxcarbazepine + lithium vs. placebo + lithium	Vieta, 2008³⁹ Moderate 18346292	Time to relapse¹ 52 week NS Kaplan Meier log-rank Relapse Rate 52 week NS	YMRS Change 52 week NS MADRS Change 52 week NS	CGI-BP-M 52 week NS p=0.45	Weight Gain ≥7% of baseline Oxcarbazapine 19.2% Placebo 6.9% Withdrawal for AE Oxcarbazapine 3 Placebo 2 Withdrawal for lack of efficacy Oxcarbazapine 0 Placebo 2	Serious Adverse Events 52 week 3 events – Oxcarbazepine 3 events – Placebo
Gabapentin + mood stabilizers vs. placebo + mood stabilizers	Vieta, 2006⁴⁰ High 16649836	Time to Relapse 52 week NS HR 1.344	YMRS Change 52 week NS HAM-D 52 week NS	CGI-BP-M Change 52 week Favors Gabapentin 1.5% (95% CI 0.5,2.5) p=0.0046	Withdrawal AE Gabapentin 1 (8%) Placebo 1 (8%) Withdrawal lack of efficacy Gabapentin 2 (15%) Placebo 1 (8%)	Gabapentin: 1 Myocardial Infarction Placebo: No Events Reported
Perphenazine + mood stabilizers vs. placebo + mood stabilizers	Zarate, 2004⁴¹ High 14702269	Time to Relapse to Depression Favors placebo Perphenazine: 157 days (SE 10) Placebo: None Occurred p<0.03 Depressive Relapse Favors placebo Perphenazine 21% Placebo 0% Manic Relapse NS Perphenazine 5% Placebo 11%	NR	NR	Overall Withdrawal Perphenazine 10 (52.6%) Placebo 3 (16.7%)	NR
Quetiapine + mood stabilizers vs. placebo + mood stabilizers	Suppes, 2009³⁴ High 19289454	Time to Recurrence of mood event 104 weeks Hazard Ratio 0.32 (0.24, 0.42) risk reduction 68% Favors Quetiapine (not dependent on rapid cycling)	70% risk reduction in time to recurrence of mania 67% risk reduction in time to recurrence of depression	Not applicable	Overall Withdrawal Quetiapine 200/310 Placebo 247/313 Loss to followup and Other categories greater than adverse event or lack of efficacy categories	SAE Quetiapine: 18 (5.8%) Placebo: 7 (2.2%)
	Vieta, 2008³⁵ High 18579216	Time to Recurrence of mood event 104 weeks Hazard Ratio 0.28 (0.21, 0.37) risk reduction 72% Favors Quetiapine	Time to Recurrence of mania 104 weeks Hazard Ratio 0.30 (0.20, 0.44 Favors Quetiapine Time to Recurrence of depression 104 weeks Hazard Ratio 0.26 (0.17, 0.41 Favors Quetiapine	Not applicable	Overall Withdrawal Quetiapine 123/336 Placebo 233/367 Loss to followup and Other categories greater than adverse event or lack of efficacy categories	SAE Quetiapine: 5 (1.5%) Placebo: 20 (5.4%)
Risperidone long acting injectable + treatment as usual vs. placebo + treatment as usual	Macfadden, 2009³⁷ High 19922552	Relapse - Time to (DSM diagnosis for acute mood episode + other complicated criteria) 52 weeks Log rank test p=0.01 Favors Risperidone	YMRS 52 weeks Favors Risperidone (only figure) MADRS 52 weeks NS	CGI-BP-S 52 weeks Favors Risperidone (only figure)	Overall Withdrawal Risperidone 26/65 (40%) Placebo 34/59 (57.6%) NS Withdrawal lack of efficacy Risperidone 13/65 (20%) Placebo 23/59 (39%) p=0.02 Withdrawal for AE Risperidone 3/65 (4.6%) Placebo 1/59 (1.7% NS	At least 1 SAE Risperidone: 9 Placebo: 13 Deaths Risperidone: 1 Placebo: 2 (1 of suicide 3 months after study) Suicide ideation: Risperidone: 1 Placebo: 3 EPS NS
Risperidone long-acting injectable + treatment as usual vs. treatment as usual	Bobo, 2011³⁶ High 22104634	Any cause relapse 52 weeks NS	YMRS 52 weeks NS MADRS 52 weeks NS	CGI-S 52 weeks NS	Overall Withdrawal Risperidone 4/25 Placebo 6/25 NS Withdrawal lack of efficacy Risperidone 6/40 (15%) Placebo 8/43 (18.6%) NS Withdrawal for AE Risperidone 0% Placebo 9.3% p=0.049	No suicide attempts; NS for suicide ideation EPS NS
Ziprasidone + mood stabilizers vs. placebo + mood stabilizers	Bowden, 2010³⁸ High 20122373 (also 22999893)	Time to relapse 26 weeks Log rank test p=0.01 Favors Ziprasidone	YMRS 26 weeks Least squares mean difference −3.27 (0.83) p<0.001 Favors Ziprasidone MADRS NS	NR	Weight gain >7% NS Overall Withdrawal Ziprasidone 43/127 Placebo 58/113 p=0.007 Withdrawal lack of efficacy Ziprasidone 9/127 Placebo 22/113 p=0.004 Withdrawal for AE Ziprasidone 11/127 Placebo 15/113 NS	SAE Ziprasidone: 11/127 8.7% Placebo: 6/112 5.4% No deaths

: Abbreviations: AE=Adverse Events; CGI-BP-M=Clinical Global Impressions Scale-Bipolar-modified (for long-term follow-up); CGI-BP-S=Clinical Global Impressions, Bipolar, Severity Scale; CGI-S=Clinical Global Impressions, Severity Scale; CI=Confidence Interval; DSM=Diagnostic and Statistical Manual of Mental Disorders; EPS=extrapyramidal symptoms; HAM-D=Hamilton Scale for Depression; HR=Hazard Ratio; MADRS=Montgomery-Asberg Depression Rating Scale; MID=Minimally important difference; NA=Not applicable; NS= Not Significant; PMID=PubMed Identification Number; RoB=Risk of Bias; SAE=Serious Adverse Events; SE=standard error; YMRS = Young Mania Rating Scale;

Appendix Table I14Strength of evidence assessment: combination therapy versus placebo for maintenance

Comparison	Outcome	# Studies/Design (n analyzed)	Finding or Summary Statistic Outcome Timing	Study Limitations	Consistency	Directness	Precision	Overall Grade/Conclusion
Aripiprazole + mood stabilizer vs. placebo + mood stabilizer	Time to Relapse 52 wks YMRS 6 mths MADRS 6 mths CGI 6 mths Withdrawals	2 RCT (n=420)	See table above	High	Unknown (over 2 time periods; Inconsistent if combined)	Direct	Imprecise	Insufficient
Divalproex + lithium vs. placebo + lithium alone	Time to relapse 26 weeks	1 RCT (n=31)	See table above	High	Unknown	Direct	Imprecise	Insufficient
Olanzapine + mood stabilizer vs. placebo + mood stabilizer	Time to relapse 18 months	1 RCT (n=99)	See table above	High	Unknown	Direct	Imprecise	Insufficient
Oxcarbazepine + lithium vs. placebo + lithium	Time to relapse 52 wks YMRS 52 wks HAM-D 52 wks CGI-BP-M 52 wks	1 RCT (n=55)	See table above	Moderate	Unknown	Direct	Imprecise	Insufficient
Gabapentin + mood stabilizers vs. placebo + mood stabilizers	Time to relapse 52 wks YMRS 52 wks HAM-D 52 wks CGI-CP-M 52 wks	1 RCT (n=25)	See table above	High	Unknown	Direct	Imprecise	Insufficient
Perphenazine + mood stabilizers vs. placebo + mood stabilizers	Time to depression relapse 6 mths	1 RCT (n=37)	See table above	High	Unknown	Direct	Imprecise	Insufficient
Quetiapine + mood stabilizers vs. placebo + mood stabilizers	Time to recurrence any mood,102 wks time to mania 102 wks time to depression 102 wks	2 RCT (n=1329)	See table above	High	Consistent	Direct	Imprecise	Insufficient
Risperidone long acting injectable + treatment as usual vs. placebo + treatment as usual	Time to recurrence 52 wks YMRS 52 wks MADRS 52 wks CGI 52 wks Withdrawals	2 RCT (n=174)	See table above	High	Inconsistent	Direct	Imprecise	Insufficient
Ziprasidone + mood stabilizers vs. placebo + mood stabilizers	Relapse Risk 26 wks YMRS 26 wks MADRS 26 wks Withdrawals	1 RCT Combination (n=240)	See table above	High	Unknown	Direct	Imprecise	Insufficient

: Abbreviations: CGI=Clinical Global Impressions; CGI-BP-M=Clinical Global Impressions Scale-Bipolar-modified (for long-term follow-up); HAM-D=Hamilton Scale for Depression; MADRS=Montgomery-Asberg Depression Rating Scale; RCT=randomized controlled trial; YMRS = Young Mania Rating Scale
: Notes:
: 1. Publication bias for antipsychotics, antidepressants, and behavioral interventions for depressive disorders is suspected.
: 2. Data were generally imprecise due to missing data from high attrition rates, which was commonly dealt with by Last Observation Carried Forward (LOCF). LOCF requires an assumption that the health status of patients who dropped out of the trial would not have changed had future observations been recorded, a strong assumption in the context of bipolar disorder research

Appendix Table I15Outcomes summary: combination therapy versus active control for maintenance

Drug	Study Comparison PMID	Responder/Remitter	Symptom	Function	Other	AE
Lithium + valproate vs. valproate vs. lithium	Balance Investigators, 2010¹⁸ 20092882	Time to new intervention for emerging mood episode 24 months Hazard ratio Favors L+V over Valproate NS for L+V vs. Lithium	NR	GAF NS	EuroQol (EQ-5D) (quality of life) NS Overall Withdrawal L+V: 21/110 Lithium: 23/110 Valproate: 23/110 Withdrawal lack of efficacy NR Withdrawal adverse events L+V: 11/110 Lithium: 6/110 Valproate: 4/110	SAE NS Valproate: 7 SAE including 3 deaths Lithium: 5 SAE including 2 deaths L+V: 4 SAE including 1 death (only one deemed due to study – did not report which)
Quetiapine + personalize treatment vs. Lithium + personalized treatment	Nierenberg, 2016⁴² 26845264	NR	YMRS 24 week NS (model-based effect) MADRS 24 week NS (model-based effect)	CGI-EI 24 weeks NS (model-based effect)	Overall Withdrawa Quetiapine: 60/180 24.8% Lithium: 58/182 24.2% Withdrawal lack of efficacy NR Withdrawal serious adverse events Quetiapine: 2/180 Lithium: 0/182 24	Death Quetiapine: 0 Lithium: 2
Valproic acid + Aripiprazole vs. Lithium + Aripiprazole	Vieta, 2010⁴³ 20429835	Remission (YMRS<12, MADRS<8) At least 2/3 patients in both groups after 40 weeks	YMRS 46 weeks Mean change + Lithium: −2.7 (−4.5, −0.7) + Valproic: −5.8 (−7.2, −4.3) MADRS 46 weeks Mean change + Lithium: −0.8 (−2.6, 1.0) + Valproic: −1.2 (−2.6, 0.3)	NR	Overall Withdrawal +Lithium: 53/108 49.1% +Valproic: 84/175 48% Withdrawal lack of efficacy +Lithium: 1/108 0.9% +Valproic: 10/175 5.7% Withdrawal adverse events +Lithium: 20/108 18.5% +Valproic: 24/175 13.7%	SAE +Lithium: 15 (14.2%) 1 patient suicidal ideation 1 lithium overdose death (50 days after last study dose) +Valproic: 15 (8.6%) EPS +Lithium: 24 (22.6%) +Valproic: 38 (21.8%)

: Abbreviations: CGI-EI=Clinical Global Impressions-Efficacy Index; EPS=extrapyramidal symptoms; GAF=General Assessment of Functioning Scale; MADRS=Montgomery-Asberg Depression Rating Scale; NR=not reported; NS=not significant; PMID=PubMed Identification Number; SAE=Severe Adverse Events; YMRS = Young Mania Rating Scale

Appendix Table I16Strength of evidence assessment: combination therapy versus active control for maintenance

Comparison	Outcome	# Studies/Design (n analyzed)	Finding or Summary Statistic Outcome Timing	Study Limitations	Consistency	Directness	Precision	Overall Grade/Conclusion
Lithium + valproate vs. valproate vs. lithium	Time to Relapse 24 mths GAF 24 mths EuroQual 24 mths Withdrawals	1 RCT (n=330)	See table above	Moderate	Unknown	Direct	Imprecise	Insufficient
Quetiapine + personalize treatment vs. Lithium + personalized treatment	YMRS 24 wks MADRS 24 wks CGI 24 wks Withdrawals	1 RCT (n=482)	See table above	Moderate	Unknown	Direct	Imprecise	Insufficient
Valproic acid + Aripiprazole vs. Lithium + Aripiprazole	Remission 46 wks YMRS 46 wks MADRS 46 wks Withdrawals	1 RCT open label extension (n=28	See table above	High	Unknown	Direct	Imprecise	Insufficient

: Abbreviations: CGI= Clinical Global Impressions; GAF=General Assessment of Functioning Scale; MADRS=Montgomery-Asberg Depression Rating Scale; RCT=randomized controlled trial; YMRS = Young Mania Rating Scale
: Notes:
: 1. Publication bias for antipsychotics, antidepressants, and behavioral interventions for depressive disorders is suspected.
: 2. Data were generally imprecise due to missing data from high attrition rates, which was commonly dealt with by Last Observation Carried Forward (LOCF). LOCF requires an assumption that the health status of patients who dropped out of the trial would not have changed had future observations been recorded, a strong assumption in the context of bipolar disorder research.

References for Appendix I

1.: Bowden CL, Calabrese JR, McElroy SL, et al A randomized, placebo-controlled 12-month trial of divalproex and lithium in treatment of outpatients with bipolar I disorder. Divalproex Maintenance Study Group. Archives of General Psychiatry. 2000 May;57(5):481–9. PMID: 10807488. [PubMed: 10807488]
2.: Gyulai L, Bowden CL, McElroy SL, et al Maintenance efficacy of divalproex in the prevention of bipolar depression. Neuropsychopharmacology. 2003 Jul;28(7):1374–82. PMID: 12784116. [PubMed: 12784116]
3.: Amsterdam JD, Shults J. Efficacy and safety of long-term fluoxetine versus lithium monotherapy of bipolar II disorder: a randomized, double-blind, placebo-substitution study. American Journal of Psychiatry. 2010 Jul;167(7):792–800. doi: http://dx.doi.org/10.1176/appi.ajp.2009.09020284. PMID: 20360317. [PMC free article: PMC2896440] [PubMed: 20360317]
4.: Bowden CL, Calabrese JR, Sachs G, et al A placebo-controlled 18-month trial of lamotrigine and lithium maintenance treatment in recently manic or hypomanic patients with bipolar I disorder. [Erratum appears in Arch Gen Psychiatry. 2004 Jul;61(7):680]. Archives of General Psychiatry. 2003 Apr;60(4):392–400. PMID: 12695317. [PubMed: 12695317]
5.: Calabrese JR, Bowden CL, Sachs G, et al A placebo-controlled 18-month trial of lamotrigine and lithium maintenance treatment in recently depressed patients with bipolar I disorder. Journal of Clinical Psychiatry. 2003 Sep;64(9):1013–24. PMID: 14628976. [PubMed: 14628976]
6.: Prien RF, Point P, Caffey EM, et al Prophylactic efficacy of lithium carbonate in manic-depressive illness: Report of the veterans administration and national institute of mental health collaborative study group. Archives of General Psychiatry. 1973;28(3):337–41. doi: 10.1001/archpsyc.1973.01750330035006. PMID: 4569674. [PubMed: 4569674] [CrossRef]
7.: Weisler RH, Nolen WA, Neijber A, et al Continuation of quetiapine versus switching to placebo or lithium for maintenance treatment of bipolar I disorder (Trial 144: a randomized controlled study). Journal of Clinical Psychiatry. 2011 Nov;72(11):1452–64. doi: http://dx.doi.org/10.4088/JCP.11m06878. PMID: 22054050. [PubMed: 22054050]
8.: Calabrese JR, Sanchez R, Jin N, et al Efficacy and safety of aripiprazole once-monthly in the maintenance treatment of bipolar i disorder: A double-blind, placebo-controlled, 52-week randomized withdrawal study. Journal of Clinical Psychiatry. 2017 March;78(3):324–31. doi: http://dx.doi.org/10.4088/JCP.16m11201. PMID: 28146613 PMID/615221995 Embase. [PubMed: 28146613]
9.: Keck PE, Jr., Calabrese JR, McQuade RD, et al A randomized, double-blind, placebo-controlled 26-week trial of aripiprazole in recently manic patients with bipolar I disorder. Journal of Clinical Psychiatry. 2006 Apr;67(4):626–37. PMID: 16669728. [PubMed: 16669728]
10.: Hartong EG, Moleman P, Hoogduin CA, et al Prophylactic efficacy of lithium versus carbamazepine in treatment-naive bipolar patients. Journal of Clinical Psychiatry. 2003 Feb;64(2):144–51. PMID: 12633122. [PubMed: 12633122]
11.: Greil W, Ludwig-Mayerhofer W, Erazo N, et al Lithium versus carbamazepine in the maintenance treatment of bipolar disorders-a randomised study. Journal of Affective Disorders. 1997 Apr;43(2):151–61. PMID: 9165384. [PubMed: 9165384]
12.: Greil W, Kleindienst N, Erazo N, et al Differential response to lithium and carbamazepine in the prophylaxis of bipolar disorder. Journal of Clinical Psychopharmacology. 1998 Dec;18(6):455–60. PMID: 9864077. [PubMed: 9864077]
13.: Greil W, Kleindienst N. The comparative prophylactic efficacy of lithium and carbamazepine in patients with bipolar I disorder. International Clinical Psychopharmacology. 1999 Sep;14(5):277–81. PMID: 10529070. [PubMed: 10529070]
14.: Greil W, Kleindienst N. Lithium versus carbamazepine in the maintenance treatment of bipolar II disorder and bipolar disorder not otherwise specified. International Clinical Psychopharmacology. 1999 Sep;14(5):283–5. PMID: 10529071. [PubMed: 10529071]
15.: Kleindienst N, Greil W. Differential efficacy of lithium and carbamazepine in the prophylaxis of bipolar disorder: results of the MAP study. Neuropsychobiology. 2000;42 Suppl 1:2–10. PMID: 11093063. [PubMed: 11093063]
16.: Calabrese JR, Shelton MD, Rapport DJ, et al A 20-month, double-blind, maintenance trial of lithium versus divalproex in rapid-cycling bipolar disorder. American Journal of Psychiatry. 2005 Nov;162(11):2152–61. PMID: 16263857. [PubMed: 16263857]
17.: Calabrese JR, Suppes T, Bowden CL, et al A double-blind, placebo-controlled, prophylaxis study of lamotrigine in rapid-cycling bipolar disorder. Lamictal 614 Study Group. Journal of Clinical Psychiatry. 2000 Nov;61(11):841–50. PMID: 11105737. [PubMed: 11105737]
18.: Balance investigators, Geddes JR, Goodwin GM, et al Lithium plus valproate combination therapy versus monotherapy for relapse prevention in bipolar I disorder (BALANCE): a randomised open-label trial. Lancet. 2010 Jan 30;375(9712):385–95. doi: http://dx.doi.org/10.1016/S0140-6736(09)61828-6. PMID: 20092882. [PubMed: 20092882]
19.: Tohen M, Calabrese JR, Sachs GS, et al Randomized, placebo-controlled trial of olanzapine as maintenance therapy in patients with bipolar I disorder responding to acute treatment with olanzapine. American Journal of Psychiatry. 2006 Feb;163(2):247–56. PMID: 16449478. [PubMed: 16449478]
20.: 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. American Journal of Psychiatry. 2003 Jul;160(7):1263–71. PMID: 12832240. [PubMed: 12832240]
21.: Tohen M, Baker RW, Altshuler LL, et al Olanzapine versus divalproex in the treatment of acute mania. [Erratum appears in Am J Psychiatry. 2005 Feb;7(1):102]. American Journal of Psychiatry. 2002 Jun;159(6):1011–7. PMID: 12042191. [PubMed: 12042191]
22.: Quiroz JA, Yatham LN, Palumbo JM, et al Risperidone long-acting injectable monotherapy in the maintenance treatment of bipolar I disorder. Biological Psychiatry. 2010 Jul 15;68(2):156–62. doi: http://dx.doi.org/10.1016/j.biopsych.2010.01.015. PMID: 20227682. [PubMed: 20227682]
23.: Vieta E, Montgomery S, Sulaiman AH, et al A randomized, double-blind, placebo-controlled trial to assess prevention of mood episodes with risperidone long-acting injectable in patients with bipolar I disorder. European Neuropsychopharmacology. 2012 Nov;22(11):825–35. doi: http://dx.doi.org/10.1016/j.euroneuro.2012.03.004. PMID: 22503488. [PubMed: 22503488]
24.: Tohen M, Greil W, Calabrese JR, et al Olanzapine versus lithium in the maintenance treatment of bipolar disorder: a 12-month, randomized, double-blind, controlled clinical trial. American Journal of Psychiatry. 2005 Jul;162(7):1281–90. PMID: 15994710. [PubMed: 15994710]
25.: Berwaerts J, Melkote R, Nuamah I, et al A randomized, placebo- and active-controlled study of paliperidone extended-release as maintenance treatment in patients with bipolar I disorder after an acute manic or mixed episode. Journal of Affective Disorders. 2012 May;138(3):247–58. doi: http://dx.doi.org/10.1016/j.jad.2012.01.047. PMID: 22377512. [PubMed: 22377512]
26.: Amsterdam JD, Lorenzo-Luaces L, Soeller I, et al Safety and effectiveness of continuation antidepressant versus mood stabilizer monotherapy for relapse-prevention of bipolar II depression: A randomized, double-blind, parallel-group, prospective study. Journal of Affective Disorders. 2015 06 Jul;185:31–7. doi: http://dx.doi.org/10.1016/j.jad.2015.05.070. PMID: 26143402 (pubmed) 2015175772 (embase). [PMC free article: PMC4540653] [PubMed: 26143402]
27.: Newport DJ, Stowe ZN, Viguera AC, et al Lamotrigine in bipolar disorder: efficacy during pregnancy. Bipolar Disorders. 2008 May;10(3):432–6. doi: http://dx.doi.org/10.1111/j.1399-5618.2007.00565.x. PMID: 18402631. [PubMed: 18402631]
28.: Marcus R, Khan A, Rollin L, et al Efficacy of aripiprazole adjunctive to lithium or valproate in the long-term treatment of patients with bipolar I disorder with an inadequate response to lithium or valproate monotherapy: a multicenter, double-blind, randomized study. Bipolar Disorders. 2011 Mar;13(2):133–44. doi: http://dx.doi.org/10.1111/j.1399-5618.2011.00898.x. PMID: 21443567. [PubMed: 21443567]
29.: Carlson BX, Ketter TA, Sun W, et al Aripiprazole in combination with lamotrigine for the long-term treatment of patients with bipolar I disorder (manic or mixed): a randomized, multicenter, double-blind study (CN138-392). Bipolar Disorders. 2012 Feb;14(1):41–53. doi: http://dx.doi.org/10.1111/j.1399-5618.2011.00974.x. PMID: 22329471. [PubMed: 22329471]
30.: Woo YS, Bahk WM, Chung MY, et al Aripiprazole plus divalproex for recently manic or mixed patients with bipolar I disorder: a 6-month, randomized, placebo-controlled, double-blind maintenance trial. Human Psychopharmacology. 2011 Dec;26(8):543–53. doi: http://dx.doi.org/10.1002/hup.1240. PMID: 22134973. [PubMed: 22134973]
31.: Kemp DE, Gao K, Ganocy SJ, et al A 6-month, double-blind, maintenance trial of lithium monotherapy versus the combination of lithium and divalproex for rapid-cycling bipolar disorder and Co-occurring substance abuse or dependence. Journal of Clinical Psychiatry. 2009 Jan;70(1):113–21. PMID: 19192457. [PMC free article: PMC3587136] [PubMed: 19192457]
32.: Tohen M, Chengappa KN, Suppes T, et al Relapse prevention in bipolar I disorder: 18-month comparison of olanzapine plus mood stabiliser v. mood stabiliser alone. British Journal of Psychiatry. 2004 Apr;184:337–45. PMID: 15056579. [PubMed: 15056579]
33.: Tohen M, Chengappa KN, Suppes T, et al Efficacy of olanzapine in combination with valproate or lithium in the treatment of mania in patients partially nonresponsive to valproate or lithium monotherapy. Archives of General Psychiatry. 2002 Jan;59(1):62–9. PMID: 11779284. [PubMed: 11779284]
34.: Suppes T, Vieta E, Liu S, et al Maintenance treatment for patients with bipolar I disorder: results from a north american study of quetiapine in combination with lithium or divalproex (trial 127). American Journal of Psychiatry. 2009 Apr;166(4):476–88. doi: http://dx.doi.org/10.1176/appi.ajp.2008.08020189. PMID: 19289454. [PubMed: 19289454]
35.: Vieta E, Suppes T, Eggens I, et al Efficacy and safety of quetiapine in combination with lithium or divalproex for maintenance of patients with bipolar I disorder (international trial 126). Journal of Affective Disorders. 2008 Aug;109(3):251–63. doi: http://dx.doi.org/10.1016/j.jad.2008.06.001. PMID: 18579216. [PubMed: 18579216]
36.: Bobo WV, Epstein RA, Lynch A, et al A randomized open comparison of long-acting injectable risperidone and treatment as usual for prevention of relapse, rehospitalization, and urgent care referral in community-treated patients with rapid cycling bipolar disorder. Clinical Neuropharmacology. 2011 Nov–Dec;34(6):224–33. doi: http://dx.doi.org/10.1097/WNF.0b013e318237709a. PMID: 22104634. [PMC free article: PMC3951021] [PubMed: 22104634]
37.: Macfadden W, Alphs L, Haskins JT, et al A randomized, double-blind, placebo-controlled study of maintenance treatment with adjunctive risperidone long-acting therapy in patients with bipolar I disorder who relapse frequently. Bipolar Disorders. 2009 Dec;11(8):827–39. doi: http://dx.doi.org/10.1111/j.1399-5618.2009.00761.x. PMID: 19922552. [PubMed: 19922552]
38.: Bowden CL, Vieta E, Ice KS, et al Ziprasidone plus a mood stabilizer in subjects with bipolar I disorder: a 6-month, randomized, placebo-controlled, double-blind trial. Journal of Clinical Psychiatry. 2010 Feb;71(2):130–7. doi: http://dx.doi.org/10.4088/JCP.09m05482yel. PMID: 20122373. [PubMed: 20122373]
39.: Vieta E, Cruz N, Garcia-Campayo J, et al A double-blind, randomized, placebo-controlled prophylaxis trial of oxcarbazepine as adjunctive treatment to lithium in the long-term treatment of bipolar I and II disorder. International Journal of Neuropsychopharmacology. 2008 Jun;11(4):445–52. doi: http://dx.doi.org/10.1017/S1461145708008596. PMID: 18346292. [PubMed: 18346292]
40.: Vieta E, Manuel Goikolea J, Martinez-Aran A, et al A double-blind, randomized, placebo-controlled, prophylaxis study of adjunctive gabapentin for bipolar disorder. Journal of Clinical Psychiatry. 2006 Mar;67(3):473–7. PMID: 16649836. [PubMed: 16649836]
41.: Zarate CA, Jr., Tohen M. Double-blind comparison of the continued use of antipsychotic treatment versus its discontinuation in remitted manic patients. American Journal of Psychiatry. 2004 Jan;161(1):169–71. PMID: 14702269. [PubMed: 14702269]
42.: Nierenberg AA, McElroy SL, Friedman ES, et al Bipolar CHOICE (Clinical Health Outcomes Initiative in Comparative Effectiveness): a pragmatic 6-month trial of lithium versus quetiapine for bipolar disorder. Journal of Clinical Psychiatry. 2016 Jan;77(1):90–9. doi: http://dx.doi.org/10.4088/JCP.14m09349. PMID: 26845264. [PubMed: 26845264]
43.: Vieta E, Owen R, Baudelet C, et al Assessment of safety, tolerability and effectiveness of adjunctive aripiprazole to lithium/valproate in bipolar mania: a 46-week, open-label extension following a 6-week double-blind study. Current Medical Research & Opinion. 2010 Jun;26(6):1485–96. doi: http://dx.doi.org/10.1185/03007991003779380. PMID: 20429835. [PubMed: 20429835]

Bookshelf ID: NBK532190

Contents