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.)
Key Points
- Evidence for maintenance treatments was scattered across 16 drugs administered alone or in combination therapy.
- Evidence was largely insufficient to draw conclusions regarding the effects of drug treatments for maintenance of euthymia in adults with bipolar disorders (BD) for the primary outcomes of interest (relapse, symptom scores, and function).
- Low-strength evidence showed longer time to recurrence of any mood state for bipolar I disorder (BD-I) patients receiving lithium compared to placebo (n=1579) in followup up to 2 years. Participants receiving lithium reported more tremor than those receiving placebo.
- Evidence was insufficient for all other outcomes across all interventions.
Eligible Studies for Maintenance Treatments
We identified 44 eligible publications reporting 36 unique studies with at least 6 months followup.56, 139, 157-192 Twenty-one studies, seven of which were three-arm studies, examined a single drug treatment for maintenance,56, 82, 99, 117, 124, 139, 160, 161, 164, 165, 167-173, 176, 178, 180, 181, 184, 187, 190, 191 and 16 examined drug combinations.88, 116, 160, 162, 163, 166, 174, 175, 177, 179, 182, 183, 185, 186, 188, 189, 192 Drugs examined included: oral aripiprazole, long-acting injectable aripiprazole, divalproex/valproate, carbamazepine, fluoxetine, gabapentin, lamotrigine, lithium, olanzapine, oxcarbazepine, paliperidone, perphenazine, long-acting injectable risperidone, quetiapine, venlafaxine, and ziprasidone. Fourteen studies were assessed as low or moderate risk of bias and 22 were assessed as high, generally due to attrition. Of 36 unique studies, 27 were industry funded. An additional 15 studies were excluded due to attrition over 50 percent and not using time to relapse outcomes.74, 75, 81, 83, 86, 87, 95, 115, 119, 120, 123, 125, 193-196 Only two studies were not RCTs.176, 185 Sample sizes ranged from 25 to 1226; 17 studies were below 200 participants, ranging from 25 to 175. Study duration ranged from 6 months to 3 years, with 24 using followup of 6 months to 1 year.
Appendix I provides detailed evidence tables, summary risk of bias assessments, and assessments of strength of evidence for key comparisons and outcomes. A summary of findings with at least low-strength evidence for other drug treatments for maintenance are provided in Table 29. Any intervention and comparison not listed in Table 29, or outcome not listed for an included intervention and comparison, was found to have an evidence base insufficient to draw conclusions.
Table 29
Summary of findings with at least low-strength evidence for maintenance studies.
Single Drug Treatments for Maintenance
Table 30 summarizes the bipolar type and major inclusion and exclusion criteria for single drug studies for maintenance. Appendix I provides details.
Table 30
Population and inclusion criteria for single drug studies for maintenance.
Single Drug for Maintenance Versus Placebo
Twelve studies examined nine different drugs versus placebo in participants with BD-I.82, 99, 117, 124, 139, 161, 164, 165, 167, 178, 180, 184, 187, 191 Five studies also included bipolar II disorder (BD-II) participants.117, 124, 163, 182, 183 Sample sizes ranged from 26 to 1226 and followup lasted from 26 weeks to 3 years.
Low-strength evidence (moderate study limitations, imprecision) from six RCTs (n=1579) showed that adults with BD-I receiving lithium over a 2 year period had longer time to recurrence of any mood state compared to those receiving a placebo.179, 186 Since the time to event outcomes account for attrition, these were the only outcomes abstracted from these studies due to the high attrition rates. Evidence was insufficient for time to manic or depressive states due to mixed results. Participants receiving lithium reported more tremor than those receiving placebo. Otherwise serious adverse events did not differ by group. Appendix I provides details.
Evidence was insufficient for all outcomes to address whether ten drugs were better than placebo for maintenance in adults with BD: long-acting aripiprazole (n=226),191 aripiprazole (n=1610,56 divalproex (n=281),165, 172 fluoxetine (n=55),162 lamotrigine (n=471; n=182 rapid cycling),124, 164, 167 olanzapine (n=855),139, 180, 184 paliperidone (n=300),139 quetiapine (n=808),187 and risperidone (n=353).99, 184 Single studies, high study limitations, small sample sizes, and strong imprecision contributed to the insufficient strength of evidence rating. Except for divalproex, results were reported as favoring the interventions for time to overall relapse. Where reported, participants using placebo experience less frequent severe events of tremor than those using divalproex, or less parkinsonism than those using olanzapine; otherwise, serious adverse events were generally not different between groups. Appendix I provides details.
While providing insufficient evidence to draw conclusions, one observational study was noteworthy for examining lamotrigine use in 26 pregnant women, recruited before conception or during first trimester, with any BD type. Women chose to discontinue all mood stabilizers or to continue on lamotrigine only. While women who chose to continue lamotrigine were less likely to have an unplanned pregnancy than those who discontinued all treatment. Risk of relapse was 3/10 women using lamotrigine versus 16/16 women who discontinued treatment.
Single Drug for Maintenance Versus Active Control
Fourteen studies (20 publications) examined 10 different drugs versus another drug.139, 157-162, 164, 165, 167-173, 178, 181, 184, 187, 190 Sample sizes ranged from 54 to 768 and followup lasted from 6 months to 3 years. Appendix I provides details.
Evidence was insufficient for all outcomes to address whether carbamazepine (n=171),168-171, 173, 190 divalproex (n=372, n=60 rapid cycling),117, 164, 172 fluoxetine (n=54),162 lamotrigine (n=390),165, 167 olanzapine (n=855),181 quetiapine (n=768),187 valproate (n=220),160 and venlafaxine (n=55)161 was better than lithium; paliperidone (n=235)139 or risperidone (n=263)184 was better than olanzapine; or olanzapine was better than divalproex (n=251)82 for maintenance in adults with BD. Single studies, high study limitations, small sample sizes, and imprecision contributed to the insufficient strength of evidence rating. Results were mixed across the studies. With the exception of participants using divalproex showing less akathisia compared to those using lithium, no differences between groups were reported for serious adverse events.
Combination Drug Treatment for Maintenance
Table 31 summarizes bipolar type and major inclusion and exclusion criteria for combination drug therapy studies for maintenance. Appendix I provides details.
Table 31
Population and inclusion criteria for combination drug treatment for maintenance studies.
Combination Drug Therapy for Maintenance Versus Placebo
Thirteen studies examined nine different combination therapies versus placebo in BD-I participants88, 116, 163, 166, 174, 175, 179, 182, 183, 186, 188, 189, 192 Four studies also included BD-II participants.116, 163, 182, 183 Sample sizes ranged from 25 to 706 and followup lasted from 26 weeks to 2 years.
Evidence was insufficient to address whether nine combinations performed better than placebo: aripiprazole plus mood stabilizers (n=771),175, 188 divalproex plus lithium (n=31),116 gabapentin plus mood stabilizers (n=25),183 olanzapine plus mood stabilizers (n=99);88 oxcarbazepine plus lithium (n=55),182 perphenazine plus mood stabilizers (n=37),189 quetiapine plus mood stabilizers (n=1329), long-acting injectable risperidone plus mood stabilizers (n=174),163, 174 and ziprasidone plus mood stabilizers (n=240).166 Single studies, high study limitations, small sample sizes, and imprecision contributed to the insufficient strength of evidence rating. Results were mixed across the studies and generally showed no differences between groups in withdrawals due to adverse events. Serious adverse events were also not different between groups.
Combination Therapy for Maintenance Versus Active Control
Three studies examined combination therapies versus active comparators in BD-I participants, each a unique, single study comparison.160, 177, 185 Only one study also enrolled participants with other types of BD.177 Sample sizes ranged from 283 to 482 and followup lasted from 24 weeks to 2 years.
Evidence was insufficient to address whether lithium plus valproate performed better than either lithium or valproate alone (n=330),160 quetiapine plus mood stabilizers performed better than lithium plus another mood stabilizer (n=482),177 or if aripiprazole plus valproate performed differently than aripiprazole plus lithium (n=283),185 generally due to high study limitations and imprecision. Overall, the trials reported no significant differences between groups. However, the three-group Balance study reported time to relapse hazard ratios favored lithium plus valproate over valproate alone, but did not significantly differ from lithium alone. Also, serious adverse events did not generally differ between groups. All studies reported at least one death, but not to significant differences between groups for such a rare outcome.
Interpreting the Findings for Drug Treatment for Maintenance
The current evidence for drug treatment for maintenance in BD is largely insufficient to draw conclusions for a number of reasons. First, 36 unique maintenance studies examined 16 different medications often resulting in a single study for a specific comparison for a specific followup duration. In addition, 22 of 36 of maintenance studies (61%) were rated as having severe study limitations (high risk of bias). Second, the high rates of attrition often led to only one usable outcome measure—time to recurrence of a bipolar episode—since this metric accounted for high attrition rates by including information from participants who dropped out due to BD episode relapse. Moreover, 17 studies had small sample sizes of less than 200 participants and 24 studies (66%) had followup between six to twelve months, precluding conclusions for long-term maintenance for most of examined treatments. Third, differences in current bipolar phase criteria across studies, ranging from any current phase (i.e., depression, hypomania, or euthymia), remission from mania, remission from any BD episode, or response or partial response to a specific acute episode treatment, made it difficult to determine for whom findings might apply.
Still, low-strength evidence showed a longer time to recurrence of any BD episode for lithium versus placebo treatment in adults with BD I during a two year followup. The evidence was insufficient for time to recurrence of depression or mania due to inconsistent findings. There was a greater rate of tremors but insufficient evidence for differences in other adverse events rates between lithium versus placebo treatment. In general, in single drug versus placebo comparisons, when reported, placebo showed less tremor than divalproex treatment and less parkinsonism signs than olanzapine treatment, but no differences in other serious adverse events. Also, comparisons between drugs and active comparators did not show differences in serious adverse events, except for less akathisia for divalproex than lithium treatment.
The nature of inclusion criteria and study populations limits the applicability of these findings for certain subpopulations of individuals with BD, such as individuals with BD II, older adults with any BD illness type, and individuals at the early stage of BD illness. For example, 20 studies included individuals with BD I only, while studies with multiple BD disorder subtypes did not report results separately by illness type. The majority of studies included younger adults with mean ages in 30s and early 40s. An additional eight studies excluded individuals experiencing first manic episode. Only two small studies looked at individuals with rapid cycling BD. Most studies did not examine whether the number of prior manic or depressive episodes affected the efficacy of drugs during maintenance phase treatment.
- Drug Treatments for Maintenance - Treatment for Bipolar Disorder in Adults: A Sy...Drug Treatments for Maintenance - Treatment for Bipolar Disorder in Adults: A Systematic Review
- PREDICTED: Mus musculus proline-rich coiled-coil 2C (Prrc2c), transcript variant...PREDICTED: Mus musculus proline-rich coiled-coil 2C (Prrc2c), transcript variant X3, mRNAgi|1907068423|ref|XM_006496768.5|Nucleotide
- Treatment for Restless Legs SyndromeTreatment for Restless Legs Syndrome
- Imaging Techniques for the Diagnosis and Staging of Hepatocellular CarcinomaImaging Techniques for the Diagnosis and Staging of Hepatocellular Carcinoma
Your browsing activity is empty.
Activity recording is turned off.
See more...