GME Logo

User login

User menu

Issue 69, Jan 2018
Share This
CBT significantly improves outcomes in bipolar disorder

Chiang KJ et al. Efficacy of cognitive-behavioral therapy in patients with bipolar disorder: A meta-analysis of randomized controlled trials. PLoS One. 2017 May 4;12(5):e0176849. PubMed PMID: 28472082; PubMed Central PMCID: PMC5417606.


Previous data have suggested that cognitive behavioral therapy (CBT) may be efficacious as an adjunct to medication treatment for bipolar disorder.

This study used meta-analysis to combine the results of previous studies on this topic.

The primary outcome measures were improvement in the severity of depressive and manic symptoms.

The secondary outcome measures were reduction in the relapse rate and improvement in psychosocial functioning. 


A thorough search of the literature was done to identify relevant randomized controlled trials.

Meta-analysis was used to combine the results of those studies.

A total of 19 randomized controlled trials with about 1400 participants were included in the meta-analysis.

Participants in these clinical trials had either bipolar I or bipolar II disorder.

Ten of these studies provided individual CBT and nine provided group CBT.

The number of CBT sessions ranged from 8 to 30. The duration of each session ranged from 45 to 120 minutes.

The duration of follow up varied but many of the studies followed patients for 9 to 24 months. 


Participants who received cognitive behavior therapy along with usual care had a statistically significantly greater improvement in both depressive and manic symptoms (the primary outcome measures). The effect size for improvement in depressive and manic symptoms was 0.5 and 0.6, respectively, indicating a moderate benefit.

They also had statistically significantly lower rate of relapse. How much lower? The odds ratio was 0.5, i.e., half of that in those who did not receive the adjunctive CBT.

Lastly, those who received cognitive behavior therapy in addition to usual care also had greater improvement in psychosocial functioning (effect size 0.5 indicating moderate benefit).

The only variable that was associated with a greater benefit of adjunctive CBT for improving depressive symptoms or manic symptoms was a duration of the CBT session of 90 minutes or more. 


Cognitive behavior therapy (CBT) as an adjunct to usual care is effective in improving depressive and manic symptoms, decreasing the relapse rate, and improving psychosocial functioning. 

Clinical Commentary

While medications are the mainstay of treatment for bipolar disorder, this meta-analysis of all relevant studies on the use of CBT as an adjunctive treatment for bipolar disorder shows convincingly that CBT improves outcomes as well.

If any medication had benefits similar what have been shown with cognitive behavior therapy, it would be considered as a possible treatment for every patient with bipolar disorder. So, greater efforts must be made to train clinicians in how to provide CBT to persons with bipolar disorder and CBT should routinely be added to medication treatment.

The specific mechanisms by which CBT improves outcomes in bipolar disorder are not clear from the research. These could include increased adherence to the medication, improved management of stress from relationship problems and life events, greater awareness of the symptoms, better self-care, and so on.

Several of the studies included in this meta-analysis used group CBT. Providing CBT in a group format may be more cost effective than individual CBT. For improvement in depressive and manic symptoms, and reduction in relapse rate, there was no difference in outcomes between those who received individual CBT or group CBT. For improvement in psychosocial functioning, though, there was a trend (not statistically significant) for individual CBT to be more efficacious. 

Expand Content
Share this Section:
Is methylphenidate much less likely than amphetamines to make children with ADHD irritable?

Stuckelman ZD et al. Risk of Irritability With Psychostimulant Treatment in Children With ADHD: A Meta-Analysis. J Clin Psychiatry. 2017 Jun;78(6):e648-e655. PubMed PMID: 28682529.


Irritability is well known to be a common side effect of psychostimulant medications.

This study aimed to evaluate the risk of irritability as a side effect of psychostimulant treatment in children with ADHD.


A search of the literature was done to identify relevant randomized, placebo-controlled trials of using psychostimulant medications for the treatment of ADHD in children. Note: in my opinion, the search strategy was not thorough enough to be confident that every relevant study was identified.

The study excluded very small trials, trials of less than one week, and those that only enrolled participants with psychiatric or medical comorbidity in addition to the ADHD.

A meta-analysis was used to determine the risk ratio (relative risk) of irritability reported as a side effect in children who received a psychostimulant compared to those who received placebo. 


The meta-analysis included 32 relevant clinical trials. A total of about 3700 children with ADHD participated in these clinical trials.

Methylphenidate preparations were associated with a statistically significantly decreased risk of irritability as a side effect of treatment compared to placebo. The risk ratio (relative risk) was 0.89. This means that children who received methylphenidate were slightly less likely to develop irritability as an adverse event than those who received a placebo.

On the other hand, children with ADHD who received an amphetamine preparation had a statistically significantly increased risk of developing irritability as an adverse event. The relative risk was 2.9. That is, the children who received an amphetamine were 2.9 times more likely to develop irritability as an adverse event compared to those who received a placebo.

This difference between the two classes of stimulant medications was not explainable by a difference in the doses used.

The duration of action of the stimulant medication was not associated with the risk of irritability as an adverse event. 


This meta-analysis suggests that an increased risk of irritability as an adverse event may be only related to amphetamine preparations and not to methylphenidate.

The authors suggested that further head-to-head trials directly comparing the effects of methylphenidate and amphetamine preparations in terms of irritability should be done. 

Clinical Commentary

The findings of this meta-analysis were unexpected, even to the authors. They were not specifically looking for this.

But, the finding is consistent with some previous data suggesting that methylphenidate may be more efficacious than amphetamines at reducing pre-existing irritability, which is a common symptom of ADHD in children.

An important clinical task is to differentiate between irritability that occurs at times when the psychostimulant medication levels are expected to be high from that which tends to occur at times when the psychostimulant is expected to be wearing off. This distinction was, unfortunately, not made in the clinical trials included in this meta-analysis. But the duration of action of the psychostimulant preparation was not related to the risk of irritability as an adverse event. This makes it less likely that the findings of this meta-analysis were due to differences in how irritability was assessed. 

Expand Content
Share this Section:
Which antidepressants are associated with an increased risk of seizures?

Wu CS et al. Seizure Risk Associated With Antidepressant Treatment Among Patients With Depressive Disorders: A Population-Based Case-Crossover Study. J Clin Psychiatry. 2017 Oct 24. pii: 16m11377. PubMed PMID: 29068610.


Antidepressant medications are associated with an increased risk of seizures.

This study aimed to compare the risk of seizures associated with different antidepressant medications. 


A Taiwanese total population health insurance database was used.

In the database, the authors identified persons who either visited the emergency department or were hospitalized with a new-onset seizure after receiving an antidepressant for a depressive disorder.

Persons with major neurological comorbidities were excluded.

The relative risk of antidepressant-related seizures was estimated by comparing the rates of antidepressant exposure during the periods when a new-onset seizure occurred compared to other (control) periods.

The effects of the class and dose of the antidepressant on the risk of a new-onset seizure were evaluated after controlling for concomitant medications. 


About 10,000 patients were included in the analysis.

Use of any antidepressant was associated with an increased risk of a new-onset seizure (odds ratio 1.5).

The odds ratios for the risk of a new-onset seizure with different antidepressants, in decreasing order, were as follows:

Bupropion 2.23

Selective serotonin reuptake inhibitors (SSRIs) 1.76

 Serotonin and norepinephrine reuptake inhibitors (SNRIs) 1.40

Mirtazapine 1.38

For all four of these classes of antidepressants, there was a statistically significant relationship between increasing dose and increasing risk of a new-onset seizure.

The risk of new-onset seizures associated with use of an antidepressant was greatest between the ages of 10 and 24 years.

The cumulative seizure rate was 0.68% (about 10,000 persons out of about 1.5 million persons prescribed an antidepressant) over a mean follow up period of 5.4 years.


Use of antidepressant medications is associated with an increased risk of a new-onset seizure. The risk increases with dose and differs between different antidepressants.

The authors recommended that the risk of a new-onset seizure, the association with dose, and differences between the various antidepressants should be considered when choosing an antidepressant. 

Clinical Commentary

In view of previous contradictory data on the association between the use of second-generation antidepressants and the risk of new-onset seizures, this study is significant.

This study is also particularly useful because Taiwan has a National Health Insurance program and data on the entire population is available in a National Health Insurance database. This is helpful because it removes certain potential biases that may occur if only one segment of a population is studied.

While the absolute risk of seizures associated with use of an antidepressant medication may be low, this association may become important in persons who also have other risk factors for having a seizure. 

Expand Content
Share this Section:
Heavy drinking during adolescence changes how the brain develops

Pfefferbaum A et al. Altered Brain Developmental Trajectories in Adolescents After Initiating Drinking. Am J Psychiatry. 2017 Oct 31:appiajp201717040469. [Epub ahead of print] PubMed PMID: 29084454.



Previous data have suggested that heavy alcohol use by adolescents is associated with changes in their brain growth.

In normal development, there is a gradual decline in gray matter volumes after the first decade of life. Supratentorial white matter volume continues to grow throughout adolescence and slows in the third decade of life.

The developing adolescent brain may be more vulnerable to environmental insults like heavy alcohol use.

This study aimed to assess whether moderate and heavy alcohol use by adolescents is associated with an alteration in the trajectory of growth of their brains.


Data from a large national, multisite, study of adolescents were used.

Importantly, data were collected prospectively, i.e., before and after initiation of alcohol use.

The participants were 483 adolescents (ages 12 to 21 years).

They were evaluated before initiation of alcohol use, one year later, and two years later.

MRIs of the brain were used to quantify cortical and white matter volumes in different regions of the brain.

The percent change per year in these volumes was the outcome measure.

Those who continued to have no or low alcohol use throughout the two year follow up served as the control group.

Standardized definitions of “no or low”, “moderate”, and “heavy” alcohol use were used. These were based on a combination of the number of days of alcohol use, the average amount per day, and the maximum amount used in a day.

No/low drinking. Less than once a month, less than two drinks on average, and a maximum of less than four drinks in a day.

Moderate drinking.  Either of the following two:

1. Low drinking frequency (e.g., less than once a month) with moderate quantity consumption (e.g., with two to three drinks per day on average and a maximum of 4 to 5 drinks in a day), or

2. Moderate frequency (e.g., once a week) and low quantity consumption (e.g., two drinks per day on average and a maximum of 3 drinks in a day).

Heavy drinking.  Either of the following two:

1. Moderate frequency (e.g., twice a month) with high quantity consumption (e.g., three to four drinks per day on average and a maximum of more than four drinks in a day), or

2. Higher frequency (e.g., once a week or more) with moderate quantity consumption (e.g., with two to three drinks on average and a maximum of more than four drinks in a day).

Those who had used cannabis more than 50 times in their lifetime were considered to be cannabis users. 


At the 2-year follow-up assessment, 356 participants had no or low alcohol consumption,

65 had started moderate alcohol use, and 62 had started heavy alcohol use.

In the control group (no or low alcohol use), as expected,:

  • Gray matter volume declined throughout adolescence. The parietal cortex showed the greatest decline.
  • Both gray matter and white matter volumes grew at faster rates at younger ages and slowed toward young adulthood.

But in those who started heavy alcohol use, there was an accelerated frontal cortical gray matter trajectory.

In those who started moderate alcohol use, the changes were not statistically significantly different from the control group. But, the changes were intermediate between no/low use and heavy use. This suggests that there may be a dose effect.

Use of cannabis along with alcohol did not affect the main findings of the study. 


Starting heavy alcohol use during adolescence is associated with changes in the trajectory of brain growth. 

Clinical Commentary

I think that this data on how heavy alcohol use is associated with changes in how the adolescent brain develops deserves serious consideration by adolescents, educators, and the public at large.

As the authors note, an escalation of maximum drinks per day has been found to be associated with greater impulsivity/compulsivity scores and with disturbance of frontoparietal control mechanisms.

I can’t help wondering, however, about the chicken-or-egg question: to what extent does the alcohol use lead to the brain changes or do the brain changes lead to the alcohol use? 

Expand Content
Share this Section:
Liraglutide for clozapine- or olanzapine-induced prediabetes and overweight/obesity

Larsen J et al. Effect of Liraglutide Treatment on Prediabetes and Overweight or Obesity in Clozapine- or Olanzapine-Treated Patients With Schizophrenia Spectrum Disorder: A Randomized Clinical Trial. JAMA Psychiatry. 2017 Jul 1;74(7):719-728. PubMed PMID: 28601891.


Persons with schizophrenia have a much higher mortality rate than the general population. This increased mortality is mainly due to cardiovascular disease.

To make things worse, second-generation antipsychotic medications, especially olanzapine and clozapine, are associated with an increased risk of weight gain, impaired glucose tolerance, and dyslipidemia.

Various interventions for these antipsychotic-induced adverse effects have shown limited benefits and additional treatment options are definitely needed.

Liraglutide (brand names Victoza® and Saxenda®) is an agonist at the glucagon−like peptide−1 receptor. Glucagon−like peptide−1, secreted from cells in the intestinal mucosa, stimulates insulin secretion and inhibits glucagon secretion. The combined effect of these is to lower plasma glucose levels.

The FDA-approved indication, as of December 2017, for Saxenda® is “as an adjunct to a reduced-calorie diet and increased physical activity for chronic weight management in adult patients with an initial body mass index (BMI) of 30 kg/m2 or greater (obese), or

27 kg/m2 or greater (overweight) in the presence of at least one weight-related comorbid condition (e.g. hypertension, type 2 diabetes mellitus, or dyslipidemia)”.

This study aimed to evaluate the efficacy and safety of liraglutide added to clozapine or olanzapine in persons with schizophrenia spectrum disorders who were overweight/obese and had prediabetes.


This was a randomized double-blind clinical trial in Denmark. Note: liraglutide is manufactured by Novo Nordisk A/S, which is based in Denmark. This was an investigator-initiated study funded, in part, by an unrestricted grant from the manufacturer.

A total of 103 participants were enrolled who:

1. Had schizophrenia spectrum disorder

2. Were being treated with either clozapine or olanzapine

3. Were overweight or obese

4. Had prediabetes

What is prediabetes? Prediabetes was defined as:

1. An elevated fasting plasma glucose level of 110 to 125mg/dL,

2. An elevated glycated hemoglobin level of 6.1% to 6.4%, and/or

3. An impaired glucose tolerance with a 2-hour plasma glucose level of at least 140 mg/dL during a 75-g oral glucose tolerance test.

Persons with type 2 diabetes (defined as a glycated hemoglobin level of at least 6.5%) and those being treated with antidiabetic medications we excluded from participation in this study.

The participants were randomized to receive either liraglutide or placebo for 16 weeks while continuing stable treatment with clozapine or olanzapine.

Both liraglutide and placebo were given as a once-daily subcutaneous injection.

The dose of liraglutide used in this clinical trial (1.8 mg/day) is the lower maximum dose contained in Victoza® rather than the higher maximum dose (3 mg/day) contained in Saxenda®.


Of the 103 participants, 58% were men and 42% were women.

Their age was variable. The mean age was 43 years with a standard deviation of 11 years.

The mean body mass index was 34 with a standard deviation of six. Note: A body mass index of 30 or more indicates obesity. So, some participants had a body mass index of 25 to 29.9, i.e., were overweight but not obese.

Participants who receive liraglutide had a statistically significantly greater improvement than those who received placebo.

The percentage of participants whose glucose tolerance became normal was 64% with liraglutide and 16% with placebo. This is a very large difference!

Similarly, compared to those who received placebo participants who received liraglutide had a greater reduction in:

1. Body weight (average drug-placebo difference 5 Kg)

2. Waist circumference (average drug-placebo difference 4 cm)

3. Systolic blood pressure (average drug-placebo difference 5 mm Hg)

4. Visceral fat (average drug-placebo difference 250 g)

5. Low-density lipoprotein levels (average drug-placebo difference 15 mg/dL)

Regarding adverse events:

- Nausea was reported by 62% of those on liraglutide and 32% of those on placebo

- Orthostatic hypotension was reported by 8% of those on liraglutide and none of those on placebo.

Could the gastrointestinal adverse effects have contributed to the weight loss? Improvement in body weight and other measures was not different between those who reported nausea and those who did not. 


Liraglutide significantly improved glucose tolerance, body weight, and other metabolic disturbances in patients with schizophrenia spectrum disorders who were being treated with clozapine or olanzapine, were overweight or obese, and had prediabetes.

Clinical Commentary

Metabolic syndrome is a serious and relatively common problem in persons with schizophrenia and related disorders.

In particular, clozapine is unique in its efficacy for the treatment of schizophrenia. Discontinuing clozapine due to metabolic syndrome is usually not an option. So, strong measures may be needed to manage the metabolic syndrome in order to allow the person to continue to take the clozapine if it is absolutely essential for the treatment of the schizophrenia.

Mental health clinicians who are considering referring a patient for possible treatment with liraglutide should be aware that liraglutide carries a boxed warning which notes that liraglutide causes thyroid C-cell tumors in rats and mice. It is unknown whether it causes thyroid C-cell tumors, including medullary thyroid carcinoma, in humans. The warning  also notes that liraglutide is contraindicated in patients with a personal or family history of

Medullary thyroid carcinoma or in patients with Multiple Endocrine Neoplasia syndrome type 2. It states that patients should be counseled regarding the potential risk of medullary thyroid carcinoma and the symptoms of thyroid tumors.

Expand Content
Share this Section:

Rajnish Mago, MD
Medical Editor, GME Research Review

GME Research Review is a monthly newsletter edited by Rajnish Mago, MD, who is author of "The Latest Antidepressants" and "Side Effects of Psychiatric Medications: Prevention, Assessment, and Management." Dr. Mago selects, summarizes, and provides a clinical commentary on the latest published research in psychiatry. 

We are always carefully evaluating which research papers to discuss in GME Research Review. Have come across a research paper published in the last 6 months that you thought is clinically relevant? Do you want me to analyze it for you and for the benefit of others? Please email Dr. Mago the citation at [email protected].

To contact GME, email us at [email protected]

GME does not provide medical advice. The website and articles are intended for informational purposes only. They are not a substitute for professional medical advice, diagnosis or treatment. Never ignore professional medical advice in seeking treatment because of something you have read on the GME Website. If you think you may have a medical emergency, immediately call your doctor or dial 911.

Log In

Join GME For Free

Become a GME subscriber and gain full access to our extensive library of 700+ psychiatric medical education videos, free CME webcasts, latest research updates, and more. To sample our content, watch the featured videos below.
Join Today!