Bipolar Disorder

HONOLULU – The antipsychotic ziprasidone does not appear to work as well in patients with bipolar disorder who are either obese or hyperglycemic, according to a study funded by the drug’s maker, Pfizer.

Among 267 acutely manic patients on ziprasidone (Geodon) monotherapy for 2-3 weeks, those with body mass indexes below 28.8 kg/m² were about twice as likely to respond to ziprasidone or go into remission during treatment than were those with BMIs above 28.8 kg/m², which roughly defines the border between being overweight and obese.

Among other findings, 52% of patients below that cut-off responded to treatment; for those above it, the response rate was 37%.

Meanwhile, patients with randomly tested blood glucose levels below 140 mg/dL were more than three times more likely to go into remission and more than five times more likely to respond to treatment than were those with blood glucose levels at or above 140 mg/dL, a level rarely reached in people with normal glucose metabolism.

More than half of patients with randomly tested glucose levels below 140 mg/dL – but only 16% of patients who tested at or above that level – responded to treatment.

Obese and hyperglycemic patients also showed less improvement on Global Assessment of Functioning scores. The findings all were statistically significant.

"Patients with bipolar disorder who have elevated blood glucose and/or elevated BMI do not respond as well to ziprasidone treatment of acute mania" and "may have a lower probability of responding" to antipsychotics in general, said lead author Dr. Roger S. McIntyre, associate professor of psychiatry and pharmacology at the University of Toronto.

Obese patients might need higher-than-typical doses to overcome greater body mass, but that’s "not clear at this point. You can increase the drug dose all you want; it may not make any difference," said Dr. McIntyre, who also is head of the mood disorders psychopharmacology unit at University Health Network in Toronto.

In any case, he said the findings offer another good reason to encourage patients to lose weight, and also argue for using antipsychotics such as ziprasidone that are less likely than others to cause weight gain, since excess weight now appears to diminish the effects of antipsychotics.

The problem with hyperglycemia might be related to insulin dysregulation; there’s emerging consensus "that insulin dysregulation manifesting as hyperglycemia might be neurotoxic," Dr. McIntyre noted.

He and his colleagues pooled data from previous Pfizer studies to gauge the effects of BMI on response. "It’s intuitive if you have an [elevated] BMI, that the psychopharmacotherapies you are taking would have different distributions and different concentrations, [but] it’s almost never been studied," Dr. McIntyre said.

Based on the results, the research community needs to rethink the effect of BMI on response, he said. "It’s an important way to stratify data."

The patients in the study were at least moderately manic, with baseline Mania Rating Scale scores of 14 or greater. Remission was defined by a score dropped below 10 by the study’s end; response was defined by a greater than 50% score reduction.

HONOLULU – The antipsychotic ziprasidone does not appear to work as well in patients with bipolar disorder who are either obese or hyperglycemic, according to a study funded by the drug’s maker, Pfizer.

Among 267 acutely manic patients on ziprasidone (Geodon) monotherapy for 2-3 weeks, those with body mass indexes below 28.8 kg/m² were about twice as likely to respond to ziprasidone or go into remission during treatment than were those with BMIs above 28.8 kg/m², which roughly defines the border between being overweight and obese.

Among other findings, 52% of patients below that cut-off responded to treatment; for those above it, the response rate was 37%.

Meanwhile, patients with randomly tested blood glucose levels below 140 mg/dL were more than three times more likely to go into remission and more than five times more likely to respond to treatment than were those with blood glucose levels at or above 140 mg/dL, a level rarely reached in people with normal glucose metabolism.

More than half of patients with randomly tested glucose levels below 140 mg/dL – but only 16% of patients who tested at or above that level – responded to treatment.

Obese and hyperglycemic patients also showed less improvement on Global Assessment of Functioning scores. The findings all were statistically significant.

"Patients with bipolar disorder who have elevated blood glucose and/or elevated BMI do not respond as well to ziprasidone treatment of acute mania" and "may have a lower probability of responding" to antipsychotics in general, said lead author Dr. Roger S. McIntyre, associate professor of psychiatry and pharmacology at the University of Toronto.

Obese patients might need higher-than-typical doses to overcome greater body mass, but that’s "not clear at this point. You can increase the drug dose all you want; it may not make any difference," said Dr. McIntyre, who also is head of the mood disorders psychopharmacology unit at University Health Network in Toronto.

In any case, he said the findings offer another good reason to encourage patients to lose weight, and also argue for using antipsychotics such as ziprasidone that are less likely than others to cause weight gain, since excess weight now appears to diminish the effects of antipsychotics.

The problem with hyperglycemia might be related to insulin dysregulation; there’s emerging consensus "that insulin dysregulation manifesting as hyperglycemia might be neurotoxic," Dr. McIntyre noted.

He and his colleagues pooled data from previous Pfizer studies to gauge the effects of BMI on response. "It’s intuitive if you have an [elevated] BMI, that the psychopharmacotherapies you are taking would have different distributions and different concentrations, [but] it’s almost never been studied," Dr. McIntyre said.

Based on the results, the research community needs to rethink the effect of BMI on response, he said. "It’s an important way to stratify data."

The patients in the study were at least moderately manic, with baseline Mania Rating Scale scores of 14 or greater. Remission was defined by a score dropped below 10 by the study’s end; response was defined by a greater than 50% score reduction.

HONOLULU – The antipsychotic ziprasidone does not appear to work as well in patients with bipolar disorder who are either obese or hyperglycemic, according to a study funded by the drug’s maker, Pfizer.

Among 267 acutely manic patients on ziprasidone (Geodon) monotherapy for 2-3 weeks, those with body mass indexes below 28.8 kg/m² were about twice as likely to respond to ziprasidone or go into remission during treatment than were those with BMIs above 28.8 kg/m², which roughly defines the border between being overweight and obese.

Among other findings, 52% of patients below that cut-off responded to treatment; for those above it, the response rate was 37%.

Meanwhile, patients with randomly tested blood glucose levels below 140 mg/dL were more than three times more likely to go into remission and more than five times more likely to respond to treatment than were those with blood glucose levels at or above 140 mg/dL, a level rarely reached in people with normal glucose metabolism.

More than half of patients with randomly tested glucose levels below 140 mg/dL – but only 16% of patients who tested at or above that level – responded to treatment.

Obese and hyperglycemic patients also showed less improvement on Global Assessment of Functioning scores. The findings all were statistically significant.

"Patients with bipolar disorder who have elevated blood glucose and/or elevated BMI do not respond as well to ziprasidone treatment of acute mania" and "may have a lower probability of responding" to antipsychotics in general, said lead author Dr. Roger S. McIntyre, associate professor of psychiatry and pharmacology at the University of Toronto.

Obese patients might need higher-than-typical doses to overcome greater body mass, but that’s "not clear at this point. You can increase the drug dose all you want; it may not make any difference," said Dr. McIntyre, who also is head of the mood disorders psychopharmacology unit at University Health Network in Toronto.

In any case, he said the findings offer another good reason to encourage patients to lose weight, and also argue for using antipsychotics such as ziprasidone that are less likely than others to cause weight gain, since excess weight now appears to diminish the effects of antipsychotics.

The problem with hyperglycemia might be related to insulin dysregulation; there’s emerging consensus "that insulin dysregulation manifesting as hyperglycemia might be neurotoxic," Dr. McIntyre noted.

He and his colleagues pooled data from previous Pfizer studies to gauge the effects of BMI on response. "It’s intuitive if you have an [elevated] BMI, that the psychopharmacotherapies you are taking would have different distributions and different concentrations, [but] it’s almost never been studied," Dr. McIntyre said.

Based on the results, the research community needs to rethink the effect of BMI on response, he said. "It’s an important way to stratify data."

The patients in the study were at least moderately manic, with baseline Mania Rating Scale scores of 14 or greater. Remission was defined by a score dropped below 10 by the study’s end; response was defined by a greater than 50% score reduction.

I needed an intellectual oasis to deal with the anguish and frustration triggered by the monumental amount of misleading information that was included in the well-written article “Not all mood swings are bipolar disorder” (Current Psychiatry, February 2011, p. 38-52). Fortunately, a commentary by Dr. Irene Abramovich (“Breaking the box,” Comments & Controversies, Current Psychiatry, February 2011, p. 59) appeared as a therapeutic elixir. I believe that the “mood swings” article is filled with examples of how dangerous “cookbook” medicine can be.

Dr. Kowatch and colleagues use an expression that can be applied to the so-called diagnoses oppositional defiant disorder (ODD) and conduct disorder (CD): “Mood swings are analogous to a fever in pediatrics—they indicate something potentially is wrong with the patient, but they are not diagnostic as an isolated symptom. “ A similar concept was my position in a debate titled “Childhood conduct disorder and oppositional-defiant disorders are common manifestations of bipolar disorder” in which I argued that ODD and CD are behavioral expressions of genuine diagnoses.¹ Besides bipolar disorder, I also have seen obsessive-compulsive disorder, social anxiety disorder, and even sexual abuse labeled as “ODD” because the child refuses to be around people (such as a classroom) or is distracted by intrusive thoughts or flashbacks and turns hostile when reproached in front of the class.

In my view, Dr. Kowatch and colleagues give undeserved credit to the behavioral scales (the “cookbooks” of psychiatry) to make diagnoses and seem to miss warning signs in patients’ family history, ie, “history of depression and anxiety” (many times this translates as agitated/dysphoric mania) and “drinking problems, “ which frequently is found in undiagnosed bipolar spectrum patients who use alcohol to “shoot down” racing thoughts that interfere with normal sleep.

From January 2010 to February 2011, I reviewed charts and interviewed patients and families of 1, 654 patients with diagnoses of attention-deficit/hyperactivity disorder co-morbid with ODD, bipolar disorder, generalized anxiety disorder, and even 2 diagnoses that are not allowed by DSM rules: autism and mental retardation. The data from this study, which covers 12 counties that represent the 5 geographical areas of Florida, are being analyzed. In the meantime, I refer readers to my poster presentation from the 2010 U. S. Psychiatric and Mental Health Congress “Extinction of oppositional-defiant symptoms following treatment with mood stabilizers. “² In this study 44 patients were followed for at least 5 years (10 patients were observed for 7 years and a similar number for 6) and none had “oppositional” behavior after the diagnoses were treated. One caveat is that I placed antipsychotics in the same category as conventional mood stabilizers because 5 patients considered to be “inattentive” and “oppositional” actually had schizophrenia.

I oppose the authors’ assertion that “it can be difficult to differentiate the mood swings and symptoms of ODD from those of pediatric BD. “ My experience is that it is simple if we consider all diagnostic possibilities and obtain a thorough family history, which usually includes alcoholism, cannabis abuse, moodiness, suicide completion, unstable lifestyle, etc.

Manuel Mota-Castillo, MD
Assistant Clinical Professor
St. Matthews University
Voluntary Faculty
University of Central Florida
Lake Mary, FL

References

1. Mota-Castillo M, Steiner H. Childhood conduct disorder and oppositional-defiant disorder are common manifestations of bipolar disorder pro and con. Journal of Bipolar Disorders: Reviews and Commentaries. 2005;3:3,15-17.

2. Mota-Castillo M. Extinction of oppositional-defiant symptoms following treatment with mood stabilizers. Poster presented at: 23rd Annual U.S. Psychiatric and Mental Health Congress; November 20 2010; Orlando, FL.

The authors respond

We never suggested that clinicians use “cookbook medicine. “ The “behavioral scales” we recommended in our article are well-validated and reliable tools that allow a clinician to effectively elicit a great deal of useful information from patients and their parents about presenting problems and symptoms. This information can be used with other clinical information to make an accurate diagnosis and subsequent treatment plan.

The purpose of our article was to share our experiences in the differential diagnosis of mood swings in children and adolescents and to suggest that there are other diagnoses that cause mood swings besides bipolar disorder. Although a family history of mood disorders is important, it is also important to recognize that a recent, state-of-the-art study by Birmaher et al¹ reported that 10% of children of parents with bipolar disorder had a bipolar spectrum disorder. That means that 90% did not have bipolar disorder. It is important to remember this when evaluating children of parents with bipolar disorder. Although these children’s risk for developing bipolar disorder is increased compared with the general population, it is more likely that they will not develop bipolar disorder.

Robert A. Kowatch, MD, PhD
Professor of Psychiatry and Pediatrics

Erin Monroe, CNS
Clinical Nurse Specialist
Division of Psychiatry

Sergio V. Delgado, MD
Associate Professor of Psychiatry
and Pediatrics
Cincinnati Children’s Hospital Medical Center
Cincinnati, OH

I needed an intellectual oasis to deal with the anguish and frustration triggered by the monumental amount of misleading information that was included in the well-written article “Not all mood swings are bipolar disorder” (Current Psychiatry, February 2011, p. 38-52). Fortunately, a commentary by Dr. Irene Abramovich (“Breaking the box,” Comments & Controversies, Current Psychiatry, February 2011, p. 59) appeared as a therapeutic elixir. I believe that the “mood swings” article is filled with examples of how dangerous “cookbook” medicine can be.

Dr. Kowatch and colleagues use an expression that can be applied to the so-called diagnoses oppositional defiant disorder (ODD) and conduct disorder (CD): “Mood swings are analogous to a fever in pediatrics—they indicate something potentially is wrong with the patient, but they are not diagnostic as an isolated symptom. “ A similar concept was my position in a debate titled “Childhood conduct disorder and oppositional-defiant disorders are common manifestations of bipolar disorder” in which I argued that ODD and CD are behavioral expressions of genuine diagnoses.¹ Besides bipolar disorder, I also have seen obsessive-compulsive disorder, social anxiety disorder, and even sexual abuse labeled as “ODD” because the child refuses to be around people (such as a classroom) or is distracted by intrusive thoughts or flashbacks and turns hostile when reproached in front of the class.

In my view, Dr. Kowatch and colleagues give undeserved credit to the behavioral scales (the “cookbooks” of psychiatry) to make diagnoses and seem to miss warning signs in patients’ family history, ie, “history of depression and anxiety” (many times this translates as agitated/dysphoric mania) and “drinking problems, “ which frequently is found in undiagnosed bipolar spectrum patients who use alcohol to “shoot down” racing thoughts that interfere with normal sleep.

From January 2010 to February 2011, I reviewed charts and interviewed patients and families of 1, 654 patients with diagnoses of attention-deficit/hyperactivity disorder co-morbid with ODD, bipolar disorder, generalized anxiety disorder, and even 2 diagnoses that are not allowed by DSM rules: autism and mental retardation. The data from this study, which covers 12 counties that represent the 5 geographical areas of Florida, are being analyzed. In the meantime, I refer readers to my poster presentation from the 2010 U. S. Psychiatric and Mental Health Congress “Extinction of oppositional-defiant symptoms following treatment with mood stabilizers. “² In this study 44 patients were followed for at least 5 years (10 patients were observed for 7 years and a similar number for 6) and none had “oppositional” behavior after the diagnoses were treated. One caveat is that I placed antipsychotics in the same category as conventional mood stabilizers because 5 patients considered to be “inattentive” and “oppositional” actually had schizophrenia.

I oppose the authors’ assertion that “it can be difficult to differentiate the mood swings and symptoms of ODD from those of pediatric BD. “ My experience is that it is simple if we consider all diagnostic possibilities and obtain a thorough family history, which usually includes alcoholism, cannabis abuse, moodiness, suicide completion, unstable lifestyle, etc.

Manuel Mota-Castillo, MD
Assistant Clinical Professor
St. Matthews University
Voluntary Faculty
University of Central Florida
Lake Mary, FL

References

1. Mota-Castillo M, Steiner H. Childhood conduct disorder and oppositional-defiant disorder are common manifestations of bipolar disorder pro and con. Journal of Bipolar Disorders: Reviews and Commentaries. 2005;3:3,15-17.

2. Mota-Castillo M. Extinction of oppositional-defiant symptoms following treatment with mood stabilizers. Poster presented at: 23rd Annual U.S. Psychiatric and Mental Health Congress; November 20 2010; Orlando, FL.

The authors respond

We never suggested that clinicians use “cookbook medicine. “ The “behavioral scales” we recommended in our article are well-validated and reliable tools that allow a clinician to effectively elicit a great deal of useful information from patients and their parents about presenting problems and symptoms. This information can be used with other clinical information to make an accurate diagnosis and subsequent treatment plan.

The purpose of our article was to share our experiences in the differential diagnosis of mood swings in children and adolescents and to suggest that there are other diagnoses that cause mood swings besides bipolar disorder. Although a family history of mood disorders is important, it is also important to recognize that a recent, state-of-the-art study by Birmaher et al¹ reported that 10% of children of parents with bipolar disorder had a bipolar spectrum disorder. That means that 90% did not have bipolar disorder. It is important to remember this when evaluating children of parents with bipolar disorder. Although these children’s risk for developing bipolar disorder is increased compared with the general population, it is more likely that they will not develop bipolar disorder.

Robert A. Kowatch, MD, PhD
Professor of Psychiatry and Pediatrics

Erin Monroe, CNS
Clinical Nurse Specialist
Division of Psychiatry

Sergio V. Delgado, MD
Associate Professor of Psychiatry
and Pediatrics
Cincinnati Children’s Hospital Medical Center
Cincinnati, OH

I needed an intellectual oasis to deal with the anguish and frustration triggered by the monumental amount of misleading information that was included in the well-written article “Not all mood swings are bipolar disorder” (Current Psychiatry, February 2011, p. 38-52). Fortunately, a commentary by Dr. Irene Abramovich (“Breaking the box,” Comments & Controversies, Current Psychiatry, February 2011, p. 59) appeared as a therapeutic elixir. I believe that the “mood swings” article is filled with examples of how dangerous “cookbook” medicine can be.

Dr. Kowatch and colleagues use an expression that can be applied to the so-called diagnoses oppositional defiant disorder (ODD) and conduct disorder (CD): “Mood swings are analogous to a fever in pediatrics—they indicate something potentially is wrong with the patient, but they are not diagnostic as an isolated symptom. “ A similar concept was my position in a debate titled “Childhood conduct disorder and oppositional-defiant disorders are common manifestations of bipolar disorder” in which I argued that ODD and CD are behavioral expressions of genuine diagnoses.¹ Besides bipolar disorder, I also have seen obsessive-compulsive disorder, social anxiety disorder, and even sexual abuse labeled as “ODD” because the child refuses to be around people (such as a classroom) or is distracted by intrusive thoughts or flashbacks and turns hostile when reproached in front of the class.

In my view, Dr. Kowatch and colleagues give undeserved credit to the behavioral scales (the “cookbooks” of psychiatry) to make diagnoses and seem to miss warning signs in patients’ family history, ie, “history of depression and anxiety” (many times this translates as agitated/dysphoric mania) and “drinking problems, “ which frequently is found in undiagnosed bipolar spectrum patients who use alcohol to “shoot down” racing thoughts that interfere with normal sleep.

From January 2010 to February 2011, I reviewed charts and interviewed patients and families of 1, 654 patients with diagnoses of attention-deficit/hyperactivity disorder co-morbid with ODD, bipolar disorder, generalized anxiety disorder, and even 2 diagnoses that are not allowed by DSM rules: autism and mental retardation. The data from this study, which covers 12 counties that represent the 5 geographical areas of Florida, are being analyzed. In the meantime, I refer readers to my poster presentation from the 2010 U. S. Psychiatric and Mental Health Congress “Extinction of oppositional-defiant symptoms following treatment with mood stabilizers. “² In this study 44 patients were followed for at least 5 years (10 patients were observed for 7 years and a similar number for 6) and none had “oppositional” behavior after the diagnoses were treated. One caveat is that I placed antipsychotics in the same category as conventional mood stabilizers because 5 patients considered to be “inattentive” and “oppositional” actually had schizophrenia.

I oppose the authors’ assertion that “it can be difficult to differentiate the mood swings and symptoms of ODD from those of pediatric BD. “ My experience is that it is simple if we consider all diagnostic possibilities and obtain a thorough family history, which usually includes alcoholism, cannabis abuse, moodiness, suicide completion, unstable lifestyle, etc.

Manuel Mota-Castillo, MD
Assistant Clinical Professor
St. Matthews University
Voluntary Faculty
University of Central Florida
Lake Mary, FL

References

1. Mota-Castillo M, Steiner H. Childhood conduct disorder and oppositional-defiant disorder are common manifestations of bipolar disorder pro and con. Journal of Bipolar Disorders: Reviews and Commentaries. 2005;3:3,15-17.

2. Mota-Castillo M. Extinction of oppositional-defiant symptoms following treatment with mood stabilizers. Poster presented at: 23rd Annual U.S. Psychiatric and Mental Health Congress; November 20 2010; Orlando, FL.

The authors respond

We never suggested that clinicians use “cookbook medicine. “ The “behavioral scales” we recommended in our article are well-validated and reliable tools that allow a clinician to effectively elicit a great deal of useful information from patients and their parents about presenting problems and symptoms. This information can be used with other clinical information to make an accurate diagnosis and subsequent treatment plan.

The purpose of our article was to share our experiences in the differential diagnosis of mood swings in children and adolescents and to suggest that there are other diagnoses that cause mood swings besides bipolar disorder. Although a family history of mood disorders is important, it is also important to recognize that a recent, state-of-the-art study by Birmaher et al¹ reported that 10% of children of parents with bipolar disorder had a bipolar spectrum disorder. That means that 90% did not have bipolar disorder. It is important to remember this when evaluating children of parents with bipolar disorder. Although these children’s risk for developing bipolar disorder is increased compared with the general population, it is more likely that they will not develop bipolar disorder.

Robert A. Kowatch, MD, PhD
Professor of Psychiatry and Pediatrics

Erin Monroe, CNS
Clinical Nurse Specialist
Division of Psychiatry

Sergio V. Delgado, MD
Associate Professor of Psychiatry
and Pediatrics
Cincinnati Children’s Hospital Medical Center
Cincinnati, OH

I thought Dr. Kowatch and colleagues took an important first step in pointing out that not all mood swings in children and adolescents are symptoms of bipolar disorder (“Not all mood swings are bipolar disorder,“ Current Psychiatry, February 2011, p. 38-52). They reviewed some of the other psychiatric conditions known to cause labile moods. One glaring omission is borderline personality disorder (BPD).

I am the medical director of a specialized unit that uses dialectical behavioral therapy (DBT) to treat children and adolescents with BPD. We have treated approximately 300 young women on the residential unit and many present similarly: multiple hospitalizations, multiple robust yet failed medication trials, severe and recurrent self-injury, suicide attempts, and a large degree of hopelessness. Most arrive with previous diagnoses of mood disorder not otherwise specified, bipolar disorder, oppositional defiant disorder, attention-deficit/hyperactivity disorder, and others. It is when their outpatient psychiatrists and mental health teams have grown frustrated at the lack of enduring progress and are faced with the treatment demands of the borderline patient that a BPD diagnosis is considered. Even though research suggests that BPD—or at least some of its symptoms—begins in the late latency period of childhood,¹ treatment typically is not sought until late adolescence. This is the case despite the fact that BPD has a better prognosis than other serious mental illnesses, such as bipolar disorder.^2,3 Adult BPD patients almost universally recognize that their inability to regulate their mood started in late childhood and early adolescence. Structural and functional neuroimaging has revealed a dysfunctional network of brain regions that seem to mediate important aspects of BPD symptomatology.^4-6

These children have marked mood swings and great difficulty regulating their moods. The mood swings of BPD are not responsive to current medication unless there is comorbid bipolar disorder, in which case treatment with mood stabilizers helps improve vegetative symptoms such as sleep and energy, and reduce racing thoughts, pressured speech, and irritability. What these medications do not treat is the “reactive” mood swings that are characteristic of BPD. The mood reactivity often is triggered by interpersonal or intrapersonal conflict and rarely is long-lived.

Many children and adolescents are moody and most do not have a major psychiatric disorder. Of those who do, it is a great risk to patients’ health to not consider BPD, especially given new and empirically validated treatments, such as DBT. Astute clinicians should keep this diagnosis in mind when treating adolescents with moodiness, particularly when the mood is predominantly reactive to life’s stressors, when other features of the presentation do not fit neatly into a bipolar picture, and when multiple medications fail. On our unit, we have seen that the cognitive-behavioral strategies of DBT help patients even when BPD is not the diagnosis.

I would like to thank Dr. Kowatch and colleagues for expanding our thinking on mood swings and encourage readers to go one step further.

Blaise Aguirre, MD
Medical Director
Adolescent DBT Residential Program
McLean Hospital
Belmont, MA
Instructor in Psychiatry
Harvard Medical School
Boston, MA

References

1. Zanarini MC, Frankenburg FR, Khera GS, et al. Treatment histories of borderline inpatients. Compr Psychiatry. 2001;42:144-150.

2. Tohen M, Hennen J, Zarate CM Jr, et al. Two-year syndromal and functional recovery in 219 cases of first-episode major affective disorder with psychotic features. Am J Psychiatry. 2000;157:220-228.

3. Coryell W, Endicott J, Maser JD, et al. The likelihood of recurrence in bipolar affective disorder: the importance of episode recency. J Affect Disord. 1995;33:201-206.

4. De La Fuente JM, Goldman S, Stanus E, et al. Brain glucose metabolism in borderline personality disorder. J Psychiatr Res. 1997;31:531-541.

5. Soloff PH, Meltzer CC, Becker C, et al. Impulsivity and prefrontal hypometabolism in borderline personality disorder. Psychiatry Res. 2003;123:153-163.

6. Juengling FD, Schmahl C, Hesslinger B, et al. Positron emission tomography in female patients with borderline personality disorder. J Psychiatr Res. 2003;37:109-115.

The authors respond

We welcome comments about the importance of a thorough diagnostic evaluation to tease out possible etiologies of “mood swings, “ including psychosocial factors, as in personality disorders. Nevertheless, the debate about diagnosing personality disorders in children and adolescents is not settled. Developmentally, children and adolescents have continuous changes in biology and brain function. There is significantly more empirical evidence of reactive attachment disorders in childhood and adolescence that integrate the affective changes seen in children who live in chaotic environments. DSM defines BPD as a pervasive pattern of instability of interpersonal relationships that begins by early adulthood.¹ Many children with diagnoses of posttraumatic stress disorder, mood disorder, bipolar disorder, oppositional defiant disorder, attention-deficit/hyperactivity disorder, etc. also can have difficulties in relating to others caused by their neurobiologic deficits, which also may limit response to medication. Furthermore, children with learning disorders also can misperceive motives of others and thus have pervasive patterns of relational instability.

The research Dr. Aguirre suggested is based on treatment histories and not rigorous study methodology. Most empirical evidence of personality disorders is strongly influenced by psychoanalytic literature regarding object relations, which is in flux because of emerging attention to attachment theory and progress in neurologic studies in the evaluation of temperamental variations related to the influence of mirror neurons.²

We also take issue with the comment that “BPD has a better prognosis than other serious men tal illn esses, such as bipolar disorder” There have been significant efforts in studying the role family can have in the outcomes of mood disorder treatment.³

Finally, there is evidence that in adults medication can be beneficial in treating the affective deregulation of patients with BPD who do not have comorbid disorders.⁴

Robert A. Kowatch, MD, PhD
Professor of Psychiatry and Pediatrics

Erin Monroe, CNS
Clinical Nurse Specialist
Division of Psychiatry

Sergio V. Delgado, MD
Associate Professor of Psychiatry
and Pediatrics
Cincinnati Children’s Hospital Medical Center
Cincinnati, OH

I thought Dr. Kowatch and colleagues took an important first step in pointing out that not all mood swings in children and adolescents are symptoms of bipolar disorder (“Not all mood swings are bipolar disorder,“ Current Psychiatry, February 2011, p. 38-52). They reviewed some of the other psychiatric conditions known to cause labile moods. One glaring omission is borderline personality disorder (BPD).

I am the medical director of a specialized unit that uses dialectical behavioral therapy (DBT) to treat children and adolescents with BPD. We have treated approximately 300 young women on the residential unit and many present similarly: multiple hospitalizations, multiple robust yet failed medication trials, severe and recurrent self-injury, suicide attempts, and a large degree of hopelessness. Most arrive with previous diagnoses of mood disorder not otherwise specified, bipolar disorder, oppositional defiant disorder, attention-deficit/hyperactivity disorder, and others. It is when their outpatient psychiatrists and mental health teams have grown frustrated at the lack of enduring progress and are faced with the treatment demands of the borderline patient that a BPD diagnosis is considered. Even though research suggests that BPD—or at least some of its symptoms—begins in the late latency period of childhood,¹ treatment typically is not sought until late adolescence. This is the case despite the fact that BPD has a better prognosis than other serious mental illnesses, such as bipolar disorder.^2,3 Adult BPD patients almost universally recognize that their inability to regulate their mood started in late childhood and early adolescence. Structural and functional neuroimaging has revealed a dysfunctional network of brain regions that seem to mediate important aspects of BPD symptomatology.^4-6

These children have marked mood swings and great difficulty regulating their moods. The mood swings of BPD are not responsive to current medication unless there is comorbid bipolar disorder, in which case treatment with mood stabilizers helps improve vegetative symptoms such as sleep and energy, and reduce racing thoughts, pressured speech, and irritability. What these medications do not treat is the “reactive” mood swings that are characteristic of BPD. The mood reactivity often is triggered by interpersonal or intrapersonal conflict and rarely is long-lived.

Many children and adolescents are moody and most do not have a major psychiatric disorder. Of those who do, it is a great risk to patients’ health to not consider BPD, especially given new and empirically validated treatments, such as DBT. Astute clinicians should keep this diagnosis in mind when treating adolescents with moodiness, particularly when the mood is predominantly reactive to life’s stressors, when other features of the presentation do not fit neatly into a bipolar picture, and when multiple medications fail. On our unit, we have seen that the cognitive-behavioral strategies of DBT help patients even when BPD is not the diagnosis.

I would like to thank Dr. Kowatch and colleagues for expanding our thinking on mood swings and encourage readers to go one step further.

Blaise Aguirre, MD
Medical Director
Adolescent DBT Residential Program
McLean Hospital
Belmont, MA
Instructor in Psychiatry
Harvard Medical School
Boston, MA

References

1. Zanarini MC, Frankenburg FR, Khera GS, et al. Treatment histories of borderline inpatients. Compr Psychiatry. 2001;42:144-150.

2. Tohen M, Hennen J, Zarate CM Jr, et al. Two-year syndromal and functional recovery in 219 cases of first-episode major affective disorder with psychotic features. Am J Psychiatry. 2000;157:220-228.

3. Coryell W, Endicott J, Maser JD, et al. The likelihood of recurrence in bipolar affective disorder: the importance of episode recency. J Affect Disord. 1995;33:201-206.

4. De La Fuente JM, Goldman S, Stanus E, et al. Brain glucose metabolism in borderline personality disorder. J Psychiatr Res. 1997;31:531-541.

5. Soloff PH, Meltzer CC, Becker C, et al. Impulsivity and prefrontal hypometabolism in borderline personality disorder. Psychiatry Res. 2003;123:153-163.

6. Juengling FD, Schmahl C, Hesslinger B, et al. Positron emission tomography in female patients with borderline personality disorder. J Psychiatr Res. 2003;37:109-115.

The authors respond

We welcome comments about the importance of a thorough diagnostic evaluation to tease out possible etiologies of “mood swings, “ including psychosocial factors, as in personality disorders. Nevertheless, the debate about diagnosing personality disorders in children and adolescents is not settled. Developmentally, children and adolescents have continuous changes in biology and brain function. There is significantly more empirical evidence of reactive attachment disorders in childhood and adolescence that integrate the affective changes seen in children who live in chaotic environments. DSM defines BPD as a pervasive pattern of instability of interpersonal relationships that begins by early adulthood.¹ Many children with diagnoses of posttraumatic stress disorder, mood disorder, bipolar disorder, oppositional defiant disorder, attention-deficit/hyperactivity disorder, etc. also can have difficulties in relating to others caused by their neurobiologic deficits, which also may limit response to medication. Furthermore, children with learning disorders also can misperceive motives of others and thus have pervasive patterns of relational instability.

The research Dr. Aguirre suggested is based on treatment histories and not rigorous study methodology. Most empirical evidence of personality disorders is strongly influenced by psychoanalytic literature regarding object relations, which is in flux because of emerging attention to attachment theory and progress in neurologic studies in the evaluation of temperamental variations related to the influence of mirror neurons.²

We also take issue with the comment that “BPD has a better prognosis than other serious men tal illn esses, such as bipolar disorder” There have been significant efforts in studying the role family can have in the outcomes of mood disorder treatment.³

Finally, there is evidence that in adults medication can be beneficial in treating the affective deregulation of patients with BPD who do not have comorbid disorders.⁴

Robert A. Kowatch, MD, PhD
Professor of Psychiatry and Pediatrics

Erin Monroe, CNS
Clinical Nurse Specialist
Division of Psychiatry

Sergio V. Delgado, MD
Associate Professor of Psychiatry
and Pediatrics
Cincinnati Children’s Hospital Medical Center
Cincinnati, OH

I thought Dr. Kowatch and colleagues took an important first step in pointing out that not all mood swings in children and adolescents are symptoms of bipolar disorder (“Not all mood swings are bipolar disorder,“ Current Psychiatry, February 2011, p. 38-52). They reviewed some of the other psychiatric conditions known to cause labile moods. One glaring omission is borderline personality disorder (BPD).

I am the medical director of a specialized unit that uses dialectical behavioral therapy (DBT) to treat children and adolescents with BPD. We have treated approximately 300 young women on the residential unit and many present similarly: multiple hospitalizations, multiple robust yet failed medication trials, severe and recurrent self-injury, suicide attempts, and a large degree of hopelessness. Most arrive with previous diagnoses of mood disorder not otherwise specified, bipolar disorder, oppositional defiant disorder, attention-deficit/hyperactivity disorder, and others. It is when their outpatient psychiatrists and mental health teams have grown frustrated at the lack of enduring progress and are faced with the treatment demands of the borderline patient that a BPD diagnosis is considered. Even though research suggests that BPD—or at least some of its symptoms—begins in the late latency period of childhood,¹ treatment typically is not sought until late adolescence. This is the case despite the fact that BPD has a better prognosis than other serious mental illnesses, such as bipolar disorder.^2,3 Adult BPD patients almost universally recognize that their inability to regulate their mood started in late childhood and early adolescence. Structural and functional neuroimaging has revealed a dysfunctional network of brain regions that seem to mediate important aspects of BPD symptomatology.^4-6

These children have marked mood swings and great difficulty regulating their moods. The mood swings of BPD are not responsive to current medication unless there is comorbid bipolar disorder, in which case treatment with mood stabilizers helps improve vegetative symptoms such as sleep and energy, and reduce racing thoughts, pressured speech, and irritability. What these medications do not treat is the “reactive” mood swings that are characteristic of BPD. The mood reactivity often is triggered by interpersonal or intrapersonal conflict and rarely is long-lived.

Many children and adolescents are moody and most do not have a major psychiatric disorder. Of those who do, it is a great risk to patients’ health to not consider BPD, especially given new and empirically validated treatments, such as DBT. Astute clinicians should keep this diagnosis in mind when treating adolescents with moodiness, particularly when the mood is predominantly reactive to life’s stressors, when other features of the presentation do not fit neatly into a bipolar picture, and when multiple medications fail. On our unit, we have seen that the cognitive-behavioral strategies of DBT help patients even when BPD is not the diagnosis.

I would like to thank Dr. Kowatch and colleagues for expanding our thinking on mood swings and encourage readers to go one step further.

Blaise Aguirre, MD
Medical Director
Adolescent DBT Residential Program
McLean Hospital
Belmont, MA
Instructor in Psychiatry
Harvard Medical School
Boston, MA

References

1. Zanarini MC, Frankenburg FR, Khera GS, et al. Treatment histories of borderline inpatients. Compr Psychiatry. 2001;42:144-150.

2. Tohen M, Hennen J, Zarate CM Jr, et al. Two-year syndromal and functional recovery in 219 cases of first-episode major affective disorder with psychotic features. Am J Psychiatry. 2000;157:220-228.

3. Coryell W, Endicott J, Maser JD, et al. The likelihood of recurrence in bipolar affective disorder: the importance of episode recency. J Affect Disord. 1995;33:201-206.

4. De La Fuente JM, Goldman S, Stanus E, et al. Brain glucose metabolism in borderline personality disorder. J Psychiatr Res. 1997;31:531-541.

5. Soloff PH, Meltzer CC, Becker C, et al. Impulsivity and prefrontal hypometabolism in borderline personality disorder. Psychiatry Res. 2003;123:153-163.

6. Juengling FD, Schmahl C, Hesslinger B, et al. Positron emission tomography in female patients with borderline personality disorder. J Psychiatr Res. 2003;37:109-115.

The authors respond

We welcome comments about the importance of a thorough diagnostic evaluation to tease out possible etiologies of “mood swings, “ including psychosocial factors, as in personality disorders. Nevertheless, the debate about diagnosing personality disorders in children and adolescents is not settled. Developmentally, children and adolescents have continuous changes in biology and brain function. There is significantly more empirical evidence of reactive attachment disorders in childhood and adolescence that integrate the affective changes seen in children who live in chaotic environments. DSM defines BPD as a pervasive pattern of instability of interpersonal relationships that begins by early adulthood.¹ Many children with diagnoses of posttraumatic stress disorder, mood disorder, bipolar disorder, oppositional defiant disorder, attention-deficit/hyperactivity disorder, etc. also can have difficulties in relating to others caused by their neurobiologic deficits, which also may limit response to medication. Furthermore, children with learning disorders also can misperceive motives of others and thus have pervasive patterns of relational instability.

The research Dr. Aguirre suggested is based on treatment histories and not rigorous study methodology. Most empirical evidence of personality disorders is strongly influenced by psychoanalytic literature regarding object relations, which is in flux because of emerging attention to attachment theory and progress in neurologic studies in the evaluation of temperamental variations related to the influence of mirror neurons.²

We also take issue with the comment that “BPD has a better prognosis than other serious men tal illn esses, such as bipolar disorder” There have been significant efforts in studying the role family can have in the outcomes of mood disorder treatment.³

Finally, there is evidence that in adults medication can be beneficial in treating the affective deregulation of patients with BPD who do not have comorbid disorders.⁴

Robert A. Kowatch, MD, PhD
Professor of Psychiatry and Pediatrics

Erin Monroe, CNS
Clinical Nurse Specialist
Division of Psychiatry

Sergio V. Delgado, MD
Associate Professor of Psychiatry
and Pediatrics
Cincinnati Children’s Hospital Medical Center
Cincinnati, OH

You’ve heard about the 2 certainties in life: death and taxes. In psychiatric practice with complex and chronic patients, there is a third certainty: polypharmacy. It ranges from thoughtful to indiscriminate and seems to be entrenched in clinical practice, possibly reflecting practitioners’ desperation in trying to manage severely ill, treatment-resistant patients, usually in the absence of evidence-based guidelines.

I never fail to encounter polypharmacy in hospitals or clinics where I consult. I always wondered how the patient’s doctor knew which drug was exerting a therapeutic effect or which drug was causing side effects (parkinsonism, akathisia, sedation, orthostasis, dizziness, headache, blurry vision, etc. ). Over time, I came to categorize polypharmacy into 4 subtypes that span the spectrum from sensible to absurd. Here is my personal classification, which I trust that you, my readers, have witnessed as well.

Necessary polypharmacy. This variant of polypharmacy is evidence-based and proven in double-blind studies to be more effective than monotherapy. The most prominent example is adding an atypical antipsychotic to a mood stabilizer in bipolar mania. In fact, the superior efficacy of combination therapy in bipolar disorder is one of the oldest forms of rational polypharmacy, is supported by FDA trials, and is indicated whenever mood stabilizer monotherapy is not sufficient. For example, combining lithium and valproate is superior to either drug alone. Another example of FDA-approved combinations is combining small doses of an atypical antipsychotic to an antidepressant for treatment-resistant depression.

Reasonable polypharmacy. Although many of the combinations in this category are not FDA-approved, controlled studies support their use for suffering patients. Examples include:

• An atypical antipsychotic added to a selective serotonin reuptake inhibitor (SSRI) for obsessive-compulsive disorder (OCD) patients who do not improve on SSRI monotherapy.
• Modafinil added to clozapine in patients who suffer substantial and persistent daytime sedation or somnolence.
• Combining 2 antidepressants for major depressive disorder patients who partially respond to 1 antidepressant.
• Combining a mood stabilizer with an antidepressant for bipolar depression to prevent mood switching.

Ridiculous polypharmacy. The sky is the limit to the variations and degrees of ridiculous polypharmacy, but the theme is the same: an absurd concoction of psychotropic drugs across several classes, often including multiple agents from 1 or several classes. Here are examples I have seen in patient records:

• Two atypicals, an anticholinergic, a mood stabilizer, an antidepressant, and 2 benzodiazepines.
• Three antipsychotics (2 atypicals and 1 typical), 2 antidepressants, 3 sedative/hypnotics, and an anticonvulsant for weight control.
• This one takes the cake: 6 antipsychotics (2 typicals and 4 atypicals), plus an anticholinergic, 3 mood stabilizers, 2 antidepressants, 2 sleeping pills, a hypoglycemic agent, 2 antihypertensives, and a statin.

Hazardous polypharmacy. In this category, serious medical complications, toxic effects, or death may occur because of careless combinations of drugs that may interact to produce dangerous kinetic interactions or exacerbate a pre-existing medical condition. Examples include:

• Combining 1 psychotropic with another that may inhibit its metabolism (eg, prescribing fluvoxamine to a severely psychotic patient who developed OCD while receiving clozapine). There have been several toxic reactions and even death because fluvoxamine inhibits cytochrome 1A2, which metabolizes clozapine, thus increasing clozapine blood level by 400% to 500%.
• Combining 2 injectable drugs for agitation that may cause a serious medical complication. An example would be injecting a benzodiazepine such as lorazepam in a patient receiving olanzapine IM, which can cause severe respiratory depression or death.
• Combining several drugs, each of which may prolong the QTc interval, resulting in syncope or torsade de pointes.

Psychopharmacology can relieve the terrible anguish of psychosis, depression, or anxiety, but it also can carry iatrogenic risks if it is not based on scientific evidence. The practice of psychopharmacology requires the fully integrated skills of medical and psychiatric training to maximize benefit while avoiding harm. It also requires basic arithmetic skills: to consider subtracting drugs, not only adding them!

You’ve heard about the 2 certainties in life: death and taxes. In psychiatric practice with complex and chronic patients, there is a third certainty: polypharmacy. It ranges from thoughtful to indiscriminate and seems to be entrenched in clinical practice, possibly reflecting practitioners’ desperation in trying to manage severely ill, treatment-resistant patients, usually in the absence of evidence-based guidelines.

I never fail to encounter polypharmacy in hospitals or clinics where I consult. I always wondered how the patient’s doctor knew which drug was exerting a therapeutic effect or which drug was causing side effects (parkinsonism, akathisia, sedation, orthostasis, dizziness, headache, blurry vision, etc. ). Over time, I came to categorize polypharmacy into 4 subtypes that span the spectrum from sensible to absurd. Here is my personal classification, which I trust that you, my readers, have witnessed as well.

Necessary polypharmacy. This variant of polypharmacy is evidence-based and proven in double-blind studies to be more effective than monotherapy. The most prominent example is adding an atypical antipsychotic to a mood stabilizer in bipolar mania. In fact, the superior efficacy of combination therapy in bipolar disorder is one of the oldest forms of rational polypharmacy, is supported by FDA trials, and is indicated whenever mood stabilizer monotherapy is not sufficient. For example, combining lithium and valproate is superior to either drug alone. Another example of FDA-approved combinations is combining small doses of an atypical antipsychotic to an antidepressant for treatment-resistant depression.

Reasonable polypharmacy. Although many of the combinations in this category are not FDA-approved, controlled studies support their use for suffering patients. Examples include:

• An atypical antipsychotic added to a selective serotonin reuptake inhibitor (SSRI) for obsessive-compulsive disorder (OCD) patients who do not improve on SSRI monotherapy.
• Modafinil added to clozapine in patients who suffer substantial and persistent daytime sedation or somnolence.
• Combining 2 antidepressants for major depressive disorder patients who partially respond to 1 antidepressant.
• Combining a mood stabilizer with an antidepressant for bipolar depression to prevent mood switching.

Ridiculous polypharmacy. The sky is the limit to the variations and degrees of ridiculous polypharmacy, but the theme is the same: an absurd concoction of psychotropic drugs across several classes, often including multiple agents from 1 or several classes. Here are examples I have seen in patient records:

• Two atypicals, an anticholinergic, a mood stabilizer, an antidepressant, and 2 benzodiazepines.
• Three antipsychotics (2 atypicals and 1 typical), 2 antidepressants, 3 sedative/hypnotics, and an anticonvulsant for weight control.
• This one takes the cake: 6 antipsychotics (2 typicals and 4 atypicals), plus an anticholinergic, 3 mood stabilizers, 2 antidepressants, 2 sleeping pills, a hypoglycemic agent, 2 antihypertensives, and a statin.

Hazardous polypharmacy. In this category, serious medical complications, toxic effects, or death may occur because of careless combinations of drugs that may interact to produce dangerous kinetic interactions or exacerbate a pre-existing medical condition. Examples include:

• Combining 1 psychotropic with another that may inhibit its metabolism (eg, prescribing fluvoxamine to a severely psychotic patient who developed OCD while receiving clozapine). There have been several toxic reactions and even death because fluvoxamine inhibits cytochrome 1A2, which metabolizes clozapine, thus increasing clozapine blood level by 400% to 500%.
• Combining 2 injectable drugs for agitation that may cause a serious medical complication. An example would be injecting a benzodiazepine such as lorazepam in a patient receiving olanzapine IM, which can cause severe respiratory depression or death.
• Combining several drugs, each of which may prolong the QTc interval, resulting in syncope or torsade de pointes.

Psychopharmacology can relieve the terrible anguish of psychosis, depression, or anxiety, but it also can carry iatrogenic risks if it is not based on scientific evidence. The practice of psychopharmacology requires the fully integrated skills of medical and psychiatric training to maximize benefit while avoiding harm. It also requires basic arithmetic skills: to consider subtracting drugs, not only adding them!

You’ve heard about the 2 certainties in life: death and taxes. In psychiatric practice with complex and chronic patients, there is a third certainty: polypharmacy. It ranges from thoughtful to indiscriminate and seems to be entrenched in clinical practice, possibly reflecting practitioners’ desperation in trying to manage severely ill, treatment-resistant patients, usually in the absence of evidence-based guidelines.

I never fail to encounter polypharmacy in hospitals or clinics where I consult. I always wondered how the patient’s doctor knew which drug was exerting a therapeutic effect or which drug was causing side effects (parkinsonism, akathisia, sedation, orthostasis, dizziness, headache, blurry vision, etc. ). Over time, I came to categorize polypharmacy into 4 subtypes that span the spectrum from sensible to absurd. Here is my personal classification, which I trust that you, my readers, have witnessed as well.

Necessary polypharmacy. This variant of polypharmacy is evidence-based and proven in double-blind studies to be more effective than monotherapy. The most prominent example is adding an atypical antipsychotic to a mood stabilizer in bipolar mania. In fact, the superior efficacy of combination therapy in bipolar disorder is one of the oldest forms of rational polypharmacy, is supported by FDA trials, and is indicated whenever mood stabilizer monotherapy is not sufficient. For example, combining lithium and valproate is superior to either drug alone. Another example of FDA-approved combinations is combining small doses of an atypical antipsychotic to an antidepressant for treatment-resistant depression.

Reasonable polypharmacy. Although many of the combinations in this category are not FDA-approved, controlled studies support their use for suffering patients. Examples include:

• An atypical antipsychotic added to a selective serotonin reuptake inhibitor (SSRI) for obsessive-compulsive disorder (OCD) patients who do not improve on SSRI monotherapy.
• Modafinil added to clozapine in patients who suffer substantial and persistent daytime sedation or somnolence.
• Combining 2 antidepressants for major depressive disorder patients who partially respond to 1 antidepressant.
• Combining a mood stabilizer with an antidepressant for bipolar depression to prevent mood switching.

Ridiculous polypharmacy. The sky is the limit to the variations and degrees of ridiculous polypharmacy, but the theme is the same: an absurd concoction of psychotropic drugs across several classes, often including multiple agents from 1 or several classes. Here are examples I have seen in patient records:

• Two atypicals, an anticholinergic, a mood stabilizer, an antidepressant, and 2 benzodiazepines.
• Three antipsychotics (2 atypicals and 1 typical), 2 antidepressants, 3 sedative/hypnotics, and an anticonvulsant for weight control.
• This one takes the cake: 6 antipsychotics (2 typicals and 4 atypicals), plus an anticholinergic, 3 mood stabilizers, 2 antidepressants, 2 sleeping pills, a hypoglycemic agent, 2 antihypertensives, and a statin.

Hazardous polypharmacy. In this category, serious medical complications, toxic effects, or death may occur because of careless combinations of drugs that may interact to produce dangerous kinetic interactions or exacerbate a pre-existing medical condition. Examples include:

• Combining 1 psychotropic with another that may inhibit its metabolism (eg, prescribing fluvoxamine to a severely psychotic patient who developed OCD while receiving clozapine). There have been several toxic reactions and even death because fluvoxamine inhibits cytochrome 1A2, which metabolizes clozapine, thus increasing clozapine blood level by 400% to 500%.
• Combining 2 injectable drugs for agitation that may cause a serious medical complication. An example would be injecting a benzodiazepine such as lorazepam in a patient receiving olanzapine IM, which can cause severe respiratory depression or death.
• Combining several drugs, each of which may prolong the QTc interval, resulting in syncope or torsade de pointes.

Psychopharmacology can relieve the terrible anguish of psychosis, depression, or anxiety, but it also can carry iatrogenic risks if it is not based on scientific evidence. The practice of psychopharmacology requires the fully integrated skills of medical and psychiatric training to maximize benefit while avoiding harm. It also requires basic arithmetic skills: to consider subtracting drugs, not only adding them!

Discuss this article at www.facebook.com/CurrentPsychiatry

Among DSM axis I diagnoses, bipolar disorder (BD) has the highest rates of comorbid substance use disorders (SUDs).^1-3 Approximately 60% of patients with bipolar I disorder have a lifetime diagnosis of an SUD.¹ Excluding tobacco, alcohol is the substance most often abused by BD patients, followed by cannabis, amphetamines, and cocaine.^1-3

BD patients with comorbid SUD usually exhibit more severe clinical presentations and poorer outcomes than their counterparts without SUDs. Compared with patients with BD alone, those with BD and SUD comorbidity (BD-SUD) experience earlier onset of mood symptoms; higher rates of anxiety disorders, suicide attempts, accidents, hospitalizations, and rapid cycling; more depressive episodes; and lower treatment compliance.^4-9

Several treatment options are available for patients with BD-SUD, including psychotherapy modalities, medications primarily used to treat BD, and medications primarily used to treat SUDs. Evidence-based support for these treatments remains limited, and no treatment of choice has emerged. This article reviews evidence on the longer-term treatment of BD-SUD, including general strategies and specific psychosocial and pharmacologic interventions. Short-term treatment strategies, such as pharmacotherapy for detoxification, are outside the scope of this review.

General strategies

The causes of BD-SUD are complex. Evidence suggests that the presence of affective symptoms is associated with an increased risk for substance misuse. This should be kept in mind when treating a patient with BD-SUD because controlling mood symptoms probably will help control substance abuse. However, evidence also shows that SUDs may be independent of mood episodes. Therefore, treating only mood symptoms in the hope that doing so will control substance abuse may not be enough.

Because the negative impact of SUDs on BD outcome is well documented, inform patients that limiting their use of alcohol and/or drugs is vital to control their mood disorder. Efforts to educate, stimulate, and support patients to moderate or stop their alcohol and/or drug use are likely to result in positive changes.¹⁰ Therefore, treatment for BD-SUD should follow, in part, the same recommendations for treatment of SUDs in patients with no comorbid axis I disorders:

• identify the problem (ie, the existence of a comorbid SUD)
• share your concerns with your patient
• offer appropriate and specific treatments, such as detoxification and/or self-help and counseling programs.¹⁰

Because SUDs usually are chronic and relapsing conditions, periods of drug and/ or alcohol use should be expected and not considered a sign of treatment failure. In addition, integrating treatment for both conditions probably is better than managing each separately. Therefore, targeting BD symptoms with mood-stabilizing medications and substance abuse with nonpharmacologic modalities such as drug counseling likely will bring about the best results.

Compared with BD patients without comorbid SUD, BD-SUD patients have a 7-fold increased risk of antidepressantinduced mania.¹¹ Therefore, antidepressants should be prescribed cautiously for patients with BD-SUD.

Integrated psychosocial therapy

BD-SUD patients may benefit from attending self-help programs such as Alcoholics Anonymous and Narcotics Anonymous, provided their mood is stable enough to allow them to participate. Other forms of psychotherapy for BD-SUD patients include standard group drug counseling and integrated group therapy that simultaneously addresses both conditions.

Integrated group therapy is based on the premise that changing maladaptive mood cognitions and behaviors will facilitate recovery from SUDs, and changing maladaptive substance use cognitions and behaviors will facilitate recovery from mood disorders.¹² In a recent randomized controlled trial, 62 BD-SUD patients were blindly assigned to integrated group therapy or standard group drug counseling and followed for 3 months.¹² Pharmacotherapy was prescribed as usual. Substance use decreased for both groups. However, compared with patients in the drug counseling group, those who participated in integrated group therapy spent fewer days using substances in general and alcohol in particular, fewer days using alcohol to intoxication, and had a shorter time from treatment initiation to the first abstinent month. There were no differences between groups in number of weeks in a mood episode.

Pharmacotherapy options

For a table that summarizes the dosages and indications of the medications used to treat BD-SUD that are described below, visit this article at CurrentPsychiatry.com.

Table

Medications used to treat substance use disorders in bipolar disorder patients*

Lithium. Given its well-documented mood stabilizing effect, lithium would seem to be a reasonable option to treat BD-SUD patients, but scant evidence supports its role as an anti-alcohol or anti-drug medication (Table 1).^13,14 Lithium’s efficacy was evaluated in a study of 25 adolescents suffering from mood disorders (mostly BD) and comorbid SUDs (mostly alcohol and cannabis) randomized to receive lithium or placebo for 6 weeks.¹³ Lithium was well tolerated and improved psychiatric symptoms. At week 3, patients receiving lithium produced fewer positive results on randomly administered urine drug screens than those receiving placebo.

However, lithium seems to have little efficacy in reducing cocaine use in cocaine-dependent patients with bipolar spectrum disorders.¹⁴ In an open-label study, 10 patients with a history of hypomania or cyclothymia received lithium monotherapy for 12 weeks. Although patients experienced improved mood symptoms and decreased cocaine use, the mean decrease was transitory and not statistically significant. Another factor that may limit lithium’s use for BD-SUD patients is that these patients are more likely to comply with valproate treatment than with lithium.¹⁵

Table 1

Lithium for BD patients with substance use disorders

Anticonvulsants. In a double-blind, placebo-controlled study of 59 alcohol-dependent bipolar I disorder patients, lithium plus divalproex sodium was superior to lithium plus placebo in decreasing number of drinking days and number of drinks per day and in increasing periods of abstinence (Table 2).^16-19 Divalproex sodium was well tolerated and liver function improved in the divalproex sodium group compared with the placebo group, which probably was a benefit of decreased alcohol consumption. In addition, there was a strong association between mood symptoms and alcohol use, which suggests that maximizing mood symptom treatment may decrease alcohol use. However, the divalproex sodium and placebo groups did not differ in measures of mood symptoms, which implies that divalproex sodium might exhibit a positive effect on drinking regardless of its mood-stabilizing properties.

Divalproex sodium also has been used to treat BD comorbid with cocaine dependence. In a small open-label study, 15 patients receiving divalproex sodium plus counseling for mood and substance use disorders were followed for 6 weeks.¹⁷ The 7 patients who completed the trial had significantly more cocaine-abstinent days, spent less money on cocaine, and experienced fewer manic and depressive symptoms. However, divalproex sodium’s effect on cocaine use cannot be determined solely from this study because there was no placebo control group.

Despite its widespread use as a mood stabilizer and potential use in alcohol detoxification, carbamazepine scarcely has been studied in BD-SUD patients. A double-blind, placebo-controlled study of 139 cocaine-dependent patients with BD or other affective disorders found that patients taking carbamazepine for 12 weeks experienced modest reductions in positive urine drug screens and increased time to cocaine use.¹⁸ They also reported less cocaine craving than patients taking placebo, and mood symptoms (mostly depressive) improved.

An open-label study used lamotrigine as adjunctive therapy or monotherapy for 62 cocaine-dependent BD patients followed for 36 weeks.¹⁹ There was some decrease in cocaine craving, money spent on cocaine, and rate of depressive and manic symptoms, but no effect on cocaine use. A placebo-controlled trial is necessary to confirm these modest effects.

No studies have evaluated the potential role of topiramate in treating BD-SUD, despite its FDA-approved indication for alcoholism treatment. Topiramate’s well-known safety and tolerability profile in BD patients make it an interesting option for those with co-occurring alcohol dependence.

Table 2

Studies suggest anticonvulsants may reduce alcohol, cocaine use in BD patients

Atypical antipsychotics. In an open-label study, 16 weeks of quetiapine monotherapy effectively decreased alcohol consumption, alcohol craving, and psychotic and affective symptoms in 28 alcoholics with a variety of psychiatric diagnoses, including BD, schizoaffective disorder, and borderline personality disorder (Table 3).^20-24 However, in a double-blind study of augmentation with quetiapine or placebo for 102 alcohol-dependent BD patients, no significant differences in alcohol use were found between groups.²¹

Quetiapine may be effective for treating BD patients with comorbid cocaine dependence. In an open-label study, 12 weeks of quetiapine augmentation in 17 cocaine-dependent BD patients was associated with decreased cocaine craving and improvement in depressive symptoms.²² In another open-label study, 80 BD patients with comorbid cocaine or amphetamine dependence were randomly assigned to receive quetiapine or risperidone as adjunctive therapy or monotherapy for 20 weeks.²³ Both groups showed significantly decreased drug use and drug craving and improved mood. This study suggests that risperidone also may be an option for BD patients with comorbid cocaine or stimulant dependence.

A 20-week, open-label study of 20 BD-SUD patients found that switching patients from their previous antipsychotic to aripiprazole resulted in less cocaine craving, less alcohol craving, and less money spent on alcohol.²⁴

Olanzapine has not been systematically studied in BD-SUD patients. Some case reports suggest that olanzapine may decrease cocaine craving and use in patients with schizoaffective disorder (bipolar type) and alcohol craving and use in BD patients with comorbid alcohol dependence.²⁵

Table 3

Evidence of efficacy for antipsychotics for BD patients with SUDs

SUD medications. Little evidence guides using medications indicated for treating SUDs—such as naltrexone, acamprosate, and disulfiram—as treatment for BD patients (Table 4).^26-29 In an open-label trial of 34 BD patients with alcohol dependence, naltrexone was well tolerated and associated with decreased alcohol craving and use and modest improvement in manic and depressive symptoms.²⁶

In a double-blind, placebo-controlled study, 50 alcohol-dependent BD patients treated with standard mood-stabilizing therapy and cognitive-behavioral therapy were randomized to receive add-on naltrexone, 50 mg/d, or placebo.²⁷ Patients receiving naltrexone showed decreased alcohol consumption, although no measures were statistically significant. Effect sizes of alcohol use decrease and alcohol craving were moderate to large compared with placebo, which suggests that naltrexone may be effective for treating alcoholism in these patients.

Two other studies evaluated naltrexone and disulfiram in patients with BD or other mood disorders.^28,29 Naltrexone was well tolerated, caused no serious adverse side effects, and was significantly more effective than placebo in decreasing drinking rates and increasing the number of abstinent days.^28,29 Disulfiram was as effective as naltrexone, but the combination of both offered no advantage over use of either drug separately.

There are reports of a new-onset manic episode associated with naltrexone use in a patient with opioid dependence, and a manic episode triggered by naltrexone in a patient with BD with comorbid alcohol dependence.^30,31 At both low and high doses, disulfiram is associated with induction of psychotic mania in alcoholic patients without a personal or family history of BD.^32,33

We found no studies that evaluated treating BD patients who abused other substances, such as cannabis or opiates. We recommend that BD patients with these substance use disorders should be referred to treatment modalities that are condition-specific, such as psychotherapy for cannabis use disorders or methadone or naltrexone treatment for opiate dependence. More severe cases of comorbid SUD probably would benefit from a referral to or consultation with a SUD specialist.

Table 4

Naltrexone and disulfiram for BD patients with alcohol dependence

Related Resource

• Tolliver BK. Bipolar disorder and substance abuse: Overcome the challenges of ‘dual diagnosis’ patients. Current Psychiatry. 2010; 9(8): 32-38.

Drug Brand Names

• Acamprosate • Campral
• Aripiprazole • Abilify
• Carbamazepine • Carbatrol, Equetro, others
• Disulfiram • Antabuse
• Divalproex sodium • Depakote,
• Depakote ER Lamotrigine • Lamictal
• Lithium • Eskalith, Lithobid
• Methadone • Dolophine
• Naltrexone • ReVia, Vivitrol
• Quetiapine • Seroquel
• Risperidone • Risperdal
• Topiramate • Topamax
• Valproate • Depacon

Disclosures

Dr. Nery held a temporary work contract as a clinical research physician with Eli Lilly and Company Brazil from May 2009 to November 2009.

Dr. Soares was partly supported by National Institute of Health grants MH 68766, MH 69774, and RR 20571. He receives grant/research support from Bristol-Myers Squibb, Cephalon, GlaxoSmithKline, and Sunovion.

Discuss this article at www.facebook.com/CurrentPsychiatry

Among DSM axis I diagnoses, bipolar disorder (BD) has the highest rates of comorbid substance use disorders (SUDs).^1-3 Approximately 60% of patients with bipolar I disorder have a lifetime diagnosis of an SUD.¹ Excluding tobacco, alcohol is the substance most often abused by BD patients, followed by cannabis, amphetamines, and cocaine.^1-3

BD patients with comorbid SUD usually exhibit more severe clinical presentations and poorer outcomes than their counterparts without SUDs. Compared with patients with BD alone, those with BD and SUD comorbidity (BD-SUD) experience earlier onset of mood symptoms; higher rates of anxiety disorders, suicide attempts, accidents, hospitalizations, and rapid cycling; more depressive episodes; and lower treatment compliance.^4-9

Several treatment options are available for patients with BD-SUD, including psychotherapy modalities, medications primarily used to treat BD, and medications primarily used to treat SUDs. Evidence-based support for these treatments remains limited, and no treatment of choice has emerged. This article reviews evidence on the longer-term treatment of BD-SUD, including general strategies and specific psychosocial and pharmacologic interventions. Short-term treatment strategies, such as pharmacotherapy for detoxification, are outside the scope of this review.

General strategies

The causes of BD-SUD are complex. Evidence suggests that the presence of affective symptoms is associated with an increased risk for substance misuse. This should be kept in mind when treating a patient with BD-SUD because controlling mood symptoms probably will help control substance abuse. However, evidence also shows that SUDs may be independent of mood episodes. Therefore, treating only mood symptoms in the hope that doing so will control substance abuse may not be enough.

Because the negative impact of SUDs on BD outcome is well documented, inform patients that limiting their use of alcohol and/or drugs is vital to control their mood disorder. Efforts to educate, stimulate, and support patients to moderate or stop their alcohol and/or drug use are likely to result in positive changes.¹⁰ Therefore, treatment for BD-SUD should follow, in part, the same recommendations for treatment of SUDs in patients with no comorbid axis I disorders:

• identify the problem (ie, the existence of a comorbid SUD)
• share your concerns with your patient
• offer appropriate and specific treatments, such as detoxification and/or self-help and counseling programs.¹⁰

Because SUDs usually are chronic and relapsing conditions, periods of drug and/ or alcohol use should be expected and not considered a sign of treatment failure. In addition, integrating treatment for both conditions probably is better than managing each separately. Therefore, targeting BD symptoms with mood-stabilizing medications and substance abuse with nonpharmacologic modalities such as drug counseling likely will bring about the best results.

Compared with BD patients without comorbid SUD, BD-SUD patients have a 7-fold increased risk of antidepressantinduced mania.¹¹ Therefore, antidepressants should be prescribed cautiously for patients with BD-SUD.

Integrated psychosocial therapy

BD-SUD patients may benefit from attending self-help programs such as Alcoholics Anonymous and Narcotics Anonymous, provided their mood is stable enough to allow them to participate. Other forms of psychotherapy for BD-SUD patients include standard group drug counseling and integrated group therapy that simultaneously addresses both conditions.

Integrated group therapy is based on the premise that changing maladaptive mood cognitions and behaviors will facilitate recovery from SUDs, and changing maladaptive substance use cognitions and behaviors will facilitate recovery from mood disorders.¹² In a recent randomized controlled trial, 62 BD-SUD patients were blindly assigned to integrated group therapy or standard group drug counseling and followed for 3 months.¹² Pharmacotherapy was prescribed as usual. Substance use decreased for both groups. However, compared with patients in the drug counseling group, those who participated in integrated group therapy spent fewer days using substances in general and alcohol in particular, fewer days using alcohol to intoxication, and had a shorter time from treatment initiation to the first abstinent month. There were no differences between groups in number of weeks in a mood episode.

Pharmacotherapy options

For a table that summarizes the dosages and indications of the medications used to treat BD-SUD that are described below, visit this article at CurrentPsychiatry.com.

Table

Medications used to treat substance use disorders in bipolar disorder patients*

Lithium. Given its well-documented mood stabilizing effect, lithium would seem to be a reasonable option to treat BD-SUD patients, but scant evidence supports its role as an anti-alcohol or anti-drug medication (Table 1).^13,14 Lithium’s efficacy was evaluated in a study of 25 adolescents suffering from mood disorders (mostly BD) and comorbid SUDs (mostly alcohol and cannabis) randomized to receive lithium or placebo for 6 weeks.¹³ Lithium was well tolerated and improved psychiatric symptoms. At week 3, patients receiving lithium produced fewer positive results on randomly administered urine drug screens than those receiving placebo.

However, lithium seems to have little efficacy in reducing cocaine use in cocaine-dependent patients with bipolar spectrum disorders.¹⁴ In an open-label study, 10 patients with a history of hypomania or cyclothymia received lithium monotherapy for 12 weeks. Although patients experienced improved mood symptoms and decreased cocaine use, the mean decrease was transitory and not statistically significant. Another factor that may limit lithium’s use for BD-SUD patients is that these patients are more likely to comply with valproate treatment than with lithium.¹⁵

Table 1

Lithium for BD patients with substance use disorders

Anticonvulsants. In a double-blind, placebo-controlled study of 59 alcohol-dependent bipolar I disorder patients, lithium plus divalproex sodium was superior to lithium plus placebo in decreasing number of drinking days and number of drinks per day and in increasing periods of abstinence (Table 2).^16-19 Divalproex sodium was well tolerated and liver function improved in the divalproex sodium group compared with the placebo group, which probably was a benefit of decreased alcohol consumption. In addition, there was a strong association between mood symptoms and alcohol use, which suggests that maximizing mood symptom treatment may decrease alcohol use. However, the divalproex sodium and placebo groups did not differ in measures of mood symptoms, which implies that divalproex sodium might exhibit a positive effect on drinking regardless of its mood-stabilizing properties.

Divalproex sodium also has been used to treat BD comorbid with cocaine dependence. In a small open-label study, 15 patients receiving divalproex sodium plus counseling for mood and substance use disorders were followed for 6 weeks.¹⁷ The 7 patients who completed the trial had significantly more cocaine-abstinent days, spent less money on cocaine, and experienced fewer manic and depressive symptoms. However, divalproex sodium’s effect on cocaine use cannot be determined solely from this study because there was no placebo control group.

Despite its widespread use as a mood stabilizer and potential use in alcohol detoxification, carbamazepine scarcely has been studied in BD-SUD patients. A double-blind, placebo-controlled study of 139 cocaine-dependent patients with BD or other affective disorders found that patients taking carbamazepine for 12 weeks experienced modest reductions in positive urine drug screens and increased time to cocaine use.¹⁸ They also reported less cocaine craving than patients taking placebo, and mood symptoms (mostly depressive) improved.

An open-label study used lamotrigine as adjunctive therapy or monotherapy for 62 cocaine-dependent BD patients followed for 36 weeks.¹⁹ There was some decrease in cocaine craving, money spent on cocaine, and rate of depressive and manic symptoms, but no effect on cocaine use. A placebo-controlled trial is necessary to confirm these modest effects.

No studies have evaluated the potential role of topiramate in treating BD-SUD, despite its FDA-approved indication for alcoholism treatment. Topiramate’s well-known safety and tolerability profile in BD patients make it an interesting option for those with co-occurring alcohol dependence.

Table 2

Studies suggest anticonvulsants may reduce alcohol, cocaine use in BD patients

Atypical antipsychotics. In an open-label study, 16 weeks of quetiapine monotherapy effectively decreased alcohol consumption, alcohol craving, and psychotic and affective symptoms in 28 alcoholics with a variety of psychiatric diagnoses, including BD, schizoaffective disorder, and borderline personality disorder (Table 3).^20-24 However, in a double-blind study of augmentation with quetiapine or placebo for 102 alcohol-dependent BD patients, no significant differences in alcohol use were found between groups.²¹

Quetiapine may be effective for treating BD patients with comorbid cocaine dependence. In an open-label study, 12 weeks of quetiapine augmentation in 17 cocaine-dependent BD patients was associated with decreased cocaine craving and improvement in depressive symptoms.²² In another open-label study, 80 BD patients with comorbid cocaine or amphetamine dependence were randomly assigned to receive quetiapine or risperidone as adjunctive therapy or monotherapy for 20 weeks.²³ Both groups showed significantly decreased drug use and drug craving and improved mood. This study suggests that risperidone also may be an option for BD patients with comorbid cocaine or stimulant dependence.

A 20-week, open-label study of 20 BD-SUD patients found that switching patients from their previous antipsychotic to aripiprazole resulted in less cocaine craving, less alcohol craving, and less money spent on alcohol.²⁴

Olanzapine has not been systematically studied in BD-SUD patients. Some case reports suggest that olanzapine may decrease cocaine craving and use in patients with schizoaffective disorder (bipolar type) and alcohol craving and use in BD patients with comorbid alcohol dependence.²⁵

Table 3

Evidence of efficacy for antipsychotics for BD patients with SUDs

SUD medications. Little evidence guides using medications indicated for treating SUDs—such as naltrexone, acamprosate, and disulfiram—as treatment for BD patients (Table 4).^26-29 In an open-label trial of 34 BD patients with alcohol dependence, naltrexone was well tolerated and associated with decreased alcohol craving and use and modest improvement in manic and depressive symptoms.²⁶

In a double-blind, placebo-controlled study, 50 alcohol-dependent BD patients treated with standard mood-stabilizing therapy and cognitive-behavioral therapy were randomized to receive add-on naltrexone, 50 mg/d, or placebo.²⁷ Patients receiving naltrexone showed decreased alcohol consumption, although no measures were statistically significant. Effect sizes of alcohol use decrease and alcohol craving were moderate to large compared with placebo, which suggests that naltrexone may be effective for treating alcoholism in these patients.

Two other studies evaluated naltrexone and disulfiram in patients with BD or other mood disorders.^28,29 Naltrexone was well tolerated, caused no serious adverse side effects, and was significantly more effective than placebo in decreasing drinking rates and increasing the number of abstinent days.^28,29 Disulfiram was as effective as naltrexone, but the combination of both offered no advantage over use of either drug separately.

There are reports of a new-onset manic episode associated with naltrexone use in a patient with opioid dependence, and a manic episode triggered by naltrexone in a patient with BD with comorbid alcohol dependence.^30,31 At both low and high doses, disulfiram is associated with induction of psychotic mania in alcoholic patients without a personal or family history of BD.^32,33

We found no studies that evaluated treating BD patients who abused other substances, such as cannabis or opiates. We recommend that BD patients with these substance use disorders should be referred to treatment modalities that are condition-specific, such as psychotherapy for cannabis use disorders or methadone or naltrexone treatment for opiate dependence. More severe cases of comorbid SUD probably would benefit from a referral to or consultation with a SUD specialist.

Table 4

Naltrexone and disulfiram for BD patients with alcohol dependence

Related Resource

• Tolliver BK. Bipolar disorder and substance abuse: Overcome the challenges of ‘dual diagnosis’ patients. Current Psychiatry. 2010; 9(8): 32-38.

Drug Brand Names

• Acamprosate • Campral
• Aripiprazole • Abilify
• Carbamazepine • Carbatrol, Equetro, others
• Disulfiram • Antabuse
• Divalproex sodium • Depakote,
• Depakote ER Lamotrigine • Lamictal
• Lithium • Eskalith, Lithobid
• Methadone • Dolophine
• Naltrexone • ReVia, Vivitrol
• Quetiapine • Seroquel
• Risperidone • Risperdal
• Topiramate • Topamax
• Valproate • Depacon

Disclosures

Dr. Nery held a temporary work contract as a clinical research physician with Eli Lilly and Company Brazil from May 2009 to November 2009.

Dr. Soares was partly supported by National Institute of Health grants MH 68766, MH 69774, and RR 20571. He receives grant/research support from Bristol-Myers Squibb, Cephalon, GlaxoSmithKline, and Sunovion.

Discuss this article at www.facebook.com/CurrentPsychiatry

Among DSM axis I diagnoses, bipolar disorder (BD) has the highest rates of comorbid substance use disorders (SUDs).^1-3 Approximately 60% of patients with bipolar I disorder have a lifetime diagnosis of an SUD.¹ Excluding tobacco, alcohol is the substance most often abused by BD patients, followed by cannabis, amphetamines, and cocaine.^1-3

BD patients with comorbid SUD usually exhibit more severe clinical presentations and poorer outcomes than their counterparts without SUDs. Compared with patients with BD alone, those with BD and SUD comorbidity (BD-SUD) experience earlier onset of mood symptoms; higher rates of anxiety disorders, suicide attempts, accidents, hospitalizations, and rapid cycling; more depressive episodes; and lower treatment compliance.^4-9

Several treatment options are available for patients with BD-SUD, including psychotherapy modalities, medications primarily used to treat BD, and medications primarily used to treat SUDs. Evidence-based support for these treatments remains limited, and no treatment of choice has emerged. This article reviews evidence on the longer-term treatment of BD-SUD, including general strategies and specific psychosocial and pharmacologic interventions. Short-term treatment strategies, such as pharmacotherapy for detoxification, are outside the scope of this review.

General strategies

The causes of BD-SUD are complex. Evidence suggests that the presence of affective symptoms is associated with an increased risk for substance misuse. This should be kept in mind when treating a patient with BD-SUD because controlling mood symptoms probably will help control substance abuse. However, evidence also shows that SUDs may be independent of mood episodes. Therefore, treating only mood symptoms in the hope that doing so will control substance abuse may not be enough.

Because the negative impact of SUDs on BD outcome is well documented, inform patients that limiting their use of alcohol and/or drugs is vital to control their mood disorder. Efforts to educate, stimulate, and support patients to moderate or stop their alcohol and/or drug use are likely to result in positive changes.¹⁰ Therefore, treatment for BD-SUD should follow, in part, the same recommendations for treatment of SUDs in patients with no comorbid axis I disorders:

• identify the problem (ie, the existence of a comorbid SUD)
• share your concerns with your patient
• offer appropriate and specific treatments, such as detoxification and/or self-help and counseling programs.¹⁰

Because SUDs usually are chronic and relapsing conditions, periods of drug and/ or alcohol use should be expected and not considered a sign of treatment failure. In addition, integrating treatment for both conditions probably is better than managing each separately. Therefore, targeting BD symptoms with mood-stabilizing medications and substance abuse with nonpharmacologic modalities such as drug counseling likely will bring about the best results.

Compared with BD patients without comorbid SUD, BD-SUD patients have a 7-fold increased risk of antidepressantinduced mania.¹¹ Therefore, antidepressants should be prescribed cautiously for patients with BD-SUD.

Integrated psychosocial therapy

BD-SUD patients may benefit from attending self-help programs such as Alcoholics Anonymous and Narcotics Anonymous, provided their mood is stable enough to allow them to participate. Other forms of psychotherapy for BD-SUD patients include standard group drug counseling and integrated group therapy that simultaneously addresses both conditions.

Integrated group therapy is based on the premise that changing maladaptive mood cognitions and behaviors will facilitate recovery from SUDs, and changing maladaptive substance use cognitions and behaviors will facilitate recovery from mood disorders.¹² In a recent randomized controlled trial, 62 BD-SUD patients were blindly assigned to integrated group therapy or standard group drug counseling and followed for 3 months.¹² Pharmacotherapy was prescribed as usual. Substance use decreased for both groups. However, compared with patients in the drug counseling group, those who participated in integrated group therapy spent fewer days using substances in general and alcohol in particular, fewer days using alcohol to intoxication, and had a shorter time from treatment initiation to the first abstinent month. There were no differences between groups in number of weeks in a mood episode.

Pharmacotherapy options

For a table that summarizes the dosages and indications of the medications used to treat BD-SUD that are described below, visit this article at CurrentPsychiatry.com.

Table

Medications used to treat substance use disorders in bipolar disorder patients*

Lithium. Given its well-documented mood stabilizing effect, lithium would seem to be a reasonable option to treat BD-SUD patients, but scant evidence supports its role as an anti-alcohol or anti-drug medication (Table 1).^13,14 Lithium’s efficacy was evaluated in a study of 25 adolescents suffering from mood disorders (mostly BD) and comorbid SUDs (mostly alcohol and cannabis) randomized to receive lithium or placebo for 6 weeks.¹³ Lithium was well tolerated and improved psychiatric symptoms. At week 3, patients receiving lithium produced fewer positive results on randomly administered urine drug screens than those receiving placebo.

However, lithium seems to have little efficacy in reducing cocaine use in cocaine-dependent patients with bipolar spectrum disorders.¹⁴ In an open-label study, 10 patients with a history of hypomania or cyclothymia received lithium monotherapy for 12 weeks. Although patients experienced improved mood symptoms and decreased cocaine use, the mean decrease was transitory and not statistically significant. Another factor that may limit lithium’s use for BD-SUD patients is that these patients are more likely to comply with valproate treatment than with lithium.¹⁵

Table 1

Lithium for BD patients with substance use disorders

Anticonvulsants. In a double-blind, placebo-controlled study of 59 alcohol-dependent bipolar I disorder patients, lithium plus divalproex sodium was superior to lithium plus placebo in decreasing number of drinking days and number of drinks per day and in increasing periods of abstinence (Table 2).^16-19 Divalproex sodium was well tolerated and liver function improved in the divalproex sodium group compared with the placebo group, which probably was a benefit of decreased alcohol consumption. In addition, there was a strong association between mood symptoms and alcohol use, which suggests that maximizing mood symptom treatment may decrease alcohol use. However, the divalproex sodium and placebo groups did not differ in measures of mood symptoms, which implies that divalproex sodium might exhibit a positive effect on drinking regardless of its mood-stabilizing properties.

Divalproex sodium also has been used to treat BD comorbid with cocaine dependence. In a small open-label study, 15 patients receiving divalproex sodium plus counseling for mood and substance use disorders were followed for 6 weeks.¹⁷ The 7 patients who completed the trial had significantly more cocaine-abstinent days, spent less money on cocaine, and experienced fewer manic and depressive symptoms. However, divalproex sodium’s effect on cocaine use cannot be determined solely from this study because there was no placebo control group.

Despite its widespread use as a mood stabilizer and potential use in alcohol detoxification, carbamazepine scarcely has been studied in BD-SUD patients. A double-blind, placebo-controlled study of 139 cocaine-dependent patients with BD or other affective disorders found that patients taking carbamazepine for 12 weeks experienced modest reductions in positive urine drug screens and increased time to cocaine use.¹⁸ They also reported less cocaine craving than patients taking placebo, and mood symptoms (mostly depressive) improved.

An open-label study used lamotrigine as adjunctive therapy or monotherapy for 62 cocaine-dependent BD patients followed for 36 weeks.¹⁹ There was some decrease in cocaine craving, money spent on cocaine, and rate of depressive and manic symptoms, but no effect on cocaine use. A placebo-controlled trial is necessary to confirm these modest effects.

No studies have evaluated the potential role of topiramate in treating BD-SUD, despite its FDA-approved indication for alcoholism treatment. Topiramate’s well-known safety and tolerability profile in BD patients make it an interesting option for those with co-occurring alcohol dependence.

Table 2

Studies suggest anticonvulsants may reduce alcohol, cocaine use in BD patients

Atypical antipsychotics. In an open-label study, 16 weeks of quetiapine monotherapy effectively decreased alcohol consumption, alcohol craving, and psychotic and affective symptoms in 28 alcoholics with a variety of psychiatric diagnoses, including BD, schizoaffective disorder, and borderline personality disorder (Table 3).^20-24 However, in a double-blind study of augmentation with quetiapine or placebo for 102 alcohol-dependent BD patients, no significant differences in alcohol use were found between groups.²¹

Quetiapine may be effective for treating BD patients with comorbid cocaine dependence. In an open-label study, 12 weeks of quetiapine augmentation in 17 cocaine-dependent BD patients was associated with decreased cocaine craving and improvement in depressive symptoms.²² In another open-label study, 80 BD patients with comorbid cocaine or amphetamine dependence were randomly assigned to receive quetiapine or risperidone as adjunctive therapy or monotherapy for 20 weeks.²³ Both groups showed significantly decreased drug use and drug craving and improved mood. This study suggests that risperidone also may be an option for BD patients with comorbid cocaine or stimulant dependence.

A 20-week, open-label study of 20 BD-SUD patients found that switching patients from their previous antipsychotic to aripiprazole resulted in less cocaine craving, less alcohol craving, and less money spent on alcohol.²⁴

Olanzapine has not been systematically studied in BD-SUD patients. Some case reports suggest that olanzapine may decrease cocaine craving and use in patients with schizoaffective disorder (bipolar type) and alcohol craving and use in BD patients with comorbid alcohol dependence.²⁵

Table 3

Evidence of efficacy for antipsychotics for BD patients with SUDs

SUD medications. Little evidence guides using medications indicated for treating SUDs—such as naltrexone, acamprosate, and disulfiram—as treatment for BD patients (Table 4).^26-29 In an open-label trial of 34 BD patients with alcohol dependence, naltrexone was well tolerated and associated with decreased alcohol craving and use and modest improvement in manic and depressive symptoms.²⁶

In a double-blind, placebo-controlled study, 50 alcohol-dependent BD patients treated with standard mood-stabilizing therapy and cognitive-behavioral therapy were randomized to receive add-on naltrexone, 50 mg/d, or placebo.²⁷ Patients receiving naltrexone showed decreased alcohol consumption, although no measures were statistically significant. Effect sizes of alcohol use decrease and alcohol craving were moderate to large compared with placebo, which suggests that naltrexone may be effective for treating alcoholism in these patients.

Two other studies evaluated naltrexone and disulfiram in patients with BD or other mood disorders.^28,29 Naltrexone was well tolerated, caused no serious adverse side effects, and was significantly more effective than placebo in decreasing drinking rates and increasing the number of abstinent days.^28,29 Disulfiram was as effective as naltrexone, but the combination of both offered no advantage over use of either drug separately.

There are reports of a new-onset manic episode associated with naltrexone use in a patient with opioid dependence, and a manic episode triggered by naltrexone in a patient with BD with comorbid alcohol dependence.^30,31 At both low and high doses, disulfiram is associated with induction of psychotic mania in alcoholic patients without a personal or family history of BD.^32,33

We found no studies that evaluated treating BD patients who abused other substances, such as cannabis or opiates. We recommend that BD patients with these substance use disorders should be referred to treatment modalities that are condition-specific, such as psychotherapy for cannabis use disorders or methadone or naltrexone treatment for opiate dependence. More severe cases of comorbid SUD probably would benefit from a referral to or consultation with a SUD specialist.

Table 4

Naltrexone and disulfiram for BD patients with alcohol dependence

Related Resource

• Tolliver BK. Bipolar disorder and substance abuse: Overcome the challenges of ‘dual diagnosis’ patients. Current Psychiatry. 2010; 9(8): 32-38.

Drug Brand Names

• Acamprosate • Campral
• Aripiprazole • Abilify
• Carbamazepine • Carbatrol, Equetro, others
• Disulfiram • Antabuse
• Divalproex sodium • Depakote,
• Depakote ER Lamotrigine • Lamictal
• Lithium • Eskalith, Lithobid
• Methadone • Dolophine
• Naltrexone • ReVia, Vivitrol
• Quetiapine • Seroquel
• Risperidone • Risperdal
• Topiramate • Topamax
• Valproate • Depacon

Disclosures

Dr. Nery held a temporary work contract as a clinical research physician with Eli Lilly and Company Brazil from May 2009 to November 2009.

Dr. Soares was partly supported by National Institute of Health grants MH 68766, MH 69774, and RR 20571. He receives grant/research support from Bristol-Myers Squibb, Cephalon, GlaxoSmithKline, and Sunovion.

Worldwide, the prevalence of bipolar disorder type I is estimated to be 0.6%, that of type II is 0.4%, and that of subthreshold bipolar disorder is 1.4%, yielding a total bipolar disorder spectrum prevalence of 2.4%, according to a new report in the March issue of the Archives of General Psychiatry.

These estimates were derived from what the investigators described as the first international data ever collected using Diagnostic and Statistical Manual–IV definitions for the full spectrum of bipolar disorders and standardized measures to survey nationally representative samples in 11 low-income, middle-income, and high-income countries.

Even though prevalence varied from one country to the next, disease severity, impact on daily life, and patterns of comorbidity remained strikingly similar, said Kathleen R. Merikangas, Ph.D., of the Intramural Research Program of the National Institute of Mental Health (NIMH), and her associates in the World Mental Health Survey Initiative.

This World Health Organization (WHO) project "aims to obtain accurate cross-national information on the prevalence, correlates, and service patterns of mental disorders." The investigators conducted in-person, population-based surveys of 61,392 adults in Brazil, Colombia, Mexico, the United States, Bulgaria, Romania, China, India, Japan, Lebanon, and New Zealand.

In addition to the overall prevalences, they found that the 1-year prevalence of bipolar disorder type I was 0.4%, that of bipolar disorder type II was 0.3%, and that of subthreshold bipolar disorder was 0.8%, for a total bipolar spectrum disorder annual prevalence of 1.5%.

In general, high-income countries had the highest prevalences of bipolar disease and low-income countries had the lowest. The United States had the highest prevalence of overall (4.4%) and annual (2.8%) disease, while India had the lowest (0.1% for both). Two exceptions to this rule were Japan, a high-income country with very low overall (0.7%) and annual (0.2%) prevalences, and Colombia, a low-income country with a high overall prevalence (2.6%).

The mean ages at onset were 18 years for bipolar disorder type I, 20 years for bipolar disorder type II, and 22 years for subthreshold bipolar disorder.

Patterns of comorbidity were remarkably consistent among the different countries. Three-fourths of patients with bipolar spectrum disorders also met criteria for other psychiatric disorders, and the majority had three or more of them. "Anxiety disorders, particularly panic attacks, were the most common comorbid conditions (63%), followed by behavior disorders (45%) and substance use disorders (37%)," Dr. Merikangas and her colleagues said (Arch. Gen. Psychiatry 2011;68:241-51).

The association between bipolar disorders and substance use disorders across the globe was particularly striking in light of the large differences between countries in rates of substance use and abuse. "This suggests that [bipolar disorder] can be considered a risk factor for the development of substance use disorders ... [and supports] the need for careful probing of a history of bipolarity among those with substance use disorders," they said.

The proportion of patients with suicidal behaviors rose with increasing severity of bipolar disease. Approximately 10% of those with subthreshold bipolar disorder, 20% of those with bipolar disorder type II, and 25% of those with bipolar disorder type I said that they had attempted suicide over the last 12 months.

This "striking" finding of suicidality, together with the early age at onset and the strong association with other mental health disorders, provides "further documentation of the individual and societal disability associated with this disorder," they noted.

A substantially higher proportion of patients in high-income countries, compared with middle- or low-income countries, reported having used mental health services. Lifetime use of such services was 50% in high-income countries, 34% in middle-income countries, and 25% in low-income countries.

"Because [bipolar disorder] has an average age at onset at one of the most critical periods of educational, occupational, and social development, its consequences often lead to lifelong disability," the investigators said, and this lack of mental health treatment, especially in low-income countries, is therefore "alarming."

"These data also provide the first aggregate international evidence, to our knowledge, that supports the validity of the spectrum concept of bipolarity," they added. The proportion of mood episodes rated as clinically severe, as well as the proportion in which patients reported severe impairment of work, home management, social life, and close relationships, were directly associated with increasingly severe forms of bipolar disease.

This study was supported by the NIMH’s Intramural Research Program, the U.S. Public Health Service, numerous organizations in the participating countries, and numerous drug companies. Dr. Merikangas reported her affiliation with the Intramural Research Program, and one of her associates, Ronald Kessler, Ph.D., reported ties to numerous pharmaceutical companies.

Worldwide, the prevalence of bipolar disorder type I is estimated to be 0.6%, that of type II is 0.4%, and that of subthreshold bipolar disorder is 1.4%, yielding a total bipolar disorder spectrum prevalence of 2.4%, according to a new report in the March issue of the Archives of General Psychiatry.

These estimates were derived from what the investigators described as the first international data ever collected using Diagnostic and Statistical Manual–IV definitions for the full spectrum of bipolar disorders and standardized measures to survey nationally representative samples in 11 low-income, middle-income, and high-income countries.

Even though prevalence varied from one country to the next, disease severity, impact on daily life, and patterns of comorbidity remained strikingly similar, said Kathleen R. Merikangas, Ph.D., of the Intramural Research Program of the National Institute of Mental Health (NIMH), and her associates in the World Mental Health Survey Initiative.

This World Health Organization (WHO) project "aims to obtain accurate cross-national information on the prevalence, correlates, and service patterns of mental disorders." The investigators conducted in-person, population-based surveys of 61,392 adults in Brazil, Colombia, Mexico, the United States, Bulgaria, Romania, China, India, Japan, Lebanon, and New Zealand.

In addition to the overall prevalences, they found that the 1-year prevalence of bipolar disorder type I was 0.4%, that of bipolar disorder type II was 0.3%, and that of subthreshold bipolar disorder was 0.8%, for a total bipolar spectrum disorder annual prevalence of 1.5%.

In general, high-income countries had the highest prevalences of bipolar disease and low-income countries had the lowest. The United States had the highest prevalence of overall (4.4%) and annual (2.8%) disease, while India had the lowest (0.1% for both). Two exceptions to this rule were Japan, a high-income country with very low overall (0.7%) and annual (0.2%) prevalences, and Colombia, a low-income country with a high overall prevalence (2.6%).

The mean ages at onset were 18 years for bipolar disorder type I, 20 years for bipolar disorder type II, and 22 years for subthreshold bipolar disorder.

Patterns of comorbidity were remarkably consistent among the different countries. Three-fourths of patients with bipolar spectrum disorders also met criteria for other psychiatric disorders, and the majority had three or more of them. "Anxiety disorders, particularly panic attacks, were the most common comorbid conditions (63%), followed by behavior disorders (45%) and substance use disorders (37%)," Dr. Merikangas and her colleagues said (Arch. Gen. Psychiatry 2011;68:241-51).

The association between bipolar disorders and substance use disorders across the globe was particularly striking in light of the large differences between countries in rates of substance use and abuse. "This suggests that [bipolar disorder] can be considered a risk factor for the development of substance use disorders ... [and supports] the need for careful probing of a history of bipolarity among those with substance use disorders," they said.

The proportion of patients with suicidal behaviors rose with increasing severity of bipolar disease. Approximately 10% of those with subthreshold bipolar disorder, 20% of those with bipolar disorder type II, and 25% of those with bipolar disorder type I said that they had attempted suicide over the last 12 months.

This "striking" finding of suicidality, together with the early age at onset and the strong association with other mental health disorders, provides "further documentation of the individual and societal disability associated with this disorder," they noted.

A substantially higher proportion of patients in high-income countries, compared with middle- or low-income countries, reported having used mental health services. Lifetime use of such services was 50% in high-income countries, 34% in middle-income countries, and 25% in low-income countries.

"Because [bipolar disorder] has an average age at onset at one of the most critical periods of educational, occupational, and social development, its consequences often lead to lifelong disability," the investigators said, and this lack of mental health treatment, especially in low-income countries, is therefore "alarming."

"These data also provide the first aggregate international evidence, to our knowledge, that supports the validity of the spectrum concept of bipolarity," they added. The proportion of mood episodes rated as clinically severe, as well as the proportion in which patients reported severe impairment of work, home management, social life, and close relationships, were directly associated with increasingly severe forms of bipolar disease.

This study was supported by the NIMH’s Intramural Research Program, the U.S. Public Health Service, numerous organizations in the participating countries, and numerous drug companies. Dr. Merikangas reported her affiliation with the Intramural Research Program, and one of her associates, Ronald Kessler, Ph.D., reported ties to numerous pharmaceutical companies.

Worldwide, the prevalence of bipolar disorder type I is estimated to be 0.6%, that of type II is 0.4%, and that of subthreshold bipolar disorder is 1.4%, yielding a total bipolar disorder spectrum prevalence of 2.4%, according to a new report in the March issue of the Archives of General Psychiatry.

These estimates were derived from what the investigators described as the first international data ever collected using Diagnostic and Statistical Manual–IV definitions for the full spectrum of bipolar disorders and standardized measures to survey nationally representative samples in 11 low-income, middle-income, and high-income countries.

Even though prevalence varied from one country to the next, disease severity, impact on daily life, and patterns of comorbidity remained strikingly similar, said Kathleen R. Merikangas, Ph.D., of the Intramural Research Program of the National Institute of Mental Health (NIMH), and her associates in the World Mental Health Survey Initiative.

This World Health Organization (WHO) project "aims to obtain accurate cross-national information on the prevalence, correlates, and service patterns of mental disorders." The investigators conducted in-person, population-based surveys of 61,392 adults in Brazil, Colombia, Mexico, the United States, Bulgaria, Romania, China, India, Japan, Lebanon, and New Zealand.

In addition to the overall prevalences, they found that the 1-year prevalence of bipolar disorder type I was 0.4%, that of bipolar disorder type II was 0.3%, and that of subthreshold bipolar disorder was 0.8%, for a total bipolar spectrum disorder annual prevalence of 1.5%.

In general, high-income countries had the highest prevalences of bipolar disease and low-income countries had the lowest. The United States had the highest prevalence of overall (4.4%) and annual (2.8%) disease, while India had the lowest (0.1% for both). Two exceptions to this rule were Japan, a high-income country with very low overall (0.7%) and annual (0.2%) prevalences, and Colombia, a low-income country with a high overall prevalence (2.6%).

The mean ages at onset were 18 years for bipolar disorder type I, 20 years for bipolar disorder type II, and 22 years for subthreshold bipolar disorder.

Patterns of comorbidity were remarkably consistent among the different countries. Three-fourths of patients with bipolar spectrum disorders also met criteria for other psychiatric disorders, and the majority had three or more of them. "Anxiety disorders, particularly panic attacks, were the most common comorbid conditions (63%), followed by behavior disorders (45%) and substance use disorders (37%)," Dr. Merikangas and her colleagues said (Arch. Gen. Psychiatry 2011;68:241-51).

The association between bipolar disorders and substance use disorders across the globe was particularly striking in light of the large differences between countries in rates of substance use and abuse. "This suggests that [bipolar disorder] can be considered a risk factor for the development of substance use disorders ... [and supports] the need for careful probing of a history of bipolarity among those with substance use disorders," they said.

The proportion of patients with suicidal behaviors rose with increasing severity of bipolar disease. Approximately 10% of those with subthreshold bipolar disorder, 20% of those with bipolar disorder type II, and 25% of those with bipolar disorder type I said that they had attempted suicide over the last 12 months.

This "striking" finding of suicidality, together with the early age at onset and the strong association with other mental health disorders, provides "further documentation of the individual and societal disability associated with this disorder," they noted.

A substantially higher proportion of patients in high-income countries, compared with middle- or low-income countries, reported having used mental health services. Lifetime use of such services was 50% in high-income countries, 34% in middle-income countries, and 25% in low-income countries.

"Because [bipolar disorder] has an average age at onset at one of the most critical periods of educational, occupational, and social development, its consequences often lead to lifelong disability," the investigators said, and this lack of mental health treatment, especially in low-income countries, is therefore "alarming."

"These data also provide the first aggregate international evidence, to our knowledge, that supports the validity of the spectrum concept of bipolarity," they added. The proportion of mood episodes rated as clinically severe, as well as the proportion in which patients reported severe impairment of work, home management, social life, and close relationships, were directly associated with increasingly severe forms of bipolar disease.

This study was supported by the NIMH’s Intramural Research Program, the U.S. Public Health Service, numerous organizations in the participating countries, and numerous drug companies. Dr. Merikangas reported her affiliation with the Intramural Research Program, and one of her associates, Ronald Kessler, Ph.D., reported ties to numerous pharmaceutical companies.

The atypical antipsychotic aripiprazole has been approved by the Food and Drug Administration for the maintenance treatment of people with bipolar I disorder as an adjunct to lithium or valproate, the manufacturers announced Feb. 17.

Aripiprazole, marketed as Abilify by the Bristol-Myers Squibb Co. and Otsuka Pharmaceutical Co., was approved in May 2008 as an adjunct to lithium or valproate for the acute treatment of manic and mixed episodes associated with bipolar I disorder.

The approval of the expanded adjunctive indication was based on a 52-week randomized, double-blind, placebo-controlled maintenance study of adults who met the DSM-IV criteria for bipolar I disorder and who had had a recent manic or mixed episode. The subjects also had a history of one or more manic or mixed episodes that had been severe enough to require hospitalization and/or treatment with a mood stabilizer or antipsychotic. All patients started treatment with lithium or valproate; after 2 weeks, those who had an inadequate response to one of the mood stabilizers alone began treatment with aripiprazole (15 mg/day, with the option to increase the dose to 30 mg/day or reduce the dose to 10 mg/day as early as the fourth day of treatment). After being stabilized for 12 consecutive weeks, 337 patients were randomized to continue treatment with the same aripiprazole dose with lithium or valproate or were switched to placebo with lithium or valproate.

During a period of up to 52 weeks, those who remained on aripiprazole experienced fewer manic episodes compared with those on placebo (7 observed episodes vs. 19). The number of depressive episodes was similar in the two groups (14 among those on aripiprazole vs. 18 among those on placebo), according to the statement and revised label. In the study, the most common adverse event associated with adjunctive aripiprazole was tremor, affecting 6%, compared with 2.4% of those on placebo, according to the company statement.

"An examination of population subgroups did not reveal any clear evidence of differential responsiveness on the basis of age and gender," but there were not enough patients in each of the ethnic groups "to adequately assess inter-group differences," according to a statement in the revised label.

Aripiprazole, an oral dopamine partial agonist, also is approved:

• For the acute treatment of manic or mixed episodes associated with bipolar I disorder as monotherapy in adults and pediatric patients aged 10-17 years.

• As monotherapy for the maintenance treatment of bipolar I disorder in adults and adolescents aged 13-17 years.

• For the treatment of schizophrenia in adults and adolescents aged 13-17 years.

• As an adjunctive treatment for adults with major depressive disorder who have an inadequate response to antidepressant therapy.

For the treatment of irritability associated with autistic disorder in pediatric patients aged 6 to 17 years.

It initially was approved by the FDA in 2002 for the treatment of schizophrenia.

The atypical antipsychotic aripiprazole has been approved by the Food and Drug Administration for the maintenance treatment of people with bipolar I disorder as an adjunct to lithium or valproate, the manufacturers announced Feb. 17.

Aripiprazole, marketed as Abilify by the Bristol-Myers Squibb Co. and Otsuka Pharmaceutical Co., was approved in May 2008 as an adjunct to lithium or valproate for the acute treatment of manic and mixed episodes associated with bipolar I disorder.

The approval of the expanded adjunctive indication was based on a 52-week randomized, double-blind, placebo-controlled maintenance study of adults who met the DSM-IV criteria for bipolar I disorder and who had had a recent manic or mixed episode. The subjects also had a history of one or more manic or mixed episodes that had been severe enough to require hospitalization and/or treatment with a mood stabilizer or antipsychotic. All patients started treatment with lithium or valproate; after 2 weeks, those who had an inadequate response to one of the mood stabilizers alone began treatment with aripiprazole (15 mg/day, with the option to increase the dose to 30 mg/day or reduce the dose to 10 mg/day as early as the fourth day of treatment). After being stabilized for 12 consecutive weeks, 337 patients were randomized to continue treatment with the same aripiprazole dose with lithium or valproate or were switched to placebo with lithium or valproate.

During a period of up to 52 weeks, those who remained on aripiprazole experienced fewer manic episodes compared with those on placebo (7 observed episodes vs. 19). The number of depressive episodes was similar in the two groups (14 among those on aripiprazole vs. 18 among those on placebo), according to the statement and revised label. In the study, the most common adverse event associated with adjunctive aripiprazole was tremor, affecting 6%, compared with 2.4% of those on placebo, according to the company statement.

"An examination of population subgroups did not reveal any clear evidence of differential responsiveness on the basis of age and gender," but there were not enough patients in each of the ethnic groups "to adequately assess inter-group differences," according to a statement in the revised label.

Aripiprazole, an oral dopamine partial agonist, also is approved:

• For the acute treatment of manic or mixed episodes associated with bipolar I disorder as monotherapy in adults and pediatric patients aged 10-17 years.

• As monotherapy for the maintenance treatment of bipolar I disorder in adults and adolescents aged 13-17 years.

• For the treatment of schizophrenia in adults and adolescents aged 13-17 years.

• As an adjunctive treatment for adults with major depressive disorder who have an inadequate response to antidepressant therapy.

For the treatment of irritability associated with autistic disorder in pediatric patients aged 6 to 17 years.

It initially was approved by the FDA in 2002 for the treatment of schizophrenia.

The atypical antipsychotic aripiprazole has been approved by the Food and Drug Administration for the maintenance treatment of people with bipolar I disorder as an adjunct to lithium or valproate, the manufacturers announced Feb. 17.

Aripiprazole, marketed as Abilify by the Bristol-Myers Squibb Co. and Otsuka Pharmaceutical Co., was approved in May 2008 as an adjunct to lithium or valproate for the acute treatment of manic and mixed episodes associated with bipolar I disorder.

The approval of the expanded adjunctive indication was based on a 52-week randomized, double-blind, placebo-controlled maintenance study of adults who met the DSM-IV criteria for bipolar I disorder and who had had a recent manic or mixed episode. The subjects also had a history of one or more manic or mixed episodes that had been severe enough to require hospitalization and/or treatment with a mood stabilizer or antipsychotic. All patients started treatment with lithium or valproate; after 2 weeks, those who had an inadequate response to one of the mood stabilizers alone began treatment with aripiprazole (15 mg/day, with the option to increase the dose to 30 mg/day or reduce the dose to 10 mg/day as early as the fourth day of treatment). After being stabilized for 12 consecutive weeks, 337 patients were randomized to continue treatment with the same aripiprazole dose with lithium or valproate or were switched to placebo with lithium or valproate.

During a period of up to 52 weeks, those who remained on aripiprazole experienced fewer manic episodes compared with those on placebo (7 observed episodes vs. 19). The number of depressive episodes was similar in the two groups (14 among those on aripiprazole vs. 18 among those on placebo), according to the statement and revised label. In the study, the most common adverse event associated with adjunctive aripiprazole was tremor, affecting 6%, compared with 2.4% of those on placebo, according to the company statement.

"An examination of population subgroups did not reveal any clear evidence of differential responsiveness on the basis of age and gender," but there were not enough patients in each of the ethnic groups "to adequately assess inter-group differences," according to a statement in the revised label.

Aripiprazole, an oral dopamine partial agonist, also is approved:

• For the acute treatment of manic or mixed episodes associated with bipolar I disorder as monotherapy in adults and pediatric patients aged 10-17 years.

• As monotherapy for the maintenance treatment of bipolar I disorder in adults and adolescents aged 13-17 years.

• For the treatment of schizophrenia in adults and adolescents aged 13-17 years.

• As an adjunctive treatment for adults with major depressive disorder who have an inadequate response to antidepressant therapy.

For the treatment of irritability associated with autistic disorder in pediatric patients aged 6 to 17 years.

It initially was approved by the FDA in 2002 for the treatment of schizophrenia.

LOS ANGELES – It’s reasonable to consider the anticonvulsant valproate as a last option for women with bipolar disorder, given the drug’s associations with the risk of developing isolated features of polycystic ovary syndrome, according to Dr. Harold Carlson.

"I don’t have any problem with that," he said, when an audience member suggested it and also noted the drug’s teratogenicity.

Women under age 25, and particularly adolescents in their midteens, are most at risk for valproate-induced PCOS, usually within the first year of treatment, said Dr. Carlson, professor of endocrinology at Stony Brook (N.Y.) University.

In one study, 9 of 86 women with bipolar disorder (10.5%) treated with valproate developed PCOS; 2 of 144 women with bipolar disorder (1.4%) developed PCOS when treated with other mood stabilizers (Biol. Psychiatry 2006;59:1078-86).

Valproate seems to pose a particular risk for the condition driven by something more than the weight gain caused by the drug.

After all, "the folks [who] gain all that weight on olanzapine don’t get PCOS," Dr. Carlson noted.

Valproate appears to act directly on the ovaries, altering their hormone production. Cultured ovarian cells produce more testosterone in its presence. The excess testosterone shuts off menstruation, and causes acne and hirsutism. Obesity, insulin resistance, and dyslipidemia are problems in PCOS, as well.

When valproate cannot be switched out for a mood stabilizer, prescribing birth control pills at the start of therapy might be a smart move, Dr. Carlson said at a psychopharmacology update, sponsored by the American Academy of Child and Adolescent Psychiatry. That appeared to help prevent PCOS in valproate-treated women in one study (Seizure 2003;12:323-9).

Baseline pelvic ultrasounds might seem like a good idea, too, but they’re "not worth doing," he said.

The reason is that 10%-15% of healthy women have cysts on their ovaries without having PCOS, and ovarian cysts aren’t always present in PCOS.

When women are started on the drug, ask them "every time you see them about their menstrual function. Look at them and see if they are getting acne and hirsutism. Ask them about it. Provide some counseling on diet and exercise to avoid the excessive weight gain, which only makes it worse," Dr. Carlson said.

Should PCOS develop, Metformin is the first-line symptom treatment. Clomiphene can induce ovulation if pregnancy is the goal.

Endocrinology, urology, or gynecology referrals also are in order to help with symptoms, Dr. Carlson said.

He said he is a consultant to Eli Lilly & Co. He also disclosed receiving research funding from GlaxoSmithKline.

LOS ANGELES – It’s reasonable to consider the anticonvulsant valproate as a last option for women with bipolar disorder, given the drug’s associations with the risk of developing isolated features of polycystic ovary syndrome, according to Dr. Harold Carlson.

"I don’t have any problem with that," he said, when an audience member suggested it and also noted the drug’s teratogenicity.

Women under age 25, and particularly adolescents in their midteens, are most at risk for valproate-induced PCOS, usually within the first year of treatment, said Dr. Carlson, professor of endocrinology at Stony Brook (N.Y.) University.

In one study, 9 of 86 women with bipolar disorder (10.5%) treated with valproate developed PCOS; 2 of 144 women with bipolar disorder (1.4%) developed PCOS when treated with other mood stabilizers (Biol. Psychiatry 2006;59:1078-86).

Valproate seems to pose a particular risk for the condition driven by something more than the weight gain caused by the drug.

After all, "the folks [who] gain all that weight on olanzapine don’t get PCOS," Dr. Carlson noted.

Valproate appears to act directly on the ovaries, altering their hormone production. Cultured ovarian cells produce more testosterone in its presence. The excess testosterone shuts off menstruation, and causes acne and hirsutism. Obesity, insulin resistance, and dyslipidemia are problems in PCOS, as well.

When valproate cannot be switched out for a mood stabilizer, prescribing birth control pills at the start of therapy might be a smart move, Dr. Carlson said at a psychopharmacology update, sponsored by the American Academy of Child and Adolescent Psychiatry. That appeared to help prevent PCOS in valproate-treated women in one study (Seizure 2003;12:323-9).

Baseline pelvic ultrasounds might seem like a good idea, too, but they’re "not worth doing," he said.

The reason is that 10%-15% of healthy women have cysts on their ovaries without having PCOS, and ovarian cysts aren’t always present in PCOS.

When women are started on the drug, ask them "every time you see them about their menstrual function. Look at them and see if they are getting acne and hirsutism. Ask them about it. Provide some counseling on diet and exercise to avoid the excessive weight gain, which only makes it worse," Dr. Carlson said.

Should PCOS develop, Metformin is the first-line symptom treatment. Clomiphene can induce ovulation if pregnancy is the goal.

Endocrinology, urology, or gynecology referrals also are in order to help with symptoms, Dr. Carlson said.

He said he is a consultant to Eli Lilly & Co. He also disclosed receiving research funding from GlaxoSmithKline.

LOS ANGELES – It’s reasonable to consider the anticonvulsant valproate as a last option for women with bipolar disorder, given the drug’s associations with the risk of developing isolated features of polycystic ovary syndrome, according to Dr. Harold Carlson.

"I don’t have any problem with that," he said, when an audience member suggested it and also noted the drug’s teratogenicity.

Women under age 25, and particularly adolescents in their midteens, are most at risk for valproate-induced PCOS, usually within the first year of treatment, said Dr. Carlson, professor of endocrinology at Stony Brook (N.Y.) University.

In one study, 9 of 86 women with bipolar disorder (10.5%) treated with valproate developed PCOS; 2 of 144 women with bipolar disorder (1.4%) developed PCOS when treated with other mood stabilizers (Biol. Psychiatry 2006;59:1078-86).

Valproate seems to pose a particular risk for the condition driven by something more than the weight gain caused by the drug.

After all, "the folks [who] gain all that weight on olanzapine don’t get PCOS," Dr. Carlson noted.

Valproate appears to act directly on the ovaries, altering their hormone production. Cultured ovarian cells produce more testosterone in its presence. The excess testosterone shuts off menstruation, and causes acne and hirsutism. Obesity, insulin resistance, and dyslipidemia are problems in PCOS, as well.

When valproate cannot be switched out for a mood stabilizer, prescribing birth control pills at the start of therapy might be a smart move, Dr. Carlson said at a psychopharmacology update, sponsored by the American Academy of Child and Adolescent Psychiatry. That appeared to help prevent PCOS in valproate-treated women in one study (Seizure 2003;12:323-9).

Baseline pelvic ultrasounds might seem like a good idea, too, but they’re "not worth doing," he said.

The reason is that 10%-15% of healthy women have cysts on their ovaries without having PCOS, and ovarian cysts aren’t always present in PCOS.

When women are started on the drug, ask them "every time you see them about their menstrual function. Look at them and see if they are getting acne and hirsutism. Ask them about it. Provide some counseling on diet and exercise to avoid the excessive weight gain, which only makes it worse," Dr. Carlson said.

Should PCOS develop, Metformin is the first-line symptom treatment. Clomiphene can induce ovulation if pregnancy is the goal.

Endocrinology, urology, or gynecology referrals also are in order to help with symptoms, Dr. Carlson said.

He said he is a consultant to Eli Lilly & Co. He also disclosed receiving research funding from GlaxoSmithKline.

M, age 13, is referred by her pediatrician with the chief complaint of “severe mood swings, rule out bipolar disorder (BD).” In the past she was treated for attention-deficit/hyperactivity disorder (ADHD) with stimulants with mixed results. M’s parents are concerned about her “flipping out” whenever she is asked to do something she does not want to do. Her mother has a history of depression and anxiety; her father had a “drinking problem.” There is no history of BD in her first- or second-degree relatives. Are M’s rapid mood swings a sign of BD or another disorder?

The differential diagnosis of “mood swings” is important because they are a common presenting symptom of many children and adolescents with mood and behavioral disorders. Mood swings often occur in children and adolescents with ADHD, oppositional defiant disorder (ODD), developmental disorders, depressive disorders, BD, anxiety disorders, and conduct disorders. Mood swings are analogous to a fever in pediatrics—they indicate something potentially is wrong with the patient, but are not diagnostic as an isolated symptom.

Mood swings in children are common, nonspecific symptoms that more often are a sign of anxiety or behavioral disorders than BD. This article discusses the differential diagnosis of mood swings in children and adolescents and how to best screen and diagnose these patients.

What are ‘mood swings’?

Mood swings is a popular term that is nonspecific and not part of DSM-IV-TR diagnostic criteria for BD. The complaint of “mood swings” may reflect severe mood lability of pediatric patients with BD. This mood lability is best described by the Kiddie-Mania Rating Scale (K-MRS) developed by Axelson and colleagues as “rapid mood variation with several mood states within a brief period of time which appears internally driven without regard to the circumstance.”¹ On K-MRS mood lability items, children with mania typically score:

• Moderate—many mood changes throughout the day, can vary from elevated mood to anger to sadness within a few hours; changes in mood are clearly out of proportion to circumstances and cause impairment in functioning
• Severe—rapid mood swings nearly all of the time, with mood intensity greatly out of proportion to circumstances
• Extreme—constant, explosive variability in mood, several mood changes occurring within minutes, difficult to identify a particular mood, changes in mood radically out of proportion to circumstances.

Patients with BD typically exhibit what is best described as a “mood cycle”—a pronounced shift in mood and energy from 1 extreme to another.² An example of this would be a child who wakes up with extreme silliness, high energy, and intrusive behavior that persists for several hours and then later in the day becomes sad, depressed, and suicidal with no precipitant for either mood cycle. BD patients also will exhibit other symptoms of mania during these mood cycling periods.

Rapid cycling is a DSM-IV course specifier that indicates ≥4 mood episodes per year in patients with BD with a typical course of mania or hypomania followed by depression, or vice versa.³ The episodes must be demarcated by full or partial remission that lasts ≥2 months or by a switch to a mood state of opposite polarity. In the past, children with frequent mood swings were described incorrectly as “rapid cycling,” but this term has been dropped because it engenders confusion between adult and pediatric BD phenomenology.²

A more precise method of describing mood symptoms in a child or adolescent is to use the FIND criteria, which include:⁴

• Frequency of symptoms per week
• Intensity of mood symptoms
• Number of mood cycles per day
• Duration of symptoms per day.

Visit this article at CurrentPsychiatry.com to view a table that outlines what to look for when using the FIND criteria to evaluate common pediatric psychiatric disorders that include mood swings. Table 1
describes clinical characteristics and tools and resources used to differentiate these and other disorders.⁴

Table 1

Clinical characteristics of psychiatric disorders that often feature mood swings

Mood swings: A chart review

We recently completed a retrospective chart review of 100 patients consecutively referred to our pediatric mood disorders clinic for evaluation of “mood swings, rule out BD.” These patients were self-referred, referred by a psychiatrist for a second opinion, or referred by their primary care physician. The mean age of these patients was 8±2.8 years and 68% were male.

Two experienced clinicians (RAK and EM) interviewed each patient and their caregivers and reviewed results of the Conners’ Parent Rating Scale-Revised: Long Form (CPRS-R:L)⁵ and other outside information.
Figure 1 illustrates these patients’ diagnoses. Diagnoses for each of these disorders were made using DSM-IV-TR criteria.³

The most common diagnoses among patients with the chief complaint of mood swings were ADHD (39%); ODD with ADHD (15%); an anxiety disorder, usually generalized anxiety disorder (GAD) (15%); BD (12%); and a secondary mood disorder, usually fetal alcohol spectrum disorder (10%). We were surprised at how often ADHD, ODD, and anxiety disorders were found to be responsible for these patients’ mood swings and how frequently the referring clinician did not recognize these disorders. In the following sections, we discuss each of these disorders and how they differ from BD.

Figure 1 Underlying diagnoses of 100 children/adolescents referred for ‘mood swings’

ADHD: attention-deficit/hyperactivity disorder; BD: bipolar disorder; MDD: major depressive disorder; ODD: oppositional defiant disorder; PDD: pervasive developmental disorder

ADHD and ODD

In our sample, patients with undiagnosed ADHD made up the largest group of those with frequent mood swings. ADHD inattentive type was missed frequently in adolescent girls who still had behavioral aspects of ADHD, including impulsivity and aggression.⁶

The CPRS-R:L is useful for screening and diagnosing children and adolescents with ADHD and ODD. It contains 80 items, can be used in males and females and patients age 3 to 17, and has validated norms by age and sex.⁵ It takes parents approximately 10 minutes to fill out this questionnaire and the results can be scored by hand. The CPRS-R:L includes the following scales: oppositional; cognitive problems/inattention; hyperactivity; anxious-shy; perfectionism; social problems; psychosomatic; Connors’ global index; DSM-IV symptom subscales; and an ADHD index. Patients with mood swings and ADHD combined typically score >2 standard deviations above their age/sex mean on the CPRS-R:L hyperactivity scale, Connors’ Global Index, and ADHD index.⁵

A common childhood disorder, ODD has multiple etiologies.⁷ The first DSM-IV criteria for ODD is “often loses temper”³—essentially mood swings that often are expressed behaviorally as anger and at times as aggressive outbursts.

Dodge and Cole⁸ categorized aggression as reactive (impulsivity with a high affective valence) or proactive (characterized by low arousal and premeditation, ie, predatory conduct disorder). Reactive aggression typically is an angry defensive response to frustration, threat, or provocation, whereas proactive aggression is deliberate, coercive behavior often used to obtain a goal.⁹ Reactive aggression is common among children with ADHD and ODD and typically begins as a mood swing that escalates into reactive aggressive behavior. In a study of 268 consecutively referred children and adolescents with ADHD and 100 community controls, Connor et al¹⁰ found significantly more reactive than proactive forms of aggression in ADHD patients.

It can be difficult to differentiate the moods swings and symptoms of ODD from those of pediatric BD. Mick et al¹¹ found that severe irritability may be a diagnostic indicator of BD in children with ADHD. Using the Kiddie Schedule for Affective Disorders and Schizophrenia (epidemiologic version) structured diagnostic interview,¹² they evaluated 274 children (mean age 10.8±3.2) with ADHD; 37% had no comorbid mood disorder, 36% had ADHD with depression, and 11% had ADHD with BD. Researchers characterized 3 types of irritability in these patients:

• ODD-type irritability characterized by a low frustration tolerance that is seen in ODD
• Mad/cranky irritability found in depressive disorders
• Super-angry/grouchy/cranky irritability with frequent, prolonged, and largely unprovoked anger episodes and characteristics of mania.

ODD-type irritability was common among all ADHD patients, was the least impairing type of irritability, and did not increase the risk of a mood disorder. Mad/cranky irritability was common only in children with ADHD and a mood disorder (depression or BD), was more impairing than ODD-type irritability, and was most predictive of unipolar depression. Super-angry/grouchy/cranky irritability was common only among children with ADHD and BD (77%), was the most impairing, and was predictive of both unipolar depression and BD. The type of irritability and clustering of DSM-IV manic symptoms best differentiated ADHD subjects from those with ADHD and BD. Figure 2 illustrates symptoms that differentiated patients with ADHD from those with ADHD and comorbid BD.¹¹

A review of pharmacotherapy for aggression in children found the largest effects for methylphenidate for aggression in ADHD (mean effect size=0.9, combined N=844).¹³ Our clinical experience has been that pediatric patients with ADHD or ODD with ADHD often have high levels of reactive aggression that presents as mood swings, and aggressively treating ADHD often results in improved mood and other ADHD symptoms.

Figure 2 Symptoms that differentiate BD from BD with comorbid ADHD

ADHD: attention-deficit/hyperactivity disorder; BD: bipolar disorder
Source: Reference 11

Anxiety disorders

The estimated prevalence of child and adolescent anxiety disorders is 10% to 20%¹⁴; in our sample the prevalence was 15%. These disorders include GAD, separation anxiety disorder, social phobias, posttraumatic stress disorder (PTSD), and obsessive-compulsive disorder. Often, children with GAD worry excessively and become upset during transitions when things don’t proceed as they expect, with resultant angry outbursts and mood swings. Mood swings and difficulty sleeping are common in children with anxiety disorders or BD. Anxiety disorders often will be missed unless specific triggers of the mood swings or angry outbursts—as well as differentiating symptoms such as excessive fear, worry, and psychosomatic symptoms—are assessed.

In our clinical experience, simply asking a child if he or she is anxious is not sufficient to uncover an anxiety disorder. Although the CPRS-L:R will screen for anxiety disorders, we have found that the Self-Report for Childhood Anxiety Related Disorders (SCARED) developed by Birmaher et al¹⁵ is more specific. This tool can be used in patients age ≥8. The parent and child versions of the SCARED contain 41 items that measure 5 factors:

• general anxiety
• separation anxiety
• social phobia
• school phobia
• physical symptoms of anxiety.

The SCARED takes 5 minutes to fill out and is available in parent and child versions.

Secondary mood disorders

Many patients in our sample had a mood disorder secondary to the neurologic effects of alcohol on the developing brain. For more about maternal alcohol use, fetal alcohol spectrum disorders, and mood swings, visit this article at CurrentPsychiatry.com.

What BD looks like in children

In our sample, 12% of patients referred for mood swings were diagnosed with bipolar I disorder (BDI), bipolar II disorder (BDII), or bipolar disorder, not otherwise specified (BD-NOS). In the United States, lifetime prevalence of BDI and BDII in adolescents age 13 to 17 is 2.9%.¹⁶ No large epidemiologic studies have looked at the lifetime prevalence of BD in children age <13.

How often a clinician sees BD in children and adolescents largely depends on the type of setting in which he or she practices. Although in the general population BD is relatively rare compared with other childhood psychiatric disorders, on child/adolescent inpatient units it is common to find that 30% to 40% of patients have BD.¹⁷

The best longitudinal study to date of the phenomenology, comorbidity, and outcome of BD in children and adolescents is the National Institute of Mental Health-funded Course and Outcome of Bipolar Youth study (COBY).¹⁸ In this ongoing, longitudinal study, 413 youths (age 7 to 17) with BDI (N=244), BDII (N=28), or BD-NOS (N=141) were rigorously diagnosed using state-of-the-art measures, including the Kiddie Schedule for Affective Disorders and Schizophrenia for School-Age Children-Present version¹⁹ and re-evaluated every 9.4 months for 4 years. When organizing this study, investigators found that DSM-IV criteria for BD-NOS were too vague to be useful and developed their own criteria (Table 2).¹⁸

For BDI patients in the COBY study, the mean age of onset for bipolar symptoms was 9.0±4.1 years and the mean duration of illness was 4.4±3.1 years. Researchers reported that at the 4-year assessment approximately 70% of patients with BD recovered from their index episode, and 50% had at least 1 syndromal recurrence, particularly depressive episodes.²⁰ Analyses of these patients’ weekly mood symptoms showed that they had syndromal or subsyndromal symptoms with numerous changes in symptoms and shifts of mood polarity 60% of the time, and psychosis 3% of the time. During this study, 20% of BDII patients progressed to BDI, and 25% of BD-NOS patients converted to BDI or BDII.

Further analysis of the COBY data revealed that onset of mood symptoms preceded onset of clear bipolar episodes by an average of 1.0±1.7 years. Depression was the most common initial and most frequent episode for adolescents; mood lability was seen more often in childhood-onset and adolescents with early-onset BD. Depressed children had more severe irritability than depressed adolescents, and older age was associated with more severe and typical mood symptomatology.²¹

The clinical picture of a child with BD that emerges from the COBY study is:

• a fairly young child with the onset of mood symptoms between age 5 to 12
• subsyndromal and less frequently clear syndromal episodes
• primarily mixed and depressed symptoms with rapid mood cycles during these episodes.²²

It is clear that there is a spectrum of bipolar disorders in children and adolescents with varying degrees of symptom expression and children differ from adolescents and adults in their initial presentation of BD.

Table 2

COBY criteria for bipolar disorder, not otherwise specified

Related Resources

• Kowatch RA, Fristad MA, Findling RL, et al. Clinical manual for management of bipolar disorder in children and adolescents. Arlington, VA: American Psychiatric Publishing, Inc.; 2009.
• Goodwin FK, Jamison KR. Manic-depressive illness. 2nd ed. Oxford, United Kingdom: Oxford University Press; 2007.
• Miklowitz DJ, Cicchetti D, eds. Understanding bipolar disorder: a developmental perspective. New York, NY: Guilford Press; 2010.

Drug Brand Name

• Methylphenidate • Ritalin, Concerta, others

Disclosures

Dr. Kowatch receives grant/research support from the National Institute of Child Health and Human Development and the National Institute of Mental Health and is a consultant to AstraZeneca, Forest Pharmaceuticals, Merck, and the REACH Foundation.

Dr. Delgado and Ms. Monroe report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Table

FIND criteria of disorders found to cause mood swings

Box

M, age 13, is referred by her pediatrician with the chief complaint of “severe mood swings, rule out bipolar disorder (BD).” In the past she was treated for attention-deficit/hyperactivity disorder (ADHD) with stimulants with mixed results. M’s parents are concerned about her “flipping out” whenever she is asked to do something she does not want to do. Her mother has a history of depression and anxiety; her father had a “drinking problem.” There is no history of BD in her first- or second-degree relatives. Are M’s rapid mood swings a sign of BD or another disorder?

The differential diagnosis of “mood swings” is important because they are a common presenting symptom of many children and adolescents with mood and behavioral disorders. Mood swings often occur in children and adolescents with ADHD, oppositional defiant disorder (ODD), developmental disorders, depressive disorders, BD, anxiety disorders, and conduct disorders. Mood swings are analogous to a fever in pediatrics—they indicate something potentially is wrong with the patient, but are not diagnostic as an isolated symptom.

Mood swings in children are common, nonspecific symptoms that more often are a sign of anxiety or behavioral disorders than BD. This article discusses the differential diagnosis of mood swings in children and adolescents and how to best screen and diagnose these patients.

What are ‘mood swings’?

Mood swings is a popular term that is nonspecific and not part of DSM-IV-TR diagnostic criteria for BD. The complaint of “mood swings” may reflect severe mood lability of pediatric patients with BD. This mood lability is best described by the Kiddie-Mania Rating Scale (K-MRS) developed by Axelson and colleagues as “rapid mood variation with several mood states within a brief period of time which appears internally driven without regard to the circumstance.”¹ On K-MRS mood lability items, children with mania typically score:

• Moderate—many mood changes throughout the day, can vary from elevated mood to anger to sadness within a few hours; changes in mood are clearly out of proportion to circumstances and cause impairment in functioning
• Severe—rapid mood swings nearly all of the time, with mood intensity greatly out of proportion to circumstances
• Extreme—constant, explosive variability in mood, several mood changes occurring within minutes, difficult to identify a particular mood, changes in mood radically out of proportion to circumstances.

Patients with BD typically exhibit what is best described as a “mood cycle”—a pronounced shift in mood and energy from 1 extreme to another.² An example of this would be a child who wakes up with extreme silliness, high energy, and intrusive behavior that persists for several hours and then later in the day becomes sad, depressed, and suicidal with no precipitant for either mood cycle. BD patients also will exhibit other symptoms of mania during these mood cycling periods.

Rapid cycling is a DSM-IV course specifier that indicates ≥4 mood episodes per year in patients with BD with a typical course of mania or hypomania followed by depression, or vice versa.³ The episodes must be demarcated by full or partial remission that lasts ≥2 months or by a switch to a mood state of opposite polarity. In the past, children with frequent mood swings were described incorrectly as “rapid cycling,” but this term has been dropped because it engenders confusion between adult and pediatric BD phenomenology.²

A more precise method of describing mood symptoms in a child or adolescent is to use the FIND criteria, which include:⁴

• Frequency of symptoms per week
• Intensity of mood symptoms
• Number of mood cycles per day
• Duration of symptoms per day.

Visit this article at CurrentPsychiatry.com to view a table that outlines what to look for when using the FIND criteria to evaluate common pediatric psychiatric disorders that include mood swings. Table 1
describes clinical characteristics and tools and resources used to differentiate these and other disorders.⁴

Table 1

Clinical characteristics of psychiatric disorders that often feature mood swings

Mood swings: A chart review

We recently completed a retrospective chart review of 100 patients consecutively referred to our pediatric mood disorders clinic for evaluation of “mood swings, rule out BD.” These patients were self-referred, referred by a psychiatrist for a second opinion, or referred by their primary care physician. The mean age of these patients was 8±2.8 years and 68% were male.

Two experienced clinicians (RAK and EM) interviewed each patient and their caregivers and reviewed results of the Conners’ Parent Rating Scale-Revised: Long Form (CPRS-R:L)⁵ and other outside information.
Figure 1 illustrates these patients’ diagnoses. Diagnoses for each of these disorders were made using DSM-IV-TR criteria.³

The most common diagnoses among patients with the chief complaint of mood swings were ADHD (39%); ODD with ADHD (15%); an anxiety disorder, usually generalized anxiety disorder (GAD) (15%); BD (12%); and a secondary mood disorder, usually fetal alcohol spectrum disorder (10%). We were surprised at how often ADHD, ODD, and anxiety disorders were found to be responsible for these patients’ mood swings and how frequently the referring clinician did not recognize these disorders. In the following sections, we discuss each of these disorders and how they differ from BD.

Figure 1 Underlying diagnoses of 100 children/adolescents referred for ‘mood swings’

ADHD: attention-deficit/hyperactivity disorder; BD: bipolar disorder; MDD: major depressive disorder; ODD: oppositional defiant disorder; PDD: pervasive developmental disorder

ADHD and ODD

In our sample, patients with undiagnosed ADHD made up the largest group of those with frequent mood swings. ADHD inattentive type was missed frequently in adolescent girls who still had behavioral aspects of ADHD, including impulsivity and aggression.⁶

The CPRS-R:L is useful for screening and diagnosing children and adolescents with ADHD and ODD. It contains 80 items, can be used in males and females and patients age 3 to 17, and has validated norms by age and sex.⁵ It takes parents approximately 10 minutes to fill out this questionnaire and the results can be scored by hand. The CPRS-R:L includes the following scales: oppositional; cognitive problems/inattention; hyperactivity; anxious-shy; perfectionism; social problems; psychosomatic; Connors’ global index; DSM-IV symptom subscales; and an ADHD index. Patients with mood swings and ADHD combined typically score >2 standard deviations above their age/sex mean on the CPRS-R:L hyperactivity scale, Connors’ Global Index, and ADHD index.⁵

A common childhood disorder, ODD has multiple etiologies.⁷ The first DSM-IV criteria for ODD is “often loses temper”³—essentially mood swings that often are expressed behaviorally as anger and at times as aggressive outbursts.

Dodge and Cole⁸ categorized aggression as reactive (impulsivity with a high affective valence) or proactive (characterized by low arousal and premeditation, ie, predatory conduct disorder). Reactive aggression typically is an angry defensive response to frustration, threat, or provocation, whereas proactive aggression is deliberate, coercive behavior often used to obtain a goal.⁹ Reactive aggression is common among children with ADHD and ODD and typically begins as a mood swing that escalates into reactive aggressive behavior. In a study of 268 consecutively referred children and adolescents with ADHD and 100 community controls, Connor et al¹⁰ found significantly more reactive than proactive forms of aggression in ADHD patients.

It can be difficult to differentiate the moods swings and symptoms of ODD from those of pediatric BD. Mick et al¹¹ found that severe irritability may be a diagnostic indicator of BD in children with ADHD. Using the Kiddie Schedule for Affective Disorders and Schizophrenia (epidemiologic version) structured diagnostic interview,¹² they evaluated 274 children (mean age 10.8±3.2) with ADHD; 37% had no comorbid mood disorder, 36% had ADHD with depression, and 11% had ADHD with BD. Researchers characterized 3 types of irritability in these patients:

• ODD-type irritability characterized by a low frustration tolerance that is seen in ODD
• Mad/cranky irritability found in depressive disorders
• Super-angry/grouchy/cranky irritability with frequent, prolonged, and largely unprovoked anger episodes and characteristics of mania.

ODD-type irritability was common among all ADHD patients, was the least impairing type of irritability, and did not increase the risk of a mood disorder. Mad/cranky irritability was common only in children with ADHD and a mood disorder (depression or BD), was more impairing than ODD-type irritability, and was most predictive of unipolar depression. Super-angry/grouchy/cranky irritability was common only among children with ADHD and BD (77%), was the most impairing, and was predictive of both unipolar depression and BD. The type of irritability and clustering of DSM-IV manic symptoms best differentiated ADHD subjects from those with ADHD and BD. Figure 2 illustrates symptoms that differentiated patients with ADHD from those with ADHD and comorbid BD.¹¹

A review of pharmacotherapy for aggression in children found the largest effects for methylphenidate for aggression in ADHD (mean effect size=0.9, combined N=844).¹³ Our clinical experience has been that pediatric patients with ADHD or ODD with ADHD often have high levels of reactive aggression that presents as mood swings, and aggressively treating ADHD often results in improved mood and other ADHD symptoms.

Figure 2 Symptoms that differentiate BD from BD with comorbid ADHD

ADHD: attention-deficit/hyperactivity disorder; BD: bipolar disorder
Source: Reference 11

Anxiety disorders

The estimated prevalence of child and adolescent anxiety disorders is 10% to 20%¹⁴; in our sample the prevalence was 15%. These disorders include GAD, separation anxiety disorder, social phobias, posttraumatic stress disorder (PTSD), and obsessive-compulsive disorder. Often, children with GAD worry excessively and become upset during transitions when things don’t proceed as they expect, with resultant angry outbursts and mood swings. Mood swings and difficulty sleeping are common in children with anxiety disorders or BD. Anxiety disorders often will be missed unless specific triggers of the mood swings or angry outbursts—as well as differentiating symptoms such as excessive fear, worry, and psychosomatic symptoms—are assessed.

In our clinical experience, simply asking a child if he or she is anxious is not sufficient to uncover an anxiety disorder. Although the CPRS-L:R will screen for anxiety disorders, we have found that the Self-Report for Childhood Anxiety Related Disorders (SCARED) developed by Birmaher et al¹⁵ is more specific. This tool can be used in patients age ≥8. The parent and child versions of the SCARED contain 41 items that measure 5 factors:

• general anxiety
• separation anxiety
• social phobia
• school phobia
• physical symptoms of anxiety.

The SCARED takes 5 minutes to fill out and is available in parent and child versions.

Secondary mood disorders

Many patients in our sample had a mood disorder secondary to the neurologic effects of alcohol on the developing brain. For more about maternal alcohol use, fetal alcohol spectrum disorders, and mood swings, visit this article at CurrentPsychiatry.com.

What BD looks like in children

In our sample, 12% of patients referred for mood swings were diagnosed with bipolar I disorder (BDI), bipolar II disorder (BDII), or bipolar disorder, not otherwise specified (BD-NOS). In the United States, lifetime prevalence of BDI and BDII in adolescents age 13 to 17 is 2.9%.¹⁶ No large epidemiologic studies have looked at the lifetime prevalence of BD in children age <13.

How often a clinician sees BD in children and adolescents largely depends on the type of setting in which he or she practices. Although in the general population BD is relatively rare compared with other childhood psychiatric disorders, on child/adolescent inpatient units it is common to find that 30% to 40% of patients have BD.¹⁷

The best longitudinal study to date of the phenomenology, comorbidity, and outcome of BD in children and adolescents is the National Institute of Mental Health-funded Course and Outcome of Bipolar Youth study (COBY).¹⁸ In this ongoing, longitudinal study, 413 youths (age 7 to 17) with BDI (N=244), BDII (N=28), or BD-NOS (N=141) were rigorously diagnosed using state-of-the-art measures, including the Kiddie Schedule for Affective Disorders and Schizophrenia for School-Age Children-Present version¹⁹ and re-evaluated every 9.4 months for 4 years. When organizing this study, investigators found that DSM-IV criteria for BD-NOS were too vague to be useful and developed their own criteria (Table 2).¹⁸

For BDI patients in the COBY study, the mean age of onset for bipolar symptoms was 9.0±4.1 years and the mean duration of illness was 4.4±3.1 years. Researchers reported that at the 4-year assessment approximately 70% of patients with BD recovered from their index episode, and 50% had at least 1 syndromal recurrence, particularly depressive episodes.²⁰ Analyses of these patients’ weekly mood symptoms showed that they had syndromal or subsyndromal symptoms with numerous changes in symptoms and shifts of mood polarity 60% of the time, and psychosis 3% of the time. During this study, 20% of BDII patients progressed to BDI, and 25% of BD-NOS patients converted to BDI or BDII.

Further analysis of the COBY data revealed that onset of mood symptoms preceded onset of clear bipolar episodes by an average of 1.0±1.7 years. Depression was the most common initial and most frequent episode for adolescents; mood lability was seen more often in childhood-onset and adolescents with early-onset BD. Depressed children had more severe irritability than depressed adolescents, and older age was associated with more severe and typical mood symptomatology.²¹

The clinical picture of a child with BD that emerges from the COBY study is:

• a fairly young child with the onset of mood symptoms between age 5 to 12
• subsyndromal and less frequently clear syndromal episodes
• primarily mixed and depressed symptoms with rapid mood cycles during these episodes.²²

It is clear that there is a spectrum of bipolar disorders in children and adolescents with varying degrees of symptom expression and children differ from adolescents and adults in their initial presentation of BD.

Table 2

COBY criteria for bipolar disorder, not otherwise specified

Related Resources

• Kowatch RA, Fristad MA, Findling RL, et al. Clinical manual for management of bipolar disorder in children and adolescents. Arlington, VA: American Psychiatric Publishing, Inc.; 2009.
• Goodwin FK, Jamison KR. Manic-depressive illness. 2nd ed. Oxford, United Kingdom: Oxford University Press; 2007.
• Miklowitz DJ, Cicchetti D, eds. Understanding bipolar disorder: a developmental perspective. New York, NY: Guilford Press; 2010.

Drug Brand Name

• Methylphenidate • Ritalin, Concerta, others

Disclosures

Dr. Kowatch receives grant/research support from the National Institute of Child Health and Human Development and the National Institute of Mental Health and is a consultant to AstraZeneca, Forest Pharmaceuticals, Merck, and the REACH Foundation.

Dr. Delgado and Ms. Monroe report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Table

FIND criteria of disorders found to cause mood swings

Box

M, age 13, is referred by her pediatrician with the chief complaint of “severe mood swings, rule out bipolar disorder (BD).” In the past she was treated for attention-deficit/hyperactivity disorder (ADHD) with stimulants with mixed results. M’s parents are concerned about her “flipping out” whenever she is asked to do something she does not want to do. Her mother has a history of depression and anxiety; her father had a “drinking problem.” There is no history of BD in her first- or second-degree relatives. Are M’s rapid mood swings a sign of BD or another disorder?

The differential diagnosis of “mood swings” is important because they are a common presenting symptom of many children and adolescents with mood and behavioral disorders. Mood swings often occur in children and adolescents with ADHD, oppositional defiant disorder (ODD), developmental disorders, depressive disorders, BD, anxiety disorders, and conduct disorders. Mood swings are analogous to a fever in pediatrics—they indicate something potentially is wrong with the patient, but are not diagnostic as an isolated symptom.

Mood swings in children are common, nonspecific symptoms that more often are a sign of anxiety or behavioral disorders than BD. This article discusses the differential diagnosis of mood swings in children and adolescents and how to best screen and diagnose these patients.

What are ‘mood swings’?

Mood swings is a popular term that is nonspecific and not part of DSM-IV-TR diagnostic criteria for BD. The complaint of “mood swings” may reflect severe mood lability of pediatric patients with BD. This mood lability is best described by the Kiddie-Mania Rating Scale (K-MRS) developed by Axelson and colleagues as “rapid mood variation with several mood states within a brief period of time which appears internally driven without regard to the circumstance.”¹ On K-MRS mood lability items, children with mania typically score:

• Moderate—many mood changes throughout the day, can vary from elevated mood to anger to sadness within a few hours; changes in mood are clearly out of proportion to circumstances and cause impairment in functioning
• Severe—rapid mood swings nearly all of the time, with mood intensity greatly out of proportion to circumstances
• Extreme—constant, explosive variability in mood, several mood changes occurring within minutes, difficult to identify a particular mood, changes in mood radically out of proportion to circumstances.

Patients with BD typically exhibit what is best described as a “mood cycle”—a pronounced shift in mood and energy from 1 extreme to another.² An example of this would be a child who wakes up with extreme silliness, high energy, and intrusive behavior that persists for several hours and then later in the day becomes sad, depressed, and suicidal with no precipitant for either mood cycle. BD patients also will exhibit other symptoms of mania during these mood cycling periods.

Rapid cycling is a DSM-IV course specifier that indicates ≥4 mood episodes per year in patients with BD with a typical course of mania or hypomania followed by depression, or vice versa.³ The episodes must be demarcated by full or partial remission that lasts ≥2 months or by a switch to a mood state of opposite polarity. In the past, children with frequent mood swings were described incorrectly as “rapid cycling,” but this term has been dropped because it engenders confusion between adult and pediatric BD phenomenology.²

A more precise method of describing mood symptoms in a child or adolescent is to use the FIND criteria, which include:⁴

• Frequency of symptoms per week
• Intensity of mood symptoms
• Number of mood cycles per day
• Duration of symptoms per day.

Visit this article at CurrentPsychiatry.com to view a table that outlines what to look for when using the FIND criteria to evaluate common pediatric psychiatric disorders that include mood swings. Table 1
describes clinical characteristics and tools and resources used to differentiate these and other disorders.⁴

Table 1

Clinical characteristics of psychiatric disorders that often feature mood swings

Mood swings: A chart review

We recently completed a retrospective chart review of 100 patients consecutively referred to our pediatric mood disorders clinic for evaluation of “mood swings, rule out BD.” These patients were self-referred, referred by a psychiatrist for a second opinion, or referred by their primary care physician. The mean age of these patients was 8±2.8 years and 68% were male.

Two experienced clinicians (RAK and EM) interviewed each patient and their caregivers and reviewed results of the Conners’ Parent Rating Scale-Revised: Long Form (CPRS-R:L)⁵ and other outside information.
Figure 1 illustrates these patients’ diagnoses. Diagnoses for each of these disorders were made using DSM-IV-TR criteria.³

The most common diagnoses among patients with the chief complaint of mood swings were ADHD (39%); ODD with ADHD (15%); an anxiety disorder, usually generalized anxiety disorder (GAD) (15%); BD (12%); and a secondary mood disorder, usually fetal alcohol spectrum disorder (10%). We were surprised at how often ADHD, ODD, and anxiety disorders were found to be responsible for these patients’ mood swings and how frequently the referring clinician did not recognize these disorders. In the following sections, we discuss each of these disorders and how they differ from BD.

Figure 1 Underlying diagnoses of 100 children/adolescents referred for ‘mood swings’

ADHD: attention-deficit/hyperactivity disorder; BD: bipolar disorder; MDD: major depressive disorder; ODD: oppositional defiant disorder; PDD: pervasive developmental disorder

ADHD and ODD

In our sample, patients with undiagnosed ADHD made up the largest group of those with frequent mood swings. ADHD inattentive type was missed frequently in adolescent girls who still had behavioral aspects of ADHD, including impulsivity and aggression.⁶

The CPRS-R:L is useful for screening and diagnosing children and adolescents with ADHD and ODD. It contains 80 items, can be used in males and females and patients age 3 to 17, and has validated norms by age and sex.⁵ It takes parents approximately 10 minutes to fill out this questionnaire and the results can be scored by hand. The CPRS-R:L includes the following scales: oppositional; cognitive problems/inattention; hyperactivity; anxious-shy; perfectionism; social problems; psychosomatic; Connors’ global index; DSM-IV symptom subscales; and an ADHD index. Patients with mood swings and ADHD combined typically score >2 standard deviations above their age/sex mean on the CPRS-R:L hyperactivity scale, Connors’ Global Index, and ADHD index.⁵

A common childhood disorder, ODD has multiple etiologies.⁷ The first DSM-IV criteria for ODD is “often loses temper”³—essentially mood swings that often are expressed behaviorally as anger and at times as aggressive outbursts.

Dodge and Cole⁸ categorized aggression as reactive (impulsivity with a high affective valence) or proactive (characterized by low arousal and premeditation, ie, predatory conduct disorder). Reactive aggression typically is an angry defensive response to frustration, threat, or provocation, whereas proactive aggression is deliberate, coercive behavior often used to obtain a goal.⁹ Reactive aggression is common among children with ADHD and ODD and typically begins as a mood swing that escalates into reactive aggressive behavior. In a study of 268 consecutively referred children and adolescents with ADHD and 100 community controls, Connor et al¹⁰ found significantly more reactive than proactive forms of aggression in ADHD patients.

It can be difficult to differentiate the moods swings and symptoms of ODD from those of pediatric BD. Mick et al¹¹ found that severe irritability may be a diagnostic indicator of BD in children with ADHD. Using the Kiddie Schedule for Affective Disorders and Schizophrenia (epidemiologic version) structured diagnostic interview,¹² they evaluated 274 children (mean age 10.8±3.2) with ADHD; 37% had no comorbid mood disorder, 36% had ADHD with depression, and 11% had ADHD with BD. Researchers characterized 3 types of irritability in these patients:

• ODD-type irritability characterized by a low frustration tolerance that is seen in ODD
• Mad/cranky irritability found in depressive disorders
• Super-angry/grouchy/cranky irritability with frequent, prolonged, and largely unprovoked anger episodes and characteristics of mania.

ODD-type irritability was common among all ADHD patients, was the least impairing type of irritability, and did not increase the risk of a mood disorder. Mad/cranky irritability was common only in children with ADHD and a mood disorder (depression or BD), was more impairing than ODD-type irritability, and was most predictive of unipolar depression. Super-angry/grouchy/cranky irritability was common only among children with ADHD and BD (77%), was the most impairing, and was predictive of both unipolar depression and BD. The type of irritability and clustering of DSM-IV manic symptoms best differentiated ADHD subjects from those with ADHD and BD. Figure 2 illustrates symptoms that differentiated patients with ADHD from those with ADHD and comorbid BD.¹¹

A review of pharmacotherapy for aggression in children found the largest effects for methylphenidate for aggression in ADHD (mean effect size=0.9, combined N=844).¹³ Our clinical experience has been that pediatric patients with ADHD or ODD with ADHD often have high levels of reactive aggression that presents as mood swings, and aggressively treating ADHD often results in improved mood and other ADHD symptoms.

Figure 2 Symptoms that differentiate BD from BD with comorbid ADHD

ADHD: attention-deficit/hyperactivity disorder; BD: bipolar disorder
Source: Reference 11

Anxiety disorders

The estimated prevalence of child and adolescent anxiety disorders is 10% to 20%¹⁴; in our sample the prevalence was 15%. These disorders include GAD, separation anxiety disorder, social phobias, posttraumatic stress disorder (PTSD), and obsessive-compulsive disorder. Often, children with GAD worry excessively and become upset during transitions when things don’t proceed as they expect, with resultant angry outbursts and mood swings. Mood swings and difficulty sleeping are common in children with anxiety disorders or BD. Anxiety disorders often will be missed unless specific triggers of the mood swings or angry outbursts—as well as differentiating symptoms such as excessive fear, worry, and psychosomatic symptoms—are assessed.

In our clinical experience, simply asking a child if he or she is anxious is not sufficient to uncover an anxiety disorder. Although the CPRS-L:R will screen for anxiety disorders, we have found that the Self-Report for Childhood Anxiety Related Disorders (SCARED) developed by Birmaher et al¹⁵ is more specific. This tool can be used in patients age ≥8. The parent and child versions of the SCARED contain 41 items that measure 5 factors:

• general anxiety
• separation anxiety
• social phobia
• school phobia
• physical symptoms of anxiety.

The SCARED takes 5 minutes to fill out and is available in parent and child versions.

Secondary mood disorders

Many patients in our sample had a mood disorder secondary to the neurologic effects of alcohol on the developing brain. For more about maternal alcohol use, fetal alcohol spectrum disorders, and mood swings, visit this article at CurrentPsychiatry.com.

What BD looks like in children

In our sample, 12% of patients referred for mood swings were diagnosed with bipolar I disorder (BDI), bipolar II disorder (BDII), or bipolar disorder, not otherwise specified (BD-NOS). In the United States, lifetime prevalence of BDI and BDII in adolescents age 13 to 17 is 2.9%.¹⁶ No large epidemiologic studies have looked at the lifetime prevalence of BD in children age <13.

How often a clinician sees BD in children and adolescents largely depends on the type of setting in which he or she practices. Although in the general population BD is relatively rare compared with other childhood psychiatric disorders, on child/adolescent inpatient units it is common to find that 30% to 40% of patients have BD.¹⁷

The best longitudinal study to date of the phenomenology, comorbidity, and outcome of BD in children and adolescents is the National Institute of Mental Health-funded Course and Outcome of Bipolar Youth study (COBY).¹⁸ In this ongoing, longitudinal study, 413 youths (age 7 to 17) with BDI (N=244), BDII (N=28), or BD-NOS (N=141) were rigorously diagnosed using state-of-the-art measures, including the Kiddie Schedule for Affective Disorders and Schizophrenia for School-Age Children-Present version¹⁹ and re-evaluated every 9.4 months for 4 years. When organizing this study, investigators found that DSM-IV criteria for BD-NOS were too vague to be useful and developed their own criteria (Table 2).¹⁸

For BDI patients in the COBY study, the mean age of onset for bipolar symptoms was 9.0±4.1 years and the mean duration of illness was 4.4±3.1 years. Researchers reported that at the 4-year assessment approximately 70% of patients with BD recovered from their index episode, and 50% had at least 1 syndromal recurrence, particularly depressive episodes.²⁰ Analyses of these patients’ weekly mood symptoms showed that they had syndromal or subsyndromal symptoms with numerous changes in symptoms and shifts of mood polarity 60% of the time, and psychosis 3% of the time. During this study, 20% of BDII patients progressed to BDI, and 25% of BD-NOS patients converted to BDI or BDII.

Further analysis of the COBY data revealed that onset of mood symptoms preceded onset of clear bipolar episodes by an average of 1.0±1.7 years. Depression was the most common initial and most frequent episode for adolescents; mood lability was seen more often in childhood-onset and adolescents with early-onset BD. Depressed children had more severe irritability than depressed adolescents, and older age was associated with more severe and typical mood symptomatology.²¹

The clinical picture of a child with BD that emerges from the COBY study is:

• a fairly young child with the onset of mood symptoms between age 5 to 12
• subsyndromal and less frequently clear syndromal episodes
• primarily mixed and depressed symptoms with rapid mood cycles during these episodes.²²

It is clear that there is a spectrum of bipolar disorders in children and adolescents with varying degrees of symptom expression and children differ from adolescents and adults in their initial presentation of BD.

Table 2

COBY criteria for bipolar disorder, not otherwise specified

Related Resources

• Kowatch RA, Fristad MA, Findling RL, et al. Clinical manual for management of bipolar disorder in children and adolescents. Arlington, VA: American Psychiatric Publishing, Inc.; 2009.
• Goodwin FK, Jamison KR. Manic-depressive illness. 2nd ed. Oxford, United Kingdom: Oxford University Press; 2007.
• Miklowitz DJ, Cicchetti D, eds. Understanding bipolar disorder: a developmental perspective. New York, NY: Guilford Press; 2010.

Drug Brand Name

• Methylphenidate • Ritalin, Concerta, others

Disclosures

Dr. Kowatch receives grant/research support from the National Institute of Child Health and Human Development and the National Institute of Mental Health and is a consultant to AstraZeneca, Forest Pharmaceuticals, Merck, and the REACH Foundation.

Dr. Delgado and Ms. Monroe report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Table

FIND criteria of disorders found to cause mood swings

Box