Current Psychiatry

The article “Subsyndromal depression: Help your bipolar patients feel better” (Current Psychiatry, August 2008) brought to mind what may seem like a hypothetical question but is one I struggle with every day. What emotions are bipolar patients allowed to feel without necessarily leading to medical intervention?

When a loved one dies is the bipolar patient permitted to feel sad, have no sense of pleasure, become tearful, and have difficulty sleeping and concentrating? An individual who experiences these feelings but is not bipolar would be considered to be grieving—a very natural human emotion. What should we do for a bipolar patient?

I tell my patients that with medication I hope to restore the normal range of emotions that an average, healthy person would feel under similar circumstances, and I would address emotional reactions that clearly are beyond that range, such as suicidality or delusional guilt. Am I misinforming my patient?

Mohamed Dattu, MD
Presto, PA

Dr. Ostacher responds

Dr. Dattu brings up what can be a lifelong dilemma for patients with bipolar disorder and their families and suggests a well-conceived response to those in his care. After suffering through a number of mood episodes, patients often find it hard to judge whether they are experiencing bipolar symptoms or merely having emotions and feelings that are normal and expected. Many of my patients report being traumatized in some way by their episodes, so the appearance of any symptoms—sadness, anxiety, excitement, or insomnia—may make them worried that they will relapse to a new mood episode. Just because a patient has bipolar disorder doesn’t mean that these experiences are pathologic.

Psychiatrists use clinical judgment to decide when to intervene. If functioning becomes significantly impaired, thoughts and risk of self-harm appear, and symptoms persist for months, then intervention—whether changes in medical or psychosocial treatment—probably is necessary.

Michael J. Ostacher, MD, MPH
Assistant professor of psychiatry
Harvard Medical School
Boston

The article “Subsyndromal depression: Help your bipolar patients feel better” (Current Psychiatry, August 2008) brought to mind what may seem like a hypothetical question but is one I struggle with every day. What emotions are bipolar patients allowed to feel without necessarily leading to medical intervention?

When a loved one dies is the bipolar patient permitted to feel sad, have no sense of pleasure, become tearful, and have difficulty sleeping and concentrating? An individual who experiences these feelings but is not bipolar would be considered to be grieving—a very natural human emotion. What should we do for a bipolar patient?

I tell my patients that with medication I hope to restore the normal range of emotions that an average, healthy person would feel under similar circumstances, and I would address emotional reactions that clearly are beyond that range, such as suicidality or delusional guilt. Am I misinforming my patient?

Mohamed Dattu, MD
Presto, PA

Dr. Ostacher responds

Dr. Dattu brings up what can be a lifelong dilemma for patients with bipolar disorder and their families and suggests a well-conceived response to those in his care. After suffering through a number of mood episodes, patients often find it hard to judge whether they are experiencing bipolar symptoms or merely having emotions and feelings that are normal and expected. Many of my patients report being traumatized in some way by their episodes, so the appearance of any symptoms—sadness, anxiety, excitement, or insomnia—may make them worried that they will relapse to a new mood episode. Just because a patient has bipolar disorder doesn’t mean that these experiences are pathologic.

Psychiatrists use clinical judgment to decide when to intervene. If functioning becomes significantly impaired, thoughts and risk of self-harm appear, and symptoms persist for months, then intervention—whether changes in medical or psychosocial treatment—probably is necessary.

Michael J. Ostacher, MD, MPH
Assistant professor of psychiatry
Harvard Medical School
Boston

The article “Subsyndromal depression: Help your bipolar patients feel better” (Current Psychiatry, August 2008) brought to mind what may seem like a hypothetical question but is one I struggle with every day. What emotions are bipolar patients allowed to feel without necessarily leading to medical intervention?

When a loved one dies is the bipolar patient permitted to feel sad, have no sense of pleasure, become tearful, and have difficulty sleeping and concentrating? An individual who experiences these feelings but is not bipolar would be considered to be grieving—a very natural human emotion. What should we do for a bipolar patient?

I tell my patients that with medication I hope to restore the normal range of emotions that an average, healthy person would feel under similar circumstances, and I would address emotional reactions that clearly are beyond that range, such as suicidality or delusional guilt. Am I misinforming my patient?

Mohamed Dattu, MD
Presto, PA

Dr. Ostacher responds

Dr. Dattu brings up what can be a lifelong dilemma for patients with bipolar disorder and their families and suggests a well-conceived response to those in his care. After suffering through a number of mood episodes, patients often find it hard to judge whether they are experiencing bipolar symptoms or merely having emotions and feelings that are normal and expected. Many of my patients report being traumatized in some way by their episodes, so the appearance of any symptoms—sadness, anxiety, excitement, or insomnia—may make them worried that they will relapse to a new mood episode. Just because a patient has bipolar disorder doesn’t mean that these experiences are pathologic.

Psychiatrists use clinical judgment to decide when to intervene. If functioning becomes significantly impaired, thoughts and risk of self-harm appear, and symptoms persist for months, then intervention—whether changes in medical or psychosocial treatment—probably is necessary.

Michael J. Ostacher, MD, MPH
Assistant professor of psychiatry
Harvard Medical School
Boston

Medication side effects are generally regarded as the Achilles’ heel of pharmacologic treatment. And who can argue with that? Adverse effects are the downside of drug treatment of psychiatric disorders and are blamed for tolerability and adherence problems. Patients dread side effects, physicians feel uncomfortable or even guilty about them, and litigation lawyers thrive on them.

Psychotropics’ package inserts are loaded with side-effect descriptions. The precautions, warnings, and black boxes frequently make patients anxious about taking medications, even when the diseases they suffer from pose far greater risk to their lives and health.

Are side effects entirely bad, or is there a possible upside lurking within them? Psychiatrists know, for example, that a common adverse effect of selective serotonin reuptake inhibitor (SSRI) antidepressants/anxiolytics is delayed orgasm. But because of this side effect, SSRIs can be dramatically helpful for treating premature ejaculation in nondepressed men.

Let’s consider some effects of atypical antipsychotics that usually are considered a downside of antipsychotic treatment yet appear to be associated with advantages for the same patients.

Therapeutic sedation

Sedation was considered a major adverse effect of quetiapine when this antipsychotic was approved for treating schizophrenia. Yet with time—and as the drug received additional indications for bipolar mania and bipolar depression—it became apparent to psychiatrists that quetiapine’s sedative effects could be useful for treating insomnia in individuals with psychotic and mood disorders. Thus, practitioners saw an advantage in using a sedating antipsychotic, instead of adding a sedative/hypnotic, for psychotic or manic patients suffering from agitation, anxiety, and insomnia. Three advantages of this approach for patients are:

• fewer medications to manage
• lower costs without an additional sedative/hypnotic
• avoiding the potential addictive effects of a sedative/hypnotic.

Hyperprolactinemia and myelin repair

Some patients receiving risperidone or paliperidone for psychotic symptoms develop sexual dysfunction because of the side effect of increased serum prolactin. On the other hand, recent research indicates that prolactin enhances the synthesis of oligodendrocytes, which are critical for myelin (white matter) integrity in brain tissue.

In a recent study, researchers used a toxin to destroy patches of white matter in the brains of nonpregnant mice. Subsequent pregnancy and prolactin elevation repaired the myelin lesions completely, whereas no changes occurred in the white matter lesions in the brains of control mice that remained nonpregnant.¹

This suggested myelin-repairing property of prolactin is highly relevant to the development of potential therapies, not only for demyelinating disorders such as multiple sclerosis but also for schizophrenia. Numerous studies have demonstrated that schizophrenia is associated with serious white-matter pathology,² which may account for many of the disorder’s thought, emotional, and cognitive impairments.

Weight gain as marker for antipsychotic efficacy

Some weight increase is observed with all antipsychotics, although certain atypicals such as olanzapine and clozapine are associated with more weight gain than others. No clinician would see anything except a downside to weight gain, which may lead to metabolic complications such as diabetes, hyperlipidemia, and hypertension.

However, some weight gain appears to be related to the efficacy of all antipsychotics (first- and second-generation), according to the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE)³ and previous sporadic observations in the literature. Even Kraepelin noted in the 1920s—long before the era of antipsychotics—that patients with psychosis gained weight when they spontaneously improved.

Thus, some weight gain appears to be a necessary correlate of improvement in persons treated with antipsychotics. The reasons are still unclear, but weight gain may be a modulator, mediator, or marker of antipsychotic efficacy.

In summary, there appears to be at least some upside to the downside of certain drug side effects. In other words, every dark cloud has a silver lining, and it might be helpful for patients and physicians to see more than just the cloud.

Medication side effects are generally regarded as the Achilles’ heel of pharmacologic treatment. And who can argue with that? Adverse effects are the downside of drug treatment of psychiatric disorders and are blamed for tolerability and adherence problems. Patients dread side effects, physicians feel uncomfortable or even guilty about them, and litigation lawyers thrive on them.

Psychotropics’ package inserts are loaded with side-effect descriptions. The precautions, warnings, and black boxes frequently make patients anxious about taking medications, even when the diseases they suffer from pose far greater risk to their lives and health.

Are side effects entirely bad, or is there a possible upside lurking within them? Psychiatrists know, for example, that a common adverse effect of selective serotonin reuptake inhibitor (SSRI) antidepressants/anxiolytics is delayed orgasm. But because of this side effect, SSRIs can be dramatically helpful for treating premature ejaculation in nondepressed men.

Let’s consider some effects of atypical antipsychotics that usually are considered a downside of antipsychotic treatment yet appear to be associated with advantages for the same patients.

Therapeutic sedation

Sedation was considered a major adverse effect of quetiapine when this antipsychotic was approved for treating schizophrenia. Yet with time—and as the drug received additional indications for bipolar mania and bipolar depression—it became apparent to psychiatrists that quetiapine’s sedative effects could be useful for treating insomnia in individuals with psychotic and mood disorders. Thus, practitioners saw an advantage in using a sedating antipsychotic, instead of adding a sedative/hypnotic, for psychotic or manic patients suffering from agitation, anxiety, and insomnia. Three advantages of this approach for patients are:

• fewer medications to manage
• lower costs without an additional sedative/hypnotic
• avoiding the potential addictive effects of a sedative/hypnotic.

Hyperprolactinemia and myelin repair

Some patients receiving risperidone or paliperidone for psychotic symptoms develop sexual dysfunction because of the side effect of increased serum prolactin. On the other hand, recent research indicates that prolactin enhances the synthesis of oligodendrocytes, which are critical for myelin (white matter) integrity in brain tissue.

In a recent study, researchers used a toxin to destroy patches of white matter in the brains of nonpregnant mice. Subsequent pregnancy and prolactin elevation repaired the myelin lesions completely, whereas no changes occurred in the white matter lesions in the brains of control mice that remained nonpregnant.¹

This suggested myelin-repairing property of prolactin is highly relevant to the development of potential therapies, not only for demyelinating disorders such as multiple sclerosis but also for schizophrenia. Numerous studies have demonstrated that schizophrenia is associated with serious white-matter pathology,² which may account for many of the disorder’s thought, emotional, and cognitive impairments.

Weight gain as marker for antipsychotic efficacy

Some weight increase is observed with all antipsychotics, although certain atypicals such as olanzapine and clozapine are associated with more weight gain than others. No clinician would see anything except a downside to weight gain, which may lead to metabolic complications such as diabetes, hyperlipidemia, and hypertension.

However, some weight gain appears to be related to the efficacy of all antipsychotics (first- and second-generation), according to the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE)³ and previous sporadic observations in the literature. Even Kraepelin noted in the 1920s—long before the era of antipsychotics—that patients with psychosis gained weight when they spontaneously improved.

Thus, some weight gain appears to be a necessary correlate of improvement in persons treated with antipsychotics. The reasons are still unclear, but weight gain may be a modulator, mediator, or marker of antipsychotic efficacy.

In summary, there appears to be at least some upside to the downside of certain drug side effects. In other words, every dark cloud has a silver lining, and it might be helpful for patients and physicians to see more than just the cloud.

Medication side effects are generally regarded as the Achilles’ heel of pharmacologic treatment. And who can argue with that? Adverse effects are the downside of drug treatment of psychiatric disorders and are blamed for tolerability and adherence problems. Patients dread side effects, physicians feel uncomfortable or even guilty about them, and litigation lawyers thrive on them.

Psychotropics’ package inserts are loaded with side-effect descriptions. The precautions, warnings, and black boxes frequently make patients anxious about taking medications, even when the diseases they suffer from pose far greater risk to their lives and health.

Are side effects entirely bad, or is there a possible upside lurking within them? Psychiatrists know, for example, that a common adverse effect of selective serotonin reuptake inhibitor (SSRI) antidepressants/anxiolytics is delayed orgasm. But because of this side effect, SSRIs can be dramatically helpful for treating premature ejaculation in nondepressed men.

Let’s consider some effects of atypical antipsychotics that usually are considered a downside of antipsychotic treatment yet appear to be associated with advantages for the same patients.

Therapeutic sedation

Sedation was considered a major adverse effect of quetiapine when this antipsychotic was approved for treating schizophrenia. Yet with time—and as the drug received additional indications for bipolar mania and bipolar depression—it became apparent to psychiatrists that quetiapine’s sedative effects could be useful for treating insomnia in individuals with psychotic and mood disorders. Thus, practitioners saw an advantage in using a sedating antipsychotic, instead of adding a sedative/hypnotic, for psychotic or manic patients suffering from agitation, anxiety, and insomnia. Three advantages of this approach for patients are:

• fewer medications to manage
• lower costs without an additional sedative/hypnotic
• avoiding the potential addictive effects of a sedative/hypnotic.

Hyperprolactinemia and myelin repair

Some patients receiving risperidone or paliperidone for psychotic symptoms develop sexual dysfunction because of the side effect of increased serum prolactin. On the other hand, recent research indicates that prolactin enhances the synthesis of oligodendrocytes, which are critical for myelin (white matter) integrity in brain tissue.

In a recent study, researchers used a toxin to destroy patches of white matter in the brains of nonpregnant mice. Subsequent pregnancy and prolactin elevation repaired the myelin lesions completely, whereas no changes occurred in the white matter lesions in the brains of control mice that remained nonpregnant.¹

This suggested myelin-repairing property of prolactin is highly relevant to the development of potential therapies, not only for demyelinating disorders such as multiple sclerosis but also for schizophrenia. Numerous studies have demonstrated that schizophrenia is associated with serious white-matter pathology,² which may account for many of the disorder’s thought, emotional, and cognitive impairments.

Weight gain as marker for antipsychotic efficacy

Some weight increase is observed with all antipsychotics, although certain atypicals such as olanzapine and clozapine are associated with more weight gain than others. No clinician would see anything except a downside to weight gain, which may lead to metabolic complications such as diabetes, hyperlipidemia, and hypertension.

However, some weight gain appears to be related to the efficacy of all antipsychotics (first- and second-generation), according to the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE)³ and previous sporadic observations in the literature. Even Kraepelin noted in the 1920s—long before the era of antipsychotics—that patients with psychosis gained weight when they spontaneously improved.

Thus, some weight gain appears to be a necessary correlate of improvement in persons treated with antipsychotics. The reasons are still unclear, but weight gain may be a modulator, mediator, or marker of antipsychotic efficacy.

In summary, there appears to be at least some upside to the downside of certain drug side effects. In other words, every dark cloud has a silver lining, and it might be helpful for patients and physicians to see more than just the cloud.

CASE: An unlikable patient

Mr. L, age 56, is admitted to the psychiatric unit at our Veterans Affairs Medical Center for active suicidal ideation; he has a history of self-injurious behaviors that include mutilation and overdose. He also has a history of alcohol dependence and multiple inpatient psychiatric admissions. He has never married and conflicts with his siblings—in whose home he has been staying—have led to frequent homelessness.

His affect is silly and shallow. He also shows signs of haughtiness, disinhibition, grandiosity, and confabulation. For example, he says that while in the Navy he had 82 sexual exploits and developed a drug that cured herpes.

We start Mr. L on divalproex, 1,500 mg/d, and quetiapine, titrated to 200 mg/d. After 3 days he is discharged, but this begins a cycle of repeated suicide gestures and readmissions—9 within the next 3 months. Each time he is discharged, Mr. L fails to follow through on treatment recommendations and is indifferent to our staff’s annoyed reactions.

The author’s observations

Some of our staff members regard Mr. L’s suicidal gestures as manipulative and feel angry and demoralized by his poor adherence to outpatient treatment plans. Their negative countertransference might have impacted how they evaluated Mr. L through repeated admissions and discharges. During Mr. L’s ninth admission, we decide to reevaluate his longitudinal history for clues to his noncompliant behavior.

History: Undocumented injury

Mr. L says he began drinking alcohol at age 16. He reports that he has grown marijuana but has not smoked it since 1991. He denies using heroin or other drugs.

Mr. L states that he suffered a head injury in 1975 after falling off a ladder on a Navy ship. He describes losing consciousness for a brief but uncertain duration. He reports that he has developed a seizure disorder since this fall and a history of amnesia secondary to past seizures. His medical records contain no witnessed seizures. Mr. L also says he was hospitalized a few years ago and placed on a ventilator for 7 days for an undetermined reason.

The authors’ observations

Based on Mr. L’s report of a possible traumatic brain injury (TBI), we order a neurologic evaluation. A year earlier, MRI of the brain without contrast demonstrated minimal, nonspecific periventricular and subcortical, punctuate hyperintensities on flair and T2 weighted sequences that are nonspecific. Overall, the impression was “diffuse involutional changes and mild nonspecific periventricular and subcortical white matter hyperintensities,” which might reflect covert vascular brain injury.

Our frustration over Mr. L’s repeated readmissions for suicidal gestures led us to seek outside evaluation and consultation from a senior psychiatrist for assistance with discharge and treatment planning. Unlike our staff, the consulting psychiatrist did not harbor strong negative feelings toward the patient.

Mr. L’s history of deterioration in psychosocial functioning prompted this psychiatrist to perform a thorough mental status examination that focused on cognitive elements and request formal neuropsychological testing.

Evalutation: Cognitive Deficits

During mental status examination, Mr. L has difficulty recalling 3 items and uses a memory strategy to assist himself. He fails to recollect in reverse order the last 5 U.S. presidents. He spells “world” backward, but has difficulty repeating 6 digits forward and 4 backward. He is unable to do serial 7 subtractions from 93 to 65 correctly. He adequately copies interlocking pentagons and draws a clock with the correct time. He achieves a score of 28/30 on the Folstein Mini Mental State Exam, missing the date by 4 days and recalling 2 of 3 words.

These results suggest Mr. L has difficulty with attention and working memory, short-term memory, fund of general information and long-term memory, and ability to perform simple calculations. Most important, they indicate the need for further study, especially a neuropsychological test battery.

Mr. L’s abnormal neuropsychological test results are summarized in the Table. He manifests concretization of thought. His loss of conceptual fluidity is documented formally by measures of perseverative errors and categories completed on the Wisconsin Card Sorting Test (WCST). These findings support a diagnosis of acquired dementia.

Table

Abnormal findings on Mr. L’s neuropsychological testing

The authors’ observations

Mr. L’s history, cognitive testing, head imaging, and behavioral observations suggest that several pathogenic factors contribute to his impaired functioning. First, he describes a TBI of unknown severity occurring in 1975. Although brain scans did not show evidence of midline shift or encephalomalacia, a direct blow to the head after falling from a height combined with possible post-injury seizures suggests a TBI of at least moderate severity.

Second, Mr. L describes an incident in which he required inpatient respiratory assistance. Although the precipitating medical event was unclear, anoxia or hypoxia is likely. A recent CT revealed low attenuation in the left temporal region that could represent an infarct.

Mr. L’s severe memory impairment and moderate to severe impairment in cognitive flexibility are commonly reported after a TBI of moderate severity. If an ischemic incident were the primary contributor, a lateralized pattern of cognitive dysfunction—which Mr. L does not exhibit—would be expected.

Although Mr. L likely has vascular dementia, his MRI findings do not indicate sufficient disease to account for his memory scores. Vascular dementia is associated with slow, stepwise cognitive deterioration, which is not consistent with severely impaired memory in a 56-year-old patient.

Finally, alcoholism is associated with cognitive difficulty in memory, visuospatial functioning, and abstract reasoning. Mr. L demonstrated significant difficulty in memory and abstract reasoning, but his visuospatial functioning was largely intact. In the absence of Wernicke’s encephalopathy, chronic alcoholics generally do not show memory decrements in line with Mr. L’s. His MRI results indicated only minimal ventricular and sulcal enlargement. Because atrophy is present in approximately 60% of chronic alcoholics, this finding provides evidence of a contribution, but the other contributory factors are associated with more definitive medical outcomes. Thus, alcoholism must be viewed as a secondary contributor to Mr. L’s impaired functioning.

Taking into account all known contributors, TBI emerges as the primary diagnosis.

Consider neurologic injury

Recognizing and characterizing personality changes related to neurologic injury and disease is often problematic and unreliable, even when psychometrically validated instruments and structured diagnostic interviews are used (Box 1).^1-5 Mr. L’s presentation differed from the more commonly reported “impulsive aggression” associated with closed head injury. Sequelae from TBI were contributing to his clinical presentation but was obscured by his shallow and silly affect, inability to accurately assess social cues, and lack of empathy.

Box 1

Mr. L’s case emphasizes the importance of considering brain injury as an etiologic factor in personality changes. It also highlights the complex—and seemingly nonoverlapping—functions and dysfunctions of the frontal lobe, including:

• source memory
  
• working memory
  
• sustained attention
  
• conceptual fluidity
  
• imaginative thinking
  
• impulse regulation
  
• planning and problem-solving skills.
  

Documenting Mr. L’s cognitive deficits and acquired dementia diagnosis changed our staff’s perception of his behavior, enabling us to overcome negative countertransference (Box 2). We no longer regarded him as deliberately manipulative and refer him for appropriate treatment.

Box 2

Outome: Residential placement

We realize Mr. L needs cognitive rehabilitation—including assistance with planning and problem solving—and arrange for his placement in a residential facility for this specialized rehabilitation. Mr. L receives supportive psychotherapy and cognitive remediation from a psychologist. He also is involved in incentive work therapy with a vocational rehabilitation specialist.

Related resource

• Silver, JM, McAllister TW, Yudofsky SC, eds. Textbook of traumatic brain injury. Washington, DC: American Psychiatric Publishing; 2005.
  

• Divalproex • Depakote
  
• Quetiapine • Seroquel
  

The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

CASE: An unlikable patient

Mr. L, age 56, is admitted to the psychiatric unit at our Veterans Affairs Medical Center for active suicidal ideation; he has a history of self-injurious behaviors that include mutilation and overdose. He also has a history of alcohol dependence and multiple inpatient psychiatric admissions. He has never married and conflicts with his siblings—in whose home he has been staying—have led to frequent homelessness.

His affect is silly and shallow. He also shows signs of haughtiness, disinhibition, grandiosity, and confabulation. For example, he says that while in the Navy he had 82 sexual exploits and developed a drug that cured herpes.

We start Mr. L on divalproex, 1,500 mg/d, and quetiapine, titrated to 200 mg/d. After 3 days he is discharged, but this begins a cycle of repeated suicide gestures and readmissions—9 within the next 3 months. Each time he is discharged, Mr. L fails to follow through on treatment recommendations and is indifferent to our staff’s annoyed reactions.

The author’s observations

Some of our staff members regard Mr. L’s suicidal gestures as manipulative and feel angry and demoralized by his poor adherence to outpatient treatment plans. Their negative countertransference might have impacted how they evaluated Mr. L through repeated admissions and discharges. During Mr. L’s ninth admission, we decide to reevaluate his longitudinal history for clues to his noncompliant behavior.

History: Undocumented injury

Mr. L says he began drinking alcohol at age 16. He reports that he has grown marijuana but has not smoked it since 1991. He denies using heroin or other drugs.

Mr. L states that he suffered a head injury in 1975 after falling off a ladder on a Navy ship. He describes losing consciousness for a brief but uncertain duration. He reports that he has developed a seizure disorder since this fall and a history of amnesia secondary to past seizures. His medical records contain no witnessed seizures. Mr. L also says he was hospitalized a few years ago and placed on a ventilator for 7 days for an undetermined reason.

The authors’ observations

Based on Mr. L’s report of a possible traumatic brain injury (TBI), we order a neurologic evaluation. A year earlier, MRI of the brain without contrast demonstrated minimal, nonspecific periventricular and subcortical, punctuate hyperintensities on flair and T2 weighted sequences that are nonspecific. Overall, the impression was “diffuse involutional changes and mild nonspecific periventricular and subcortical white matter hyperintensities,” which might reflect covert vascular brain injury.

Our frustration over Mr. L’s repeated readmissions for suicidal gestures led us to seek outside evaluation and consultation from a senior psychiatrist for assistance with discharge and treatment planning. Unlike our staff, the consulting psychiatrist did not harbor strong negative feelings toward the patient.

Mr. L’s history of deterioration in psychosocial functioning prompted this psychiatrist to perform a thorough mental status examination that focused on cognitive elements and request formal neuropsychological testing.

Evalutation: Cognitive Deficits

During mental status examination, Mr. L has difficulty recalling 3 items and uses a memory strategy to assist himself. He fails to recollect in reverse order the last 5 U.S. presidents. He spells “world” backward, but has difficulty repeating 6 digits forward and 4 backward. He is unable to do serial 7 subtractions from 93 to 65 correctly. He adequately copies interlocking pentagons and draws a clock with the correct time. He achieves a score of 28/30 on the Folstein Mini Mental State Exam, missing the date by 4 days and recalling 2 of 3 words.

These results suggest Mr. L has difficulty with attention and working memory, short-term memory, fund of general information and long-term memory, and ability to perform simple calculations. Most important, they indicate the need for further study, especially a neuropsychological test battery.

Mr. L’s abnormal neuropsychological test results are summarized in the Table. He manifests concretization of thought. His loss of conceptual fluidity is documented formally by measures of perseverative errors and categories completed on the Wisconsin Card Sorting Test (WCST). These findings support a diagnosis of acquired dementia.

Table

Abnormal findings on Mr. L’s neuropsychological testing

The authors’ observations

Mr. L’s history, cognitive testing, head imaging, and behavioral observations suggest that several pathogenic factors contribute to his impaired functioning. First, he describes a TBI of unknown severity occurring in 1975. Although brain scans did not show evidence of midline shift or encephalomalacia, a direct blow to the head after falling from a height combined with possible post-injury seizures suggests a TBI of at least moderate severity.

Second, Mr. L describes an incident in which he required inpatient respiratory assistance. Although the precipitating medical event was unclear, anoxia or hypoxia is likely. A recent CT revealed low attenuation in the left temporal region that could represent an infarct.

Mr. L’s severe memory impairment and moderate to severe impairment in cognitive flexibility are commonly reported after a TBI of moderate severity. If an ischemic incident were the primary contributor, a lateralized pattern of cognitive dysfunction—which Mr. L does not exhibit—would be expected.

Although Mr. L likely has vascular dementia, his MRI findings do not indicate sufficient disease to account for his memory scores. Vascular dementia is associated with slow, stepwise cognitive deterioration, which is not consistent with severely impaired memory in a 56-year-old patient.

Finally, alcoholism is associated with cognitive difficulty in memory, visuospatial functioning, and abstract reasoning. Mr. L demonstrated significant difficulty in memory and abstract reasoning, but his visuospatial functioning was largely intact. In the absence of Wernicke’s encephalopathy, chronic alcoholics generally do not show memory decrements in line with Mr. L’s. His MRI results indicated only minimal ventricular and sulcal enlargement. Because atrophy is present in approximately 60% of chronic alcoholics, this finding provides evidence of a contribution, but the other contributory factors are associated with more definitive medical outcomes. Thus, alcoholism must be viewed as a secondary contributor to Mr. L’s impaired functioning.

Taking into account all known contributors, TBI emerges as the primary diagnosis.

Consider neurologic injury

Recognizing and characterizing personality changes related to neurologic injury and disease is often problematic and unreliable, even when psychometrically validated instruments and structured diagnostic interviews are used (Box 1).^1-5 Mr. L’s presentation differed from the more commonly reported “impulsive aggression” associated with closed head injury. Sequelae from TBI were contributing to his clinical presentation but was obscured by his shallow and silly affect, inability to accurately assess social cues, and lack of empathy.

Box 1

Mr. L’s case emphasizes the importance of considering brain injury as an etiologic factor in personality changes. It also highlights the complex—and seemingly nonoverlapping—functions and dysfunctions of the frontal lobe, including:

• source memory
  
• working memory
  
• sustained attention
  
• conceptual fluidity
  
• imaginative thinking
  
• impulse regulation
  
• planning and problem-solving skills.
  

Documenting Mr. L’s cognitive deficits and acquired dementia diagnosis changed our staff’s perception of his behavior, enabling us to overcome negative countertransference (Box 2). We no longer regarded him as deliberately manipulative and refer him for appropriate treatment.

Box 2

Outome: Residential placement

We realize Mr. L needs cognitive rehabilitation—including assistance with planning and problem solving—and arrange for his placement in a residential facility for this specialized rehabilitation. Mr. L receives supportive psychotherapy and cognitive remediation from a psychologist. He also is involved in incentive work therapy with a vocational rehabilitation specialist.

Related resource

• Silver, JM, McAllister TW, Yudofsky SC, eds. Textbook of traumatic brain injury. Washington, DC: American Psychiatric Publishing; 2005.
  

• Divalproex • Depakote
  
• Quetiapine • Seroquel
  

The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

CASE: An unlikable patient

Mr. L, age 56, is admitted to the psychiatric unit at our Veterans Affairs Medical Center for active suicidal ideation; he has a history of self-injurious behaviors that include mutilation and overdose. He also has a history of alcohol dependence and multiple inpatient psychiatric admissions. He has never married and conflicts with his siblings—in whose home he has been staying—have led to frequent homelessness.

His affect is silly and shallow. He also shows signs of haughtiness, disinhibition, grandiosity, and confabulation. For example, he says that while in the Navy he had 82 sexual exploits and developed a drug that cured herpes.

We start Mr. L on divalproex, 1,500 mg/d, and quetiapine, titrated to 200 mg/d. After 3 days he is discharged, but this begins a cycle of repeated suicide gestures and readmissions—9 within the next 3 months. Each time he is discharged, Mr. L fails to follow through on treatment recommendations and is indifferent to our staff’s annoyed reactions.

The author’s observations

Some of our staff members regard Mr. L’s suicidal gestures as manipulative and feel angry and demoralized by his poor adherence to outpatient treatment plans. Their negative countertransference might have impacted how they evaluated Mr. L through repeated admissions and discharges. During Mr. L’s ninth admission, we decide to reevaluate his longitudinal history for clues to his noncompliant behavior.

History: Undocumented injury

Mr. L says he began drinking alcohol at age 16. He reports that he has grown marijuana but has not smoked it since 1991. He denies using heroin or other drugs.

Mr. L states that he suffered a head injury in 1975 after falling off a ladder on a Navy ship. He describes losing consciousness for a brief but uncertain duration. He reports that he has developed a seizure disorder since this fall and a history of amnesia secondary to past seizures. His medical records contain no witnessed seizures. Mr. L also says he was hospitalized a few years ago and placed on a ventilator for 7 days for an undetermined reason.

The authors’ observations

Based on Mr. L’s report of a possible traumatic brain injury (TBI), we order a neurologic evaluation. A year earlier, MRI of the brain without contrast demonstrated minimal, nonspecific periventricular and subcortical, punctuate hyperintensities on flair and T2 weighted sequences that are nonspecific. Overall, the impression was “diffuse involutional changes and mild nonspecific periventricular and subcortical white matter hyperintensities,” which might reflect covert vascular brain injury.

Our frustration over Mr. L’s repeated readmissions for suicidal gestures led us to seek outside evaluation and consultation from a senior psychiatrist for assistance with discharge and treatment planning. Unlike our staff, the consulting psychiatrist did not harbor strong negative feelings toward the patient.

Mr. L’s history of deterioration in psychosocial functioning prompted this psychiatrist to perform a thorough mental status examination that focused on cognitive elements and request formal neuropsychological testing.

Evalutation: Cognitive Deficits

During mental status examination, Mr. L has difficulty recalling 3 items and uses a memory strategy to assist himself. He fails to recollect in reverse order the last 5 U.S. presidents. He spells “world” backward, but has difficulty repeating 6 digits forward and 4 backward. He is unable to do serial 7 subtractions from 93 to 65 correctly. He adequately copies interlocking pentagons and draws a clock with the correct time. He achieves a score of 28/30 on the Folstein Mini Mental State Exam, missing the date by 4 days and recalling 2 of 3 words.

These results suggest Mr. L has difficulty with attention and working memory, short-term memory, fund of general information and long-term memory, and ability to perform simple calculations. Most important, they indicate the need for further study, especially a neuropsychological test battery.

Mr. L’s abnormal neuropsychological test results are summarized in the Table. He manifests concretization of thought. His loss of conceptual fluidity is documented formally by measures of perseverative errors and categories completed on the Wisconsin Card Sorting Test (WCST). These findings support a diagnosis of acquired dementia.

Table

Abnormal findings on Mr. L’s neuropsychological testing

The authors’ observations

Mr. L’s history, cognitive testing, head imaging, and behavioral observations suggest that several pathogenic factors contribute to his impaired functioning. First, he describes a TBI of unknown severity occurring in 1975. Although brain scans did not show evidence of midline shift or encephalomalacia, a direct blow to the head after falling from a height combined with possible post-injury seizures suggests a TBI of at least moderate severity.

Second, Mr. L describes an incident in which he required inpatient respiratory assistance. Although the precipitating medical event was unclear, anoxia or hypoxia is likely. A recent CT revealed low attenuation in the left temporal region that could represent an infarct.

Mr. L’s severe memory impairment and moderate to severe impairment in cognitive flexibility are commonly reported after a TBI of moderate severity. If an ischemic incident were the primary contributor, a lateralized pattern of cognitive dysfunction—which Mr. L does not exhibit—would be expected.

Although Mr. L likely has vascular dementia, his MRI findings do not indicate sufficient disease to account for his memory scores. Vascular dementia is associated with slow, stepwise cognitive deterioration, which is not consistent with severely impaired memory in a 56-year-old patient.

Finally, alcoholism is associated with cognitive difficulty in memory, visuospatial functioning, and abstract reasoning. Mr. L demonstrated significant difficulty in memory and abstract reasoning, but his visuospatial functioning was largely intact. In the absence of Wernicke’s encephalopathy, chronic alcoholics generally do not show memory decrements in line with Mr. L’s. His MRI results indicated only minimal ventricular and sulcal enlargement. Because atrophy is present in approximately 60% of chronic alcoholics, this finding provides evidence of a contribution, but the other contributory factors are associated with more definitive medical outcomes. Thus, alcoholism must be viewed as a secondary contributor to Mr. L’s impaired functioning.

Taking into account all known contributors, TBI emerges as the primary diagnosis.

Consider neurologic injury

Recognizing and characterizing personality changes related to neurologic injury and disease is often problematic and unreliable, even when psychometrically validated instruments and structured diagnostic interviews are used (Box 1).^1-5 Mr. L’s presentation differed from the more commonly reported “impulsive aggression” associated with closed head injury. Sequelae from TBI were contributing to his clinical presentation but was obscured by his shallow and silly affect, inability to accurately assess social cues, and lack of empathy.

Box 1

Mr. L’s case emphasizes the importance of considering brain injury as an etiologic factor in personality changes. It also highlights the complex—and seemingly nonoverlapping—functions and dysfunctions of the frontal lobe, including:

• source memory
  
• working memory
  
• sustained attention
  
• conceptual fluidity
  
• imaginative thinking
  
• impulse regulation
  
• planning and problem-solving skills.
  

Documenting Mr. L’s cognitive deficits and acquired dementia diagnosis changed our staff’s perception of his behavior, enabling us to overcome negative countertransference (Box 2). We no longer regarded him as deliberately manipulative and refer him for appropriate treatment.

Box 2

Outome: Residential placement

We realize Mr. L needs cognitive rehabilitation—including assistance with planning and problem solving—and arrange for his placement in a residential facility for this specialized rehabilitation. Mr. L receives supportive psychotherapy and cognitive remediation from a psychologist. He also is involved in incentive work therapy with a vocational rehabilitation specialist.

Related resource

• Silver, JM, McAllister TW, Yudofsky SC, eds. Textbook of traumatic brain injury. Washington, DC: American Psychiatric Publishing; 2005.
  

• Divalproex • Depakote
  
• Quetiapine • Seroquel
  

The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Symptoms of perimenopausal depression are not inherently different from those of depression diagnosed at any other time in life, but they present in a unique context:

• Hormonal fluctuations may persist for a long duration.
• Women experiencing hormonal fluctuations may be vulnerable to mood problems.
• Psychosocial/psychodynamic stressors often complicate this life transition.

Managing perimenopausal depression has become more complicated since the Women’s Health Initiative (WHI) studies found fewer benefits and greater risks with hormone replacement therapy (HRT) than had been perceived. This article discusses the clinical presentation of perimenopausal depression, its risk factors, and treatment options in post-WHI psychiatric practice.

Who is at risk?

Perimenopausal depression is diagnosed when onset of major depressive disorder (MDD) is associated with menstrual cycle irregularity and/or somatic symptoms of the menopausal transition.¹ Diagnosis is based on the overall clinical picture, and treatment requires a thoughtful exploration of the complex relationship between hormonal function and mood regulation.

Presentation. For many women, perimenopause is characterized by mild to severe vasomotor, cognitive, and mood symptoms (Table 1). Thus, in your workup of depression in midlife women, document somatic symptoms—such as hot flushes, vaginal dryness, and incontinence—and affective/behavioral symp toms such as mood and sleep disturbances.

Table 1

Vasomotor, cognitive, and mood symptoms of perimenopause

Explore psychiatric and medical histories of your patient and her close relatives. Ask about depression, dysthymia, hypomania, or mood fluctuations around hormonal events such as menses, pregnancy, postpartum, or starting/stopping oral contraceptives. In the differential diagnosis, consider:

• Is low mood temporally connected with hot flushes and disturbed sleep?
• Is low mood secondary to stressful life events?
• Does the patient have another medical illness (such as thyroid disorder) with symptoms similar to depression?
• Is low mood secondary to anxiety or another psychiatric disorder?

Screening. Menopause is considered to have been reached after 12 months of amenorrhea not due to another cause. Median ages for this transition in the United States are 47.5 for perimenopause and 51 for menopause, with an average of 8 years between regular cycles and amenorrhea.² Therefore, begin talking with women about perimenopausal symptoms when they turn 40.

Evidence supports screening perimenopausal women for depressive symptoms even when their primary complaints are vasomotor. The Greene Climacteric Scale³ is convenient for quantifying and monitoring perimenopausal symptoms. It includes depressive symptoms plus physical and cognitive markers. The Quick Inventory of Depressive Symptomatology—Self Report (QIDS-SR)⁴ questionnaire:

• takes minutes to complete
• is easy to score
• quantitates the number and severity of depressive symptoms (see Related Resources).

Psychosocial factors can predict depression at any time in life, but some are specific to the menopausal transition (Table 2).⁵ The “empty nest syndrome,” for example, is often used to explain depressive symptoms in midlife mothers, but no evidence links mood lability with the maturation and departure of children. What may be more stressing for women is supporting adolescents/young adults in their exit to independence while caring for aging parents.

Table 2

Risk factors for depression in women

Sociocultural beliefs about sexuality and menopause may play a role in how your patient experiences and reports her symptoms. In some cultures, menopause elevates a woman’s social status and is associated with increased respect and authority. In others, such as Western societies that emphasize youth and beauty, women may view menopause and its physical changes in a negative light.⁶

Therefore, give careful attention to the psychosocial context of menopause to your patient and the social resources available to her. Questions to ask include:

• Has your lifestyle changed recently?
• Have your husband, family members, or close friends noticed any changes in your functioning?
• Is there anyone in your life that you feel comfortable confiding in?

Explaining the complexity of this life transition may ease her anxiety by normalizing her experience, helping her understand her symptoms, and validating her distress.

What might be the cause?

Although the exact pathophysiology of perimenopausal depression is unknown, hormonal changes,⁷ general health, the experience of menopause,⁸ and the psychosocial context² likely work together to increase vulnerability for depressive symptoms (Figure).

Figure Biopsychosocial milieu of depression during perimenopauseHormonal fluctuation. The estrogen withdrawal theory⁷ explains depressive symptoms as resulting from a sustained decline in ovarian estrogen in tandem with spiking secretions of follicle-stimulating hormone by the pituitary. The finding that women with surgical menopause have a higher incidence of depressive symptoms than women with natural menopause supports this hypothesis.

Mood disorders occur across various female reproductive events, and increased risk appears to be associated with fluctuating gonadal hormones. Thus, declining estrogen may be less causative of perimenopausal depression than extreme fluctuations in estradiol activity.^9,10

Estrogen interacts with dopamine, norepinephrine, beta-endorphin, and serotonin metabolism. In particular, estrogen facilitates serotonin delivery to neurons across the brain. These findings—and the success of selective serotonin reuptake inhibitors (SSRIs) in treating mood disorders—support the theory that fluctuating estrogen affects the serotonergic system and may cause depressive symptoms.

‘Domino theory.’ Others have hypothesized that depressive symptoms are the secondhand result of somatic symptoms of perimenopause. In a “domino effect,” hot flushes and night sweats disrupt women’s sleep, bringing fatigue and impaired daytime concentration, which lead to irritability and feelings of being overwhelmed.⁸

This theory, which incorporates perimenopausal hormone changes, is supported by elevated levels of depression in women who report frequent and intense vasomotor symptoms persisting >27 months.²

The psychosocial theory suggests that depression results from increased stress or adverse events.² Midlife women with depressive symptoms report many possible sources of stress:

• demanding jobs
• family responsibilities
• dual demands of career and family
• little time for self
• poverty or employment stressors
• not enough sleep
• changing social relationships.

Negative interpretations of aging or the menopausal transition also have been implicated in cross-cultural studies.⁶ The predictive nature of psychosocial issues for depression during perimenopause supports this theory.

Evidence-based treatment

HRT. Research and clinical reports suggest that estrogen may have antidepressant effects, either alone or as an adjunct to antidepressant medication.¹¹ Before the WHI studies, expert consensus guidelines on treating depression in women recommended HRT as first-line treatment for patients experiencing a first lifetime onset of mild to moderate depression during perimenopause.¹² WHI findings since 2002 that associated HRT with increased risk of stroke, deep vein thrombosis, and pulmonary embolism—without clear protection against coronary heart disease or cognitive decline—have left HRT a controversial option for treating perimenopausal depression. In the WHI trials:

• 10,739 postmenopausal women age 50 to 79 without a uterus received unopposed conjugated equine estrogens, 0.625 mg/d, or placebo for an average 6.8 years.¹³
• 16,608 postmenopausal women age 50 to 79 with an intact uterus received combination HRT (conjugated equine estrogens, 0.625 mg/d, plus 2.5 mg of medroxyprogesterone), or placebo for an average 5.6 years.¹⁴

The study using combination HRT found increased risks of breast cancer, ischemic stroke, blood clots, and coronary heart disease.¹⁵ A follow-up study showed that vasomotor symptoms returned in more than one-half the women after they stopped using combination HRT.¹⁵

A companion WHI trial found that estrogen, 0.625 mg/d—given unopposed or with a progestin—did not prevent cognitive decline in women age 65 to 79 and may have been associated with a slightly greater risk of probable dementia.^16,17

The FDA recommends that women who want to use HRT to control menopausal symptoms use the lowest effective dose for the shortest time necessary.¹⁸

Antidepressants. SSRIs may be more useful than estrogen for producing MDD remission in perimenopausal women.¹⁹ SSRIs and other psychotropics may reduce perimenopausal vasomotor symptoms in addition to addressing depressive symptoms (Table 3). When choosing antidepressant therapy, consider the patient’s dominant presenting perimenopausal symptoms and side effects associated with treatment.²⁰

Table 3

Nonhormone medications for perimenopausal depression: Evidence-based dosages and target symptoms

Nonpharmacologic interventions are viable options for women who are reluctant to begin HRT or psychotropics.

Psychotherapy. Interpersonal psychotherapy (IPT) and cognitive-behavioral therapy (CBT) have been recommended to address psychosocial elements of perimenopausal mood lability.²¹ For women with climacteric depression, IPT focuses on role transitions, loss, and interpersonal support, whereas CBT focuses on identifying and altering negative thoughts and beliefs.

Although no randomized trials have examined psychotherapies for perimenopausal depression, a pilot open trial provided group CBT—psychoeducation, group discussion, and coping skills training—to 30 women with climacteric symptoms. Anxiety, depression, partnership relations, overall sexuality, hot flushes, and cardiac complaints improved significantly, based on pre- and post-intervention surveys. Sexual satisfaction and the stressfulness of menopausal symptoms did not change.²²

Integrative medicine. Plant-based substances and herbal remedies such as phytoestrogens, red-clover isoflavones, black cohosh, and evening primrose oil have been included in a few research investigations, and the evidence is equivocal. Because of potential interactions between alternative therapies and medications, inquire about their use. Although a comprehensive review of integrative medicine for perimenopausal symptoms is beyond the scope of this article, see suggested readings (Box).

Box

Clinical recommendations

Explore options with your patient; discuss side effects, risks, and expected minimum duration of treatment. Antidepressants, hormonal therapies, psychotherapy, and complementary and alternative treatments each might have a role in managing perimenopausal depression. A patient’s preferences, psychiatric history, and depression severity help determine which options to consider and in what order. How she responded to past treatments also can help you individualize a plan.

HRT may be appropriate for women who express a preference for HRT, have responded well to past hormone therapy, and have no personal history or high-risk factors for breast cancer. Based on the WHI findings, we consider a history of breast cancer in the patient or a first- or second-degree relative a contraindication to HRT.

Estrogen can be used alone or with an antidepressant. Studies support 17β-estradiol, 0.1 to 0.3 mg/d, for 8 to 12 weeks.^11,23 Concomitant progesterone may be indicated to offset the effects of unopposed estrogen in women with an intact uterus. This option calls for an informed discussion with the patient about risks and benefits.

No data support long-term use of estrogen for recurrent or chronic depression. Because HRT’s risks and benefits vary with the length of exposure, individualize the extended use of estrogen solely to augment treatment for depression. Because vasomotor symptoms may recur when HRT is discontinued,¹⁵ we recommend that women make an informed decision in consultation with a gynecologist or primary care physician.

Antidepressants that have serotonergic activity—such as SSRIs and serotonin-norepinephrine reuptake inhibitors (SNRIs)—appear most promising for treating comorbid depressive and vasomotor symptoms. If a patient has had a good response to an antidepressant in the past, consider starting with that medication.

Common antidepressant side effects are difficult to assess in perimenopausal patients with MDD because the symptoms attributed to antidepressant side effects—such as low libido, sleep disturbance, and weight changes—also can be caused by mood disorders and hormonal changes. Therefore, inquire about these symptoms when you initiate antidepressant therapy and at follow-up assessments.

Psychotherapy. We recommend that all women who present with perimenopausal depression receive information about psychotherapy. Psychotherapy alone often is adequate for mild depression, and adding psychotherapy to antidepressant treatment usually enhances recovery from moderate and severe depression episodes. In addition, patients who engage in psychotherapy for depression may have a lower rate of relapse.²⁴

Individual psychotherapy can help patients with perimenopausal depression:

• accept this life transition
• recognize the benefits of menopause, such as no need for contraception
• develop awareness of personal potential in the years ahead.

Because depression often occurs in an interpersonal context, consider including family members in psychotherapy to improve the patient’s interpersonal support.

Integrative therapies. A full evaluation and consideration of standard treatment options is indicated for all women with MDD. Integrative medicine appeals to many patients but has not been sufficiently studied for perimenopausal depression. Supplemental omega-3 fatty acids and folate are reasonable adjuncts to the treatment of MDD^25-27 and deserve study in perimenopausal MDD.

Related resources

• Greene Climacteric Scale. www.menopausematters.co.uk/greenescale.php.
• Quick Inventory of Depressive Symptomatology. www.idsqids.org.
• National Center for Complementary and Alternative Medicine. National Institutes of Health. http://nccam.nih.gov.

Drug brand names

• Citalopram • Celexa
• Duloxetine • Cymbalta
• Escitalopram • Lexapro
• Estradiol • various
• Eszopiclone • Lunesta
• Fluoxetine • Prozac
• Gabapentin • Neurontin
• Medroxyprogesterone • Provera
• Mirtazapine • Remeron
• Paroxetine • Paxil
• Sertraline • Zoloft
• Venlafaxine • Effexor
• Zolpidem • Ambien

Disclosures

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

Dr. Freeman receives research support from GlaxoSmithKline, Forest Pharmaceuticals, and Eli Lilly and Company. She was associate professor of psychiatry at the University of Texas Southwestern Medical Center at Dallas when this article was written and is now on the faculty at Harvard Medical School and Massachusetts General Hospital, Boston.

Symptoms of perimenopausal depression are not inherently different from those of depression diagnosed at any other time in life, but they present in a unique context:

• Hormonal fluctuations may persist for a long duration.
• Women experiencing hormonal fluctuations may be vulnerable to mood problems.
• Psychosocial/psychodynamic stressors often complicate this life transition.

Managing perimenopausal depression has become more complicated since the Women’s Health Initiative (WHI) studies found fewer benefits and greater risks with hormone replacement therapy (HRT) than had been perceived. This article discusses the clinical presentation of perimenopausal depression, its risk factors, and treatment options in post-WHI psychiatric practice.

Who is at risk?

Perimenopausal depression is diagnosed when onset of major depressive disorder (MDD) is associated with menstrual cycle irregularity and/or somatic symptoms of the menopausal transition.¹ Diagnosis is based on the overall clinical picture, and treatment requires a thoughtful exploration of the complex relationship between hormonal function and mood regulation.

Presentation. For many women, perimenopause is characterized by mild to severe vasomotor, cognitive, and mood symptoms (Table 1). Thus, in your workup of depression in midlife women, document somatic symptoms—such as hot flushes, vaginal dryness, and incontinence—and affective/behavioral symp toms such as mood and sleep disturbances.

Table 1

Vasomotor, cognitive, and mood symptoms of perimenopause

Explore psychiatric and medical histories of your patient and her close relatives. Ask about depression, dysthymia, hypomania, or mood fluctuations around hormonal events such as menses, pregnancy, postpartum, or starting/stopping oral contraceptives. In the differential diagnosis, consider:

• Is low mood temporally connected with hot flushes and disturbed sleep?
• Is low mood secondary to stressful life events?
• Does the patient have another medical illness (such as thyroid disorder) with symptoms similar to depression?
• Is low mood secondary to anxiety or another psychiatric disorder?

Screening. Menopause is considered to have been reached after 12 months of amenorrhea not due to another cause. Median ages for this transition in the United States are 47.5 for perimenopause and 51 for menopause, with an average of 8 years between regular cycles and amenorrhea.² Therefore, begin talking with women about perimenopausal symptoms when they turn 40.

Evidence supports screening perimenopausal women for depressive symptoms even when their primary complaints are vasomotor. The Greene Climacteric Scale³ is convenient for quantifying and monitoring perimenopausal symptoms. It includes depressive symptoms plus physical and cognitive markers. The Quick Inventory of Depressive Symptomatology—Self Report (QIDS-SR)⁴ questionnaire:

• takes minutes to complete
• is easy to score
• quantitates the number and severity of depressive symptoms (see Related Resources).

Psychosocial factors can predict depression at any time in life, but some are specific to the menopausal transition (Table 2).⁵ The “empty nest syndrome,” for example, is often used to explain depressive symptoms in midlife mothers, but no evidence links mood lability with the maturation and departure of children. What may be more stressing for women is supporting adolescents/young adults in their exit to independence while caring for aging parents.

Table 2

Risk factors for depression in women

Sociocultural beliefs about sexuality and menopause may play a role in how your patient experiences and reports her symptoms. In some cultures, menopause elevates a woman’s social status and is associated with increased respect and authority. In others, such as Western societies that emphasize youth and beauty, women may view menopause and its physical changes in a negative light.⁶

Therefore, give careful attention to the psychosocial context of menopause to your patient and the social resources available to her. Questions to ask include:

• Has your lifestyle changed recently?
• Have your husband, family members, or close friends noticed any changes in your functioning?
• Is there anyone in your life that you feel comfortable confiding in?

Explaining the complexity of this life transition may ease her anxiety by normalizing her experience, helping her understand her symptoms, and validating her distress.

What might be the cause?

Although the exact pathophysiology of perimenopausal depression is unknown, hormonal changes,⁷ general health, the experience of menopause,⁸ and the psychosocial context² likely work together to increase vulnerability for depressive symptoms (Figure).

Figure Biopsychosocial milieu of depression during perimenopauseHormonal fluctuation. The estrogen withdrawal theory⁷ explains depressive symptoms as resulting from a sustained decline in ovarian estrogen in tandem with spiking secretions of follicle-stimulating hormone by the pituitary. The finding that women with surgical menopause have a higher incidence of depressive symptoms than women with natural menopause supports this hypothesis.

Mood disorders occur across various female reproductive events, and increased risk appears to be associated with fluctuating gonadal hormones. Thus, declining estrogen may be less causative of perimenopausal depression than extreme fluctuations in estradiol activity.^9,10

Estrogen interacts with dopamine, norepinephrine, beta-endorphin, and serotonin metabolism. In particular, estrogen facilitates serotonin delivery to neurons across the brain. These findings—and the success of selective serotonin reuptake inhibitors (SSRIs) in treating mood disorders—support the theory that fluctuating estrogen affects the serotonergic system and may cause depressive symptoms.

‘Domino theory.’ Others have hypothesized that depressive symptoms are the secondhand result of somatic symptoms of perimenopause. In a “domino effect,” hot flushes and night sweats disrupt women’s sleep, bringing fatigue and impaired daytime concentration, which lead to irritability and feelings of being overwhelmed.⁸

This theory, which incorporates perimenopausal hormone changes, is supported by elevated levels of depression in women who report frequent and intense vasomotor symptoms persisting >27 months.²

The psychosocial theory suggests that depression results from increased stress or adverse events.² Midlife women with depressive symptoms report many possible sources of stress:

• demanding jobs
• family responsibilities
• dual demands of career and family
• little time for self
• poverty or employment stressors
• not enough sleep
• changing social relationships.

Negative interpretations of aging or the menopausal transition also have been implicated in cross-cultural studies.⁶ The predictive nature of psychosocial issues for depression during perimenopause supports this theory.

Evidence-based treatment

HRT. Research and clinical reports suggest that estrogen may have antidepressant effects, either alone or as an adjunct to antidepressant medication.¹¹ Before the WHI studies, expert consensus guidelines on treating depression in women recommended HRT as first-line treatment for patients experiencing a first lifetime onset of mild to moderate depression during perimenopause.¹² WHI findings since 2002 that associated HRT with increased risk of stroke, deep vein thrombosis, and pulmonary embolism—without clear protection against coronary heart disease or cognitive decline—have left HRT a controversial option for treating perimenopausal depression. In the WHI trials:

• 10,739 postmenopausal women age 50 to 79 without a uterus received unopposed conjugated equine estrogens, 0.625 mg/d, or placebo for an average 6.8 years.¹³
• 16,608 postmenopausal women age 50 to 79 with an intact uterus received combination HRT (conjugated equine estrogens, 0.625 mg/d, plus 2.5 mg of medroxyprogesterone), or placebo for an average 5.6 years.¹⁴

The study using combination HRT found increased risks of breast cancer, ischemic stroke, blood clots, and coronary heart disease.¹⁵ A follow-up study showed that vasomotor symptoms returned in more than one-half the women after they stopped using combination HRT.¹⁵

A companion WHI trial found that estrogen, 0.625 mg/d—given unopposed or with a progestin—did not prevent cognitive decline in women age 65 to 79 and may have been associated with a slightly greater risk of probable dementia.^16,17

The FDA recommends that women who want to use HRT to control menopausal symptoms use the lowest effective dose for the shortest time necessary.¹⁸

Antidepressants. SSRIs may be more useful than estrogen for producing MDD remission in perimenopausal women.¹⁹ SSRIs and other psychotropics may reduce perimenopausal vasomotor symptoms in addition to addressing depressive symptoms (Table 3). When choosing antidepressant therapy, consider the patient’s dominant presenting perimenopausal symptoms and side effects associated with treatment.²⁰

Table 3

Nonhormone medications for perimenopausal depression: Evidence-based dosages and target symptoms

Nonpharmacologic interventions are viable options for women who are reluctant to begin HRT or psychotropics.

Psychotherapy. Interpersonal psychotherapy (IPT) and cognitive-behavioral therapy (CBT) have been recommended to address psychosocial elements of perimenopausal mood lability.²¹ For women with climacteric depression, IPT focuses on role transitions, loss, and interpersonal support, whereas CBT focuses on identifying and altering negative thoughts and beliefs.

Although no randomized trials have examined psychotherapies for perimenopausal depression, a pilot open trial provided group CBT—psychoeducation, group discussion, and coping skills training—to 30 women with climacteric symptoms. Anxiety, depression, partnership relations, overall sexuality, hot flushes, and cardiac complaints improved significantly, based on pre- and post-intervention surveys. Sexual satisfaction and the stressfulness of menopausal symptoms did not change.²²

Integrative medicine. Plant-based substances and herbal remedies such as phytoestrogens, red-clover isoflavones, black cohosh, and evening primrose oil have been included in a few research investigations, and the evidence is equivocal. Because of potential interactions between alternative therapies and medications, inquire about their use. Although a comprehensive review of integrative medicine for perimenopausal symptoms is beyond the scope of this article, see suggested readings (Box).

Box

Clinical recommendations

Explore options with your patient; discuss side effects, risks, and expected minimum duration of treatment. Antidepressants, hormonal therapies, psychotherapy, and complementary and alternative treatments each might have a role in managing perimenopausal depression. A patient’s preferences, psychiatric history, and depression severity help determine which options to consider and in what order. How she responded to past treatments also can help you individualize a plan.

HRT may be appropriate for women who express a preference for HRT, have responded well to past hormone therapy, and have no personal history or high-risk factors for breast cancer. Based on the WHI findings, we consider a history of breast cancer in the patient or a first- or second-degree relative a contraindication to HRT.

Estrogen can be used alone or with an antidepressant. Studies support 17β-estradiol, 0.1 to 0.3 mg/d, for 8 to 12 weeks.^11,23 Concomitant progesterone may be indicated to offset the effects of unopposed estrogen in women with an intact uterus. This option calls for an informed discussion with the patient about risks and benefits.

No data support long-term use of estrogen for recurrent or chronic depression. Because HRT’s risks and benefits vary with the length of exposure, individualize the extended use of estrogen solely to augment treatment for depression. Because vasomotor symptoms may recur when HRT is discontinued,¹⁵ we recommend that women make an informed decision in consultation with a gynecologist or primary care physician.

Antidepressants that have serotonergic activity—such as SSRIs and serotonin-norepinephrine reuptake inhibitors (SNRIs)—appear most promising for treating comorbid depressive and vasomotor symptoms. If a patient has had a good response to an antidepressant in the past, consider starting with that medication.

Common antidepressant side effects are difficult to assess in perimenopausal patients with MDD because the symptoms attributed to antidepressant side effects—such as low libido, sleep disturbance, and weight changes—also can be caused by mood disorders and hormonal changes. Therefore, inquire about these symptoms when you initiate antidepressant therapy and at follow-up assessments.

Psychotherapy. We recommend that all women who present with perimenopausal depression receive information about psychotherapy. Psychotherapy alone often is adequate for mild depression, and adding psychotherapy to antidepressant treatment usually enhances recovery from moderate and severe depression episodes. In addition, patients who engage in psychotherapy for depression may have a lower rate of relapse.²⁴

Individual psychotherapy can help patients with perimenopausal depression:

• accept this life transition
• recognize the benefits of menopause, such as no need for contraception
• develop awareness of personal potential in the years ahead.

Because depression often occurs in an interpersonal context, consider including family members in psychotherapy to improve the patient’s interpersonal support.

Integrative therapies. A full evaluation and consideration of standard treatment options is indicated for all women with MDD. Integrative medicine appeals to many patients but has not been sufficiently studied for perimenopausal depression. Supplemental omega-3 fatty acids and folate are reasonable adjuncts to the treatment of MDD^25-27 and deserve study in perimenopausal MDD.

Related resources

• Greene Climacteric Scale. www.menopausematters.co.uk/greenescale.php.
• Quick Inventory of Depressive Symptomatology. www.idsqids.org.
• National Center for Complementary and Alternative Medicine. National Institutes of Health. http://nccam.nih.gov.

Drug brand names

• Citalopram • Celexa
• Duloxetine • Cymbalta
• Escitalopram • Lexapro
• Estradiol • various
• Eszopiclone • Lunesta
• Fluoxetine • Prozac
• Gabapentin • Neurontin
• Medroxyprogesterone • Provera
• Mirtazapine • Remeron
• Paroxetine • Paxil
• Sertraline • Zoloft
• Venlafaxine • Effexor
• Zolpidem • Ambien

Disclosures

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

Dr. Freeman receives research support from GlaxoSmithKline, Forest Pharmaceuticals, and Eli Lilly and Company. She was associate professor of psychiatry at the University of Texas Southwestern Medical Center at Dallas when this article was written and is now on the faculty at Harvard Medical School and Massachusetts General Hospital, Boston.

Symptoms of perimenopausal depression are not inherently different from those of depression diagnosed at any other time in life, but they present in a unique context:

• Hormonal fluctuations may persist for a long duration.
• Women experiencing hormonal fluctuations may be vulnerable to mood problems.
• Psychosocial/psychodynamic stressors often complicate this life transition.

Managing perimenopausal depression has become more complicated since the Women’s Health Initiative (WHI) studies found fewer benefits and greater risks with hormone replacement therapy (HRT) than had been perceived. This article discusses the clinical presentation of perimenopausal depression, its risk factors, and treatment options in post-WHI psychiatric practice.

Who is at risk?

Perimenopausal depression is diagnosed when onset of major depressive disorder (MDD) is associated with menstrual cycle irregularity and/or somatic symptoms of the menopausal transition.¹ Diagnosis is based on the overall clinical picture, and treatment requires a thoughtful exploration of the complex relationship between hormonal function and mood regulation.

Presentation. For many women, perimenopause is characterized by mild to severe vasomotor, cognitive, and mood symptoms (Table 1). Thus, in your workup of depression in midlife women, document somatic symptoms—such as hot flushes, vaginal dryness, and incontinence—and affective/behavioral symp toms such as mood and sleep disturbances.

Table 1

Vasomotor, cognitive, and mood symptoms of perimenopause

Explore psychiatric and medical histories of your patient and her close relatives. Ask about depression, dysthymia, hypomania, or mood fluctuations around hormonal events such as menses, pregnancy, postpartum, or starting/stopping oral contraceptives. In the differential diagnosis, consider:

• Is low mood temporally connected with hot flushes and disturbed sleep?
• Is low mood secondary to stressful life events?
• Does the patient have another medical illness (such as thyroid disorder) with symptoms similar to depression?
• Is low mood secondary to anxiety or another psychiatric disorder?

Screening. Menopause is considered to have been reached after 12 months of amenorrhea not due to another cause. Median ages for this transition in the United States are 47.5 for perimenopause and 51 for menopause, with an average of 8 years between regular cycles and amenorrhea.² Therefore, begin talking with women about perimenopausal symptoms when they turn 40.

Evidence supports screening perimenopausal women for depressive symptoms even when their primary complaints are vasomotor. The Greene Climacteric Scale³ is convenient for quantifying and monitoring perimenopausal symptoms. It includes depressive symptoms plus physical and cognitive markers. The Quick Inventory of Depressive Symptomatology—Self Report (QIDS-SR)⁴ questionnaire:

• takes minutes to complete
• is easy to score
• quantitates the number and severity of depressive symptoms (see Related Resources).

Psychosocial factors can predict depression at any time in life, but some are specific to the menopausal transition (Table 2).⁵ The “empty nest syndrome,” for example, is often used to explain depressive symptoms in midlife mothers, but no evidence links mood lability with the maturation and departure of children. What may be more stressing for women is supporting adolescents/young adults in their exit to independence while caring for aging parents.

Table 2

Risk factors for depression in women

Sociocultural beliefs about sexuality and menopause may play a role in how your patient experiences and reports her symptoms. In some cultures, menopause elevates a woman’s social status and is associated with increased respect and authority. In others, such as Western societies that emphasize youth and beauty, women may view menopause and its physical changes in a negative light.⁶

Therefore, give careful attention to the psychosocial context of menopause to your patient and the social resources available to her. Questions to ask include:

• Has your lifestyle changed recently?
• Have your husband, family members, or close friends noticed any changes in your functioning?
• Is there anyone in your life that you feel comfortable confiding in?

Explaining the complexity of this life transition may ease her anxiety by normalizing her experience, helping her understand her symptoms, and validating her distress.

What might be the cause?

Although the exact pathophysiology of perimenopausal depression is unknown, hormonal changes,⁷ general health, the experience of menopause,⁸ and the psychosocial context² likely work together to increase vulnerability for depressive symptoms (Figure).

Figure Biopsychosocial milieu of depression during perimenopauseHormonal fluctuation. The estrogen withdrawal theory⁷ explains depressive symptoms as resulting from a sustained decline in ovarian estrogen in tandem with spiking secretions of follicle-stimulating hormone by the pituitary. The finding that women with surgical menopause have a higher incidence of depressive symptoms than women with natural menopause supports this hypothesis.

Mood disorders occur across various female reproductive events, and increased risk appears to be associated with fluctuating gonadal hormones. Thus, declining estrogen may be less causative of perimenopausal depression than extreme fluctuations in estradiol activity.^9,10

Estrogen interacts with dopamine, norepinephrine, beta-endorphin, and serotonin metabolism. In particular, estrogen facilitates serotonin delivery to neurons across the brain. These findings—and the success of selective serotonin reuptake inhibitors (SSRIs) in treating mood disorders—support the theory that fluctuating estrogen affects the serotonergic system and may cause depressive symptoms.

‘Domino theory.’ Others have hypothesized that depressive symptoms are the secondhand result of somatic symptoms of perimenopause. In a “domino effect,” hot flushes and night sweats disrupt women’s sleep, bringing fatigue and impaired daytime concentration, which lead to irritability and feelings of being overwhelmed.⁸

This theory, which incorporates perimenopausal hormone changes, is supported by elevated levels of depression in women who report frequent and intense vasomotor symptoms persisting >27 months.²

The psychosocial theory suggests that depression results from increased stress or adverse events.² Midlife women with depressive symptoms report many possible sources of stress:

• demanding jobs
• family responsibilities
• dual demands of career and family
• little time for self
• poverty or employment stressors
• not enough sleep
• changing social relationships.

Negative interpretations of aging or the menopausal transition also have been implicated in cross-cultural studies.⁶ The predictive nature of psychosocial issues for depression during perimenopause supports this theory.

Evidence-based treatment

HRT. Research and clinical reports suggest that estrogen may have antidepressant effects, either alone or as an adjunct to antidepressant medication.¹¹ Before the WHI studies, expert consensus guidelines on treating depression in women recommended HRT as first-line treatment for patients experiencing a first lifetime onset of mild to moderate depression during perimenopause.¹² WHI findings since 2002 that associated HRT with increased risk of stroke, deep vein thrombosis, and pulmonary embolism—without clear protection against coronary heart disease or cognitive decline—have left HRT a controversial option for treating perimenopausal depression. In the WHI trials:

• 10,739 postmenopausal women age 50 to 79 without a uterus received unopposed conjugated equine estrogens, 0.625 mg/d, or placebo for an average 6.8 years.¹³
• 16,608 postmenopausal women age 50 to 79 with an intact uterus received combination HRT (conjugated equine estrogens, 0.625 mg/d, plus 2.5 mg of medroxyprogesterone), or placebo for an average 5.6 years.¹⁴

The study using combination HRT found increased risks of breast cancer, ischemic stroke, blood clots, and coronary heart disease.¹⁵ A follow-up study showed that vasomotor symptoms returned in more than one-half the women after they stopped using combination HRT.¹⁵

A companion WHI trial found that estrogen, 0.625 mg/d—given unopposed or with a progestin—did not prevent cognitive decline in women age 65 to 79 and may have been associated with a slightly greater risk of probable dementia.^16,17

The FDA recommends that women who want to use HRT to control menopausal symptoms use the lowest effective dose for the shortest time necessary.¹⁸

Antidepressants. SSRIs may be more useful than estrogen for producing MDD remission in perimenopausal women.¹⁹ SSRIs and other psychotropics may reduce perimenopausal vasomotor symptoms in addition to addressing depressive symptoms (Table 3). When choosing antidepressant therapy, consider the patient’s dominant presenting perimenopausal symptoms and side effects associated with treatment.²⁰

Table 3

Nonhormone medications for perimenopausal depression: Evidence-based dosages and target symptoms

Nonpharmacologic interventions are viable options for women who are reluctant to begin HRT or psychotropics.

Psychotherapy. Interpersonal psychotherapy (IPT) and cognitive-behavioral therapy (CBT) have been recommended to address psychosocial elements of perimenopausal mood lability.²¹ For women with climacteric depression, IPT focuses on role transitions, loss, and interpersonal support, whereas CBT focuses on identifying and altering negative thoughts and beliefs.

Although no randomized trials have examined psychotherapies for perimenopausal depression, a pilot open trial provided group CBT—psychoeducation, group discussion, and coping skills training—to 30 women with climacteric symptoms. Anxiety, depression, partnership relations, overall sexuality, hot flushes, and cardiac complaints improved significantly, based on pre- and post-intervention surveys. Sexual satisfaction and the stressfulness of menopausal symptoms did not change.²²

Integrative medicine. Plant-based substances and herbal remedies such as phytoestrogens, red-clover isoflavones, black cohosh, and evening primrose oil have been included in a few research investigations, and the evidence is equivocal. Because of potential interactions between alternative therapies and medications, inquire about their use. Although a comprehensive review of integrative medicine for perimenopausal symptoms is beyond the scope of this article, see suggested readings (Box).

Box

Clinical recommendations

Explore options with your patient; discuss side effects, risks, and expected minimum duration of treatment. Antidepressants, hormonal therapies, psychotherapy, and complementary and alternative treatments each might have a role in managing perimenopausal depression. A patient’s preferences, psychiatric history, and depression severity help determine which options to consider and in what order. How she responded to past treatments also can help you individualize a plan.

HRT may be appropriate for women who express a preference for HRT, have responded well to past hormone therapy, and have no personal history or high-risk factors for breast cancer. Based on the WHI findings, we consider a history of breast cancer in the patient or a first- or second-degree relative a contraindication to HRT.

Estrogen can be used alone or with an antidepressant. Studies support 17β-estradiol, 0.1 to 0.3 mg/d, for 8 to 12 weeks.^11,23 Concomitant progesterone may be indicated to offset the effects of unopposed estrogen in women with an intact uterus. This option calls for an informed discussion with the patient about risks and benefits.

No data support long-term use of estrogen for recurrent or chronic depression. Because HRT’s risks and benefits vary with the length of exposure, individualize the extended use of estrogen solely to augment treatment for depression. Because vasomotor symptoms may recur when HRT is discontinued,¹⁵ we recommend that women make an informed decision in consultation with a gynecologist or primary care physician.

Antidepressants that have serotonergic activity—such as SSRIs and serotonin-norepinephrine reuptake inhibitors (SNRIs)—appear most promising for treating comorbid depressive and vasomotor symptoms. If a patient has had a good response to an antidepressant in the past, consider starting with that medication.

Common antidepressant side effects are difficult to assess in perimenopausal patients with MDD because the symptoms attributed to antidepressant side effects—such as low libido, sleep disturbance, and weight changes—also can be caused by mood disorders and hormonal changes. Therefore, inquire about these symptoms when you initiate antidepressant therapy and at follow-up assessments.

Psychotherapy. We recommend that all women who present with perimenopausal depression receive information about psychotherapy. Psychotherapy alone often is adequate for mild depression, and adding psychotherapy to antidepressant treatment usually enhances recovery from moderate and severe depression episodes. In addition, patients who engage in psychotherapy for depression may have a lower rate of relapse.²⁴

Individual psychotherapy can help patients with perimenopausal depression:

• accept this life transition
• recognize the benefits of menopause, such as no need for contraception
• develop awareness of personal potential in the years ahead.

Because depression often occurs in an interpersonal context, consider including family members in psychotherapy to improve the patient’s interpersonal support.

Integrative therapies. A full evaluation and consideration of standard treatment options is indicated for all women with MDD. Integrative medicine appeals to many patients but has not been sufficiently studied for perimenopausal depression. Supplemental omega-3 fatty acids and folate are reasonable adjuncts to the treatment of MDD^25-27 and deserve study in perimenopausal MDD.

Related resources

• Greene Climacteric Scale. www.menopausematters.co.uk/greenescale.php.
• Quick Inventory of Depressive Symptomatology. www.idsqids.org.
• National Center for Complementary and Alternative Medicine. National Institutes of Health. http://nccam.nih.gov.

Drug brand names

• Citalopram • Celexa
• Duloxetine • Cymbalta
• Escitalopram • Lexapro
• Estradiol • various
• Eszopiclone • Lunesta
• Fluoxetine • Prozac
• Gabapentin • Neurontin
• Medroxyprogesterone • Provera
• Mirtazapine • Remeron
• Paroxetine • Paxil
• Sertraline • Zoloft
• Venlafaxine • Effexor
• Zolpidem • Ambien

Disclosures

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

Dr. Freeman receives research support from GlaxoSmithKline, Forest Pharmaceuticals, and Eli Lilly and Company. She was associate professor of psychiatry at the University of Texas Southwestern Medical Center at Dallas when this article was written and is now on the faculty at Harvard Medical School and Massachusetts General Hospital, Boston.

From SIG: E CAPS to CAGE and WWHHHHIMPS, mnemonics help practitioners and trainees recall important lists (such as criteria for depression, screening questions for alcoholism, or life-threatening causes of delirium, respectively). Mnemonics’ efficacy rests on the principle that grouped information is easier to remember than individual points of data.

Not everyone loves mnemonics, but recollecting diagnostic criteria is useful in clinical practice and research, on board examinations, and for insurance reimbursement. Thus, tools that assist in recalling diagnostic criteria have a role in psychiatric practice and teaching.

In this article, we present 32 mnemonics to help clinicians diagnose:

• affective disorders (Box 1)^1,2
• anxiety disorders (Box 2)^3-6
• medication adverse effects (Box 3)^7,8
• personality disorders (Box 4)^9-11
• addiction disorders (Box 5)^12,13
• causes of delirium (Box 6).¹⁴

We also discuss how mnemonics improve one’s memory, based on the principles of learning theory.

How mnemonics work

A mnemonic—from the Greek word “mnemonikos” (“of memory”)—links new data with previously learned information. Mnemonics assist in learning by reducing the amount of information (“cognitive load”) that needs to be stored for long-term processing and retrieval.¹⁵

Memory, defined as the “persistence of learning in a state that can be revealed at a later time,”¹⁶ can be divided into 2 types:

• declarative (a conscious recollection of facts, such as remembering a relative’s birthday)
• procedural (skills-based learning, such as riding a bicycle).

Declarative memory has a conscious component and may be mediated by the medial temporal lobe and cortical association structures. Procedural memory has less of a conscious component; it may involve the basal ganglia, cerebellum, and a variety of cortical sensory-perceptive regions.¹⁷

BOX 1.

Declarative memory can be subdivided into working memory and long-term memory.

With working memory, new items of information are held briefly so that encoding and eventual storage can take place.

Working memory guides decision-making and future planning and is intricately related to attention.^18-21 Functional MRI and positron emission tomography as well as neurocognitive testing have shown that working memory tasks activate the prefrontal cortex and brain regions specific to language and visuospatial memory.

The hippocampus is thought to rapidly absorb new information, and this data is consolidated and permanently stored via the prefrontal cortex.^22-26 Given the hippocampus’ limited storage capacity, new information (such as what you ate for breakfast 3 weeks ago) will disappear if it is not repeated regularly.¹⁷

Long-term memory, on the other hand, is encoded knowledge that is linked to facts learned in the past; it is consolidated in the brain and can be readily retrieved. Neuroimaging studies have demonstrated opposing patterns of activation in the hippocampus and prefrontal cortex, depending on whether the memory being recalled is:

• new (high hippocampal activity, low prefrontal cortex activity)
• old (low hippocampal activity, high prefrontal cortex activity).²⁷

Mnemonics are thought to affect working memory by reducing the introduced cognitive load and increasing the efficiency of memory acquisition and encoding. They reduce cognitive load by grouping objects into a single verbal or visual cue that can be introduced into working memory. Learning is optimized when the load on working memory is minimized, enabling long-term memory to be facilitated.²⁸

BOX 2.

BOX 3.

Mnemonics may use rhyme, music, or visual cues to enhance memory. Most mnemonics used in medical practice and education are word-based, including:

• Acronyms—words, each letter of which stands for a particular piece of information to be recalled (such as RICE for treatment of a sprained joint: rest, ice, compression, elevation).
• Acrostics—sentences with the first letter of each word prompting the desired recollection (such as “To Zanzibar by motor car” for the branches of the facial nerve: temporal, zygomatic, buccal, mandibular, cervical).
• Alphabetical sequences (such as ABCDE of trauma assessment: airway, breathing, circulation, disability, exposure).²⁹

An appropriate teaching tool?

Dozens of mnemonics addressing psychiatric diagnosis and treatment have been published, but relatively few are widely used. Psychiatric educators may resist teaching with mnemonics, believing they might erode a humanistic approach to patients by reducing psychopathology to “a laundry list” of symptoms and the art of psychiatric diagnosis to a “check-box” endeavor. Mnemonics that use humor may be rejected as irreverent or unprofessional.³⁰ Publishing a novel mnemonic may be viewed with disdain by some as an “easy” way of padding a curriculum vitae.

BOX 4.

Entire Web sites exist to share mnemonics for medical education (see Related Resources). Thus it is likely that trainees are using them with or without their teachers’ supervision. Psychiatric educators need to be aware of the mnemonics their trainees are using and to:

• screen these tools for factual errors (such as incomplete diagnostic criteria)
• remind trainees that although mnemonics are useful, psychiatrists should approach patients as individuals without the prejudice of a potentially pejorative label.

Our methodology

In preparing this article, we gathered numerous mnemonics (some published and some novel) designed to capture the learner’s attention and impart information pertinent to psychiatric diagnosis and treatment. Whenever possible, we credited each mnemonic to its creator, but—given the difficulty in confirming authorship of (what in many cases has become) oral history—we’ve listed some mnemonics without citation.

Our list is far from complete because we likely are unaware of many mnemonics, and we have excluded some that seemed obscure, unwieldy, or redundant. We have not excluded mnemonics that some may view as pejorative but merely report their existence. Including them does not mean that we endorse them.

This article lists 32 mnemonics related to psychiatric diagnosis. Thus, it seems odd that an informal survey of >60 residents at the Massachusetts General Hospital (MGH)/McLean Residency Training Program in Psychiatry revealed that most were aware of only 2 or 3 psychiatric mnemonics, typically:

• SIG: E CAPS (a tool to recall the criteria for depression)
• DIG FAST (a list of criteria for diagnosing mania)
• WWHHHHIMPS (a tool for recalling life-threatening causes of delirium).

Although this unscientific survey may be biased because faculty or trainees at MGH created the above 3 mnemonics, it nonetheless begs the question of what qualities make a mnemonic memorable.

Learning theory provides several clues. George Miller’s classic 1956 paper, “The magical number seven, plus or minus two: some limits on our capacity for processing information,” discussed the finding that 7 seems to be the upper limit of individual pieces of data that can be easily remembered.³¹ Research also has shown that recruiting the limbic system (potentially through the use of humor) aids in the recall of otherwise dry, cortical information.^32,33

Intuitively, it would seem that nonrepeating letters would facilitate the recall of the linked data, allowing each letter to provide a distinct cue, without any clouding by redundancy. Of the 3 most popular psychiatric mnemonics, however, only DIG FAST fits the learning theory. It contains 7 letters, repeats no letters, and has the limbic cue of allowing the learner to imagine a person with mania digging furiously.

BOX 5.

SIG: E CAPS falls within the range of 7 plus or minus 2, includes a limbic cue (although often forgotten, it refers to the prescription of energy capsules for depression), but repeats the letter S.

WWHHHHIMPS, with 10 letters, exceeds the recommended range, repeats the W (appearing twice) and the H (appearing 4 times), and provides no clear limbic cue.

BOX 6.

It may be that recruiting the limbic system provides the greatest likelihood of recall. Recruiting this system may add increased valence to a particular mnemonic for a specific individual, but this same limbic valence may limit its usefulness in a professional context.

Related resources

• Free searchable database of medical mnemonics. www.medicalmnemonics.com.
• Robinson DJ. Mnemonics and more for psychiatry. Port Huron, MI: Rapid Psychler Press, 2001.

From SIG: E CAPS to CAGE and WWHHHHIMPS, mnemonics help practitioners and trainees recall important lists (such as criteria for depression, screening questions for alcoholism, or life-threatening causes of delirium, respectively). Mnemonics’ efficacy rests on the principle that grouped information is easier to remember than individual points of data.

Not everyone loves mnemonics, but recollecting diagnostic criteria is useful in clinical practice and research, on board examinations, and for insurance reimbursement. Thus, tools that assist in recalling diagnostic criteria have a role in psychiatric practice and teaching.

In this article, we present 32 mnemonics to help clinicians diagnose:

• affective disorders (Box 1)^1,2
• anxiety disorders (Box 2)^3-6
• medication adverse effects (Box 3)^7,8
• personality disorders (Box 4)^9-11
• addiction disorders (Box 5)^12,13
• causes of delirium (Box 6).¹⁴

We also discuss how mnemonics improve one’s memory, based on the principles of learning theory.

How mnemonics work

A mnemonic—from the Greek word “mnemonikos” (“of memory”)—links new data with previously learned information. Mnemonics assist in learning by reducing the amount of information (“cognitive load”) that needs to be stored for long-term processing and retrieval.¹⁵

Memory, defined as the “persistence of learning in a state that can be revealed at a later time,”¹⁶ can be divided into 2 types:

• declarative (a conscious recollection of facts, such as remembering a relative’s birthday)
• procedural (skills-based learning, such as riding a bicycle).

Declarative memory has a conscious component and may be mediated by the medial temporal lobe and cortical association structures. Procedural memory has less of a conscious component; it may involve the basal ganglia, cerebellum, and a variety of cortical sensory-perceptive regions.¹⁷

BOX 1.

Declarative memory can be subdivided into working memory and long-term memory.

With working memory, new items of information are held briefly so that encoding and eventual storage can take place.

Working memory guides decision-making and future planning and is intricately related to attention.^18-21 Functional MRI and positron emission tomography as well as neurocognitive testing have shown that working memory tasks activate the prefrontal cortex and brain regions specific to language and visuospatial memory.

The hippocampus is thought to rapidly absorb new information, and this data is consolidated and permanently stored via the prefrontal cortex.^22-26 Given the hippocampus’ limited storage capacity, new information (such as what you ate for breakfast 3 weeks ago) will disappear if it is not repeated regularly.¹⁷

Long-term memory, on the other hand, is encoded knowledge that is linked to facts learned in the past; it is consolidated in the brain and can be readily retrieved. Neuroimaging studies have demonstrated opposing patterns of activation in the hippocampus and prefrontal cortex, depending on whether the memory being recalled is:

• new (high hippocampal activity, low prefrontal cortex activity)
• old (low hippocampal activity, high prefrontal cortex activity).²⁷

Mnemonics are thought to affect working memory by reducing the introduced cognitive load and increasing the efficiency of memory acquisition and encoding. They reduce cognitive load by grouping objects into a single verbal or visual cue that can be introduced into working memory. Learning is optimized when the load on working memory is minimized, enabling long-term memory to be facilitated.²⁸

BOX 2.

BOX 3.

Mnemonics may use rhyme, music, or visual cues to enhance memory. Most mnemonics used in medical practice and education are word-based, including:

• Acronyms—words, each letter of which stands for a particular piece of information to be recalled (such as RICE for treatment of a sprained joint: rest, ice, compression, elevation).
• Acrostics—sentences with the first letter of each word prompting the desired recollection (such as “To Zanzibar by motor car” for the branches of the facial nerve: temporal, zygomatic, buccal, mandibular, cervical).
• Alphabetical sequences (such as ABCDE of trauma assessment: airway, breathing, circulation, disability, exposure).²⁹

An appropriate teaching tool?

Dozens of mnemonics addressing psychiatric diagnosis and treatment have been published, but relatively few are widely used. Psychiatric educators may resist teaching with mnemonics, believing they might erode a humanistic approach to patients by reducing psychopathology to “a laundry list” of symptoms and the art of psychiatric diagnosis to a “check-box” endeavor. Mnemonics that use humor may be rejected as irreverent or unprofessional.³⁰ Publishing a novel mnemonic may be viewed with disdain by some as an “easy” way of padding a curriculum vitae.

BOX 4.

Entire Web sites exist to share mnemonics for medical education (see Related Resources). Thus it is likely that trainees are using them with or without their teachers’ supervision. Psychiatric educators need to be aware of the mnemonics their trainees are using and to:

• screen these tools for factual errors (such as incomplete diagnostic criteria)
• remind trainees that although mnemonics are useful, psychiatrists should approach patients as individuals without the prejudice of a potentially pejorative label.

Our methodology

In preparing this article, we gathered numerous mnemonics (some published and some novel) designed to capture the learner’s attention and impart information pertinent to psychiatric diagnosis and treatment. Whenever possible, we credited each mnemonic to its creator, but—given the difficulty in confirming authorship of (what in many cases has become) oral history—we’ve listed some mnemonics without citation.

Our list is far from complete because we likely are unaware of many mnemonics, and we have excluded some that seemed obscure, unwieldy, or redundant. We have not excluded mnemonics that some may view as pejorative but merely report their existence. Including them does not mean that we endorse them.

This article lists 32 mnemonics related to psychiatric diagnosis. Thus, it seems odd that an informal survey of >60 residents at the Massachusetts General Hospital (MGH)/McLean Residency Training Program in Psychiatry revealed that most were aware of only 2 or 3 psychiatric mnemonics, typically:

• SIG: E CAPS (a tool to recall the criteria for depression)
• DIG FAST (a list of criteria for diagnosing mania)
• WWHHHHIMPS (a tool for recalling life-threatening causes of delirium).

Although this unscientific survey may be biased because faculty or trainees at MGH created the above 3 mnemonics, it nonetheless begs the question of what qualities make a mnemonic memorable.

Learning theory provides several clues. George Miller’s classic 1956 paper, “The magical number seven, plus or minus two: some limits on our capacity for processing information,” discussed the finding that 7 seems to be the upper limit of individual pieces of data that can be easily remembered.³¹ Research also has shown that recruiting the limbic system (potentially through the use of humor) aids in the recall of otherwise dry, cortical information.^32,33

Intuitively, it would seem that nonrepeating letters would facilitate the recall of the linked data, allowing each letter to provide a distinct cue, without any clouding by redundancy. Of the 3 most popular psychiatric mnemonics, however, only DIG FAST fits the learning theory. It contains 7 letters, repeats no letters, and has the limbic cue of allowing the learner to imagine a person with mania digging furiously.

BOX 5.

SIG: E CAPS falls within the range of 7 plus or minus 2, includes a limbic cue (although often forgotten, it refers to the prescription of energy capsules for depression), but repeats the letter S.

WWHHHHIMPS, with 10 letters, exceeds the recommended range, repeats the W (appearing twice) and the H (appearing 4 times), and provides no clear limbic cue.

BOX 6.

It may be that recruiting the limbic system provides the greatest likelihood of recall. Recruiting this system may add increased valence to a particular mnemonic for a specific individual, but this same limbic valence may limit its usefulness in a professional context.

Related resources

• Free searchable database of medical mnemonics. www.medicalmnemonics.com.
• Robinson DJ. Mnemonics and more for psychiatry. Port Huron, MI: Rapid Psychler Press, 2001.

From SIG: E CAPS to CAGE and WWHHHHIMPS, mnemonics help practitioners and trainees recall important lists (such as criteria for depression, screening questions for alcoholism, or life-threatening causes of delirium, respectively). Mnemonics’ efficacy rests on the principle that grouped information is easier to remember than individual points of data.

Not everyone loves mnemonics, but recollecting diagnostic criteria is useful in clinical practice and research, on board examinations, and for insurance reimbursement. Thus, tools that assist in recalling diagnostic criteria have a role in psychiatric practice and teaching.

In this article, we present 32 mnemonics to help clinicians diagnose:

• affective disorders (Box 1)^1,2
• anxiety disorders (Box 2)^3-6
• medication adverse effects (Box 3)^7,8
• personality disorders (Box 4)^9-11
• addiction disorders (Box 5)^12,13
• causes of delirium (Box 6).¹⁴

We also discuss how mnemonics improve one’s memory, based on the principles of learning theory.

How mnemonics work

A mnemonic—from the Greek word “mnemonikos” (“of memory”)—links new data with previously learned information. Mnemonics assist in learning by reducing the amount of information (“cognitive load”) that needs to be stored for long-term processing and retrieval.¹⁵

Memory, defined as the “persistence of learning in a state that can be revealed at a later time,”¹⁶ can be divided into 2 types:

• declarative (a conscious recollection of facts, such as remembering a relative’s birthday)
• procedural (skills-based learning, such as riding a bicycle).

Declarative memory has a conscious component and may be mediated by the medial temporal lobe and cortical association structures. Procedural memory has less of a conscious component; it may involve the basal ganglia, cerebellum, and a variety of cortical sensory-perceptive regions.¹⁷

BOX 1.

Declarative memory can be subdivided into working memory and long-term memory.

With working memory, new items of information are held briefly so that encoding and eventual storage can take place.

Working memory guides decision-making and future planning and is intricately related to attention.^18-21 Functional MRI and positron emission tomography as well as neurocognitive testing have shown that working memory tasks activate the prefrontal cortex and brain regions specific to language and visuospatial memory.

The hippocampus is thought to rapidly absorb new information, and this data is consolidated and permanently stored via the prefrontal cortex.^22-26 Given the hippocampus’ limited storage capacity, new information (such as what you ate for breakfast 3 weeks ago) will disappear if it is not repeated regularly.¹⁷

Long-term memory, on the other hand, is encoded knowledge that is linked to facts learned in the past; it is consolidated in the brain and can be readily retrieved. Neuroimaging studies have demonstrated opposing patterns of activation in the hippocampus and prefrontal cortex, depending on whether the memory being recalled is:

• new (high hippocampal activity, low prefrontal cortex activity)
• old (low hippocampal activity, high prefrontal cortex activity).²⁷

Mnemonics are thought to affect working memory by reducing the introduced cognitive load and increasing the efficiency of memory acquisition and encoding. They reduce cognitive load by grouping objects into a single verbal or visual cue that can be introduced into working memory. Learning is optimized when the load on working memory is minimized, enabling long-term memory to be facilitated.²⁸

BOX 2.

BOX 3.

Mnemonics may use rhyme, music, or visual cues to enhance memory. Most mnemonics used in medical practice and education are word-based, including:

• Acronyms—words, each letter of which stands for a particular piece of information to be recalled (such as RICE for treatment of a sprained joint: rest, ice, compression, elevation).
• Acrostics—sentences with the first letter of each word prompting the desired recollection (such as “To Zanzibar by motor car” for the branches of the facial nerve: temporal, zygomatic, buccal, mandibular, cervical).
• Alphabetical sequences (such as ABCDE of trauma assessment: airway, breathing, circulation, disability, exposure).²⁹

An appropriate teaching tool?

Dozens of mnemonics addressing psychiatric diagnosis and treatment have been published, but relatively few are widely used. Psychiatric educators may resist teaching with mnemonics, believing they might erode a humanistic approach to patients by reducing psychopathology to “a laundry list” of symptoms and the art of psychiatric diagnosis to a “check-box” endeavor. Mnemonics that use humor may be rejected as irreverent or unprofessional.³⁰ Publishing a novel mnemonic may be viewed with disdain by some as an “easy” way of padding a curriculum vitae.

BOX 4.

Entire Web sites exist to share mnemonics for medical education (see Related Resources). Thus it is likely that trainees are using them with or without their teachers’ supervision. Psychiatric educators need to be aware of the mnemonics their trainees are using and to:

• screen these tools for factual errors (such as incomplete diagnostic criteria)
• remind trainees that although mnemonics are useful, psychiatrists should approach patients as individuals without the prejudice of a potentially pejorative label.

Our methodology

In preparing this article, we gathered numerous mnemonics (some published and some novel) designed to capture the learner’s attention and impart information pertinent to psychiatric diagnosis and treatment. Whenever possible, we credited each mnemonic to its creator, but—given the difficulty in confirming authorship of (what in many cases has become) oral history—we’ve listed some mnemonics without citation.

Our list is far from complete because we likely are unaware of many mnemonics, and we have excluded some that seemed obscure, unwieldy, or redundant. We have not excluded mnemonics that some may view as pejorative but merely report their existence. Including them does not mean that we endorse them.

This article lists 32 mnemonics related to psychiatric diagnosis. Thus, it seems odd that an informal survey of >60 residents at the Massachusetts General Hospital (MGH)/McLean Residency Training Program in Psychiatry revealed that most were aware of only 2 or 3 psychiatric mnemonics, typically:

• SIG: E CAPS (a tool to recall the criteria for depression)
• DIG FAST (a list of criteria for diagnosing mania)
• WWHHHHIMPS (a tool for recalling life-threatening causes of delirium).

Although this unscientific survey may be biased because faculty or trainees at MGH created the above 3 mnemonics, it nonetheless begs the question of what qualities make a mnemonic memorable.

Learning theory provides several clues. George Miller’s classic 1956 paper, “The magical number seven, plus or minus two: some limits on our capacity for processing information,” discussed the finding that 7 seems to be the upper limit of individual pieces of data that can be easily remembered.³¹ Research also has shown that recruiting the limbic system (potentially through the use of humor) aids in the recall of otherwise dry, cortical information.^32,33

Intuitively, it would seem that nonrepeating letters would facilitate the recall of the linked data, allowing each letter to provide a distinct cue, without any clouding by redundancy. Of the 3 most popular psychiatric mnemonics, however, only DIG FAST fits the learning theory. It contains 7 letters, repeats no letters, and has the limbic cue of allowing the learner to imagine a person with mania digging furiously.

BOX 5.

SIG: E CAPS falls within the range of 7 plus or minus 2, includes a limbic cue (although often forgotten, it refers to the prescription of energy capsules for depression), but repeats the letter S.

WWHHHHIMPS, with 10 letters, exceeds the recommended range, repeats the W (appearing twice) and the H (appearing 4 times), and provides no clear limbic cue.

BOX 6.

It may be that recruiting the limbic system provides the greatest likelihood of recall. Recruiting this system may add increased valence to a particular mnemonic for a specific individual, but this same limbic valence may limit its usefulness in a professional context.

Related resources

• Free searchable database of medical mnemonics. www.medicalmnemonics.com.
• Robinson DJ. Mnemonics and more for psychiatry. Port Huron, MI: Rapid Psychler Press, 2001.