Current Psychiatry

Split treatment occurs when a psychiatrist and psychotherapist share a patient’s care. These relationships create opportunities and challenges for both clinicians and patients. Some patients may experience positive transference, such as nurturing and idealization, whereas others may have negative transference, including narcissistic wounds and devaluation of 1 or both clinicians.¹ This article will help you navigate split treatment relationships.

Types of relationships

Duplicative services involve a patient receiving psychotherapy or psychopharmacology from ≥2 clinicians; the latter is more common. Duplicative psychopharmacology can occur inadvertently when drugs prescribed by a nonpsychiatrist overlap with psychotropics in their effects.

Duplicative opinions. When sought with the knowledge of all parties, second opinions can strengthen the doctor-patient relationship and improve outcomes. Duplicative opinions may be pathologic when patients do not disclose their opinion-seeking to both clinicians. This form of “doctor shopping” is common among patients with personality disorders, especially borderline personality disorder, and those who seek drugs from multiple providers to abuse, stockpile, or sell them.

Strategies for split treatment

During the first session, discuss the patient’s expectations. Establish a treatment plan, setting forth what you are and are not responsible for managing. If the patient is transferring treatment to you, ask if he or she has terminated treatment with the prior clinician.

If psychotropics are indicated, ask the patient to agree that only you will manage medications for mood, attention, cognition, energy, and sleep, except in an emergency such as an allergic reaction. Inform your patient that having multiple physicians managing these symptoms puts him or her at risk for drug-drug interactions and ineffective treatment. Ask your patient to notify you about any new medications.

A patient’s psychodynamic issues with medications can complicate pharmacotherapy. For example, a patient may experience transference reactions to medications, such as issues relating to control of the treatment process or psychogenic side effects. Ideally, the psychotherapist and psychiatrist work together to determine which clinician could better address these issues.

If a patient reports that another clinician devalued your treatment, don’t react immediately. Explore with your patient how the discussion with the other clinician came about. Patients can misinterpret what other clinicians say or lie in an attempt to create conflict between you and your colleagues. Give the other clinician the benefit of the doubt and if possible, speak to him or her before commenting about the alleged statements. Look at these situations as an opportunity to set limits with patients.

Split treatment occurs when a psychiatrist and psychotherapist share a patient’s care. These relationships create opportunities and challenges for both clinicians and patients. Some patients may experience positive transference, such as nurturing and idealization, whereas others may have negative transference, including narcissistic wounds and devaluation of 1 or both clinicians.¹ This article will help you navigate split treatment relationships.

Types of relationships

Duplicative services involve a patient receiving psychotherapy or psychopharmacology from ≥2 clinicians; the latter is more common. Duplicative psychopharmacology can occur inadvertently when drugs prescribed by a nonpsychiatrist overlap with psychotropics in their effects.

Duplicative opinions. When sought with the knowledge of all parties, second opinions can strengthen the doctor-patient relationship and improve outcomes. Duplicative opinions may be pathologic when patients do not disclose their opinion-seeking to both clinicians. This form of “doctor shopping” is common among patients with personality disorders, especially borderline personality disorder, and those who seek drugs from multiple providers to abuse, stockpile, or sell them.

Strategies for split treatment

During the first session, discuss the patient’s expectations. Establish a treatment plan, setting forth what you are and are not responsible for managing. If the patient is transferring treatment to you, ask if he or she has terminated treatment with the prior clinician.

If psychotropics are indicated, ask the patient to agree that only you will manage medications for mood, attention, cognition, energy, and sleep, except in an emergency such as an allergic reaction. Inform your patient that having multiple physicians managing these symptoms puts him or her at risk for drug-drug interactions and ineffective treatment. Ask your patient to notify you about any new medications.

A patient’s psychodynamic issues with medications can complicate pharmacotherapy. For example, a patient may experience transference reactions to medications, such as issues relating to control of the treatment process or psychogenic side effects. Ideally, the psychotherapist and psychiatrist work together to determine which clinician could better address these issues.

If a patient reports that another clinician devalued your treatment, don’t react immediately. Explore with your patient how the discussion with the other clinician came about. Patients can misinterpret what other clinicians say or lie in an attempt to create conflict between you and your colleagues. Give the other clinician the benefit of the doubt and if possible, speak to him or her before commenting about the alleged statements. Look at these situations as an opportunity to set limits with patients.

Split treatment occurs when a psychiatrist and psychotherapist share a patient’s care. These relationships create opportunities and challenges for both clinicians and patients. Some patients may experience positive transference, such as nurturing and idealization, whereas others may have negative transference, including narcissistic wounds and devaluation of 1 or both clinicians.¹ This article will help you navigate split treatment relationships.

Types of relationships

Duplicative services involve a patient receiving psychotherapy or psychopharmacology from ≥2 clinicians; the latter is more common. Duplicative psychopharmacology can occur inadvertently when drugs prescribed by a nonpsychiatrist overlap with psychotropics in their effects.

Duplicative opinions. When sought with the knowledge of all parties, second opinions can strengthen the doctor-patient relationship and improve outcomes. Duplicative opinions may be pathologic when patients do not disclose their opinion-seeking to both clinicians. This form of “doctor shopping” is common among patients with personality disorders, especially borderline personality disorder, and those who seek drugs from multiple providers to abuse, stockpile, or sell them.

Strategies for split treatment

During the first session, discuss the patient’s expectations. Establish a treatment plan, setting forth what you are and are not responsible for managing. If the patient is transferring treatment to you, ask if he or she has terminated treatment with the prior clinician.

If psychotropics are indicated, ask the patient to agree that only you will manage medications for mood, attention, cognition, energy, and sleep, except in an emergency such as an allergic reaction. Inform your patient that having multiple physicians managing these symptoms puts him or her at risk for drug-drug interactions and ineffective treatment. Ask your patient to notify you about any new medications.

A patient’s psychodynamic issues with medications can complicate pharmacotherapy. For example, a patient may experience transference reactions to medications, such as issues relating to control of the treatment process or psychogenic side effects. Ideally, the psychotherapist and psychiatrist work together to determine which clinician could better address these issues.

If a patient reports that another clinician devalued your treatment, don’t react immediately. Explore with your patient how the discussion with the other clinician came about. Patients can misinterpret what other clinicians say or lie in an attempt to create conflict between you and your colleagues. Give the other clinician the benefit of the doubt and if possible, speak to him or her before commenting about the alleged statements. Look at these situations as an opportunity to set limits with patients.

Personalized care is at the heart of good medical care. It is an indispensable ingredient for optimal clinical outcomes because each patient is unique, as an individual and as a patient, and requires customized treatment.

If 10 patients with depression walk into a psychiatrist’s office on any given day, each will be different and should be treated accordingly. Their symptoms may be similar thematically but they differ widely in presentation and content. Their medical and psychiatric histories and social, educational, religious, ethnic, socioeconomic, and attitudinal diversity can be stunning in complexity and disparity. Just as patients’ symptoms can be similar yet different, so can their response to a specific antidepressant or psychotherapy. Their clinical and functional outcomes will vary widely in degree and valence. Every psychiatrist expects (and enjoys) the richness of patient backgrounds and manages each individually.

Given these individual differences among our psychiatric patients, why are practitioners being barraged by various entities to abandon the traditional medical approach to their patients? Why is there a push to transform personalized clinical care to an assembly-line system, where patients are defined by their disease and are managed like “human widgets” as though they can be “processed” in an identical, protocolized, mechanical manner? This is completely antithetical to the magnificent personal approach inherent in the classic and highly effective doctor-patient relationship.

There is nothing wrong with treating patients based on up-to-date practice guidelines and evidence-based principles of clinical effectiveness. The issue is whether clinical decisions should be made by the physician, one patient at a time, rather than imposing the dreaded “cookie-cutter” approach of protocols or algorithms on a population of patients whose only commonality is a DSM-IV-TR diagnosis. The not-so-hidden agenda of the business-oriented managed care systems is to lower costs, not to provide the best personalized medical care. Who came up with the absurd notion that there is such a thing as “an average patient” who would respond to a prepackaged, economically efficient “average treatment”? That is a serious disservice to the spectrum of patients suffering from psychiatric illnesses and an insult to skilled, compassionate psychiatrists who can provide customized care to each patient.

It is certainly paradoxical that at a time when personalized medicine is advocated as “best practice” in medical care, managed care health systems are propagating and implementing a contrarian movement of homogenizing treatment into rigid protocols with a preset, algorithmic approach. These competing messages create a confusing state of cognitive dissonance, especially for trainees, as to how clinicians should deliver medical care for their patients.

It is well known that a large proportion of psychiatric disorders (>80%) have no evidence-based, FDA-approved treatments, and no practice guidelines, protocols, or standards of care.¹ This is where psychiatrists have to use more art than science—including the necessary, but often maligned, off-label treatments—to help reduce their patients’ suffering. In these situations, the physician-patient relationship simply cannot be superseded by any prepackaged protocol, and physicians should decide what is best for their patient.

So let physicians unite behind what makes medicine such a noble profession: combining the best available scientific knowledge with experience and well-honed clinical judgment to deliver customized care, one patient at a time. We must treat our patients exactly as we want to be treated when we inevitably suffer from an illness.

Personalized care is at the heart of good medical care. It is an indispensable ingredient for optimal clinical outcomes because each patient is unique, as an individual and as a patient, and requires customized treatment.

If 10 patients with depression walk into a psychiatrist’s office on any given day, each will be different and should be treated accordingly. Their symptoms may be similar thematically but they differ widely in presentation and content. Their medical and psychiatric histories and social, educational, religious, ethnic, socioeconomic, and attitudinal diversity can be stunning in complexity and disparity. Just as patients’ symptoms can be similar yet different, so can their response to a specific antidepressant or psychotherapy. Their clinical and functional outcomes will vary widely in degree and valence. Every psychiatrist expects (and enjoys) the richness of patient backgrounds and manages each individually.

Given these individual differences among our psychiatric patients, why are practitioners being barraged by various entities to abandon the traditional medical approach to their patients? Why is there a push to transform personalized clinical care to an assembly-line system, where patients are defined by their disease and are managed like “human widgets” as though they can be “processed” in an identical, protocolized, mechanical manner? This is completely antithetical to the magnificent personal approach inherent in the classic and highly effective doctor-patient relationship.

There is nothing wrong with treating patients based on up-to-date practice guidelines and evidence-based principles of clinical effectiveness. The issue is whether clinical decisions should be made by the physician, one patient at a time, rather than imposing the dreaded “cookie-cutter” approach of protocols or algorithms on a population of patients whose only commonality is a DSM-IV-TR diagnosis. The not-so-hidden agenda of the business-oriented managed care systems is to lower costs, not to provide the best personalized medical care. Who came up with the absurd notion that there is such a thing as “an average patient” who would respond to a prepackaged, economically efficient “average treatment”? That is a serious disservice to the spectrum of patients suffering from psychiatric illnesses and an insult to skilled, compassionate psychiatrists who can provide customized care to each patient.

It is certainly paradoxical that at a time when personalized medicine is advocated as “best practice” in medical care, managed care health systems are propagating and implementing a contrarian movement of homogenizing treatment into rigid protocols with a preset, algorithmic approach. These competing messages create a confusing state of cognitive dissonance, especially for trainees, as to how clinicians should deliver medical care for their patients.

It is well known that a large proportion of psychiatric disorders (>80%) have no evidence-based, FDA-approved treatments, and no practice guidelines, protocols, or standards of care.¹ This is where psychiatrists have to use more art than science—including the necessary, but often maligned, off-label treatments—to help reduce their patients’ suffering. In these situations, the physician-patient relationship simply cannot be superseded by any prepackaged protocol, and physicians should decide what is best for their patient.

So let physicians unite behind what makes medicine such a noble profession: combining the best available scientific knowledge with experience and well-honed clinical judgment to deliver customized care, one patient at a time. We must treat our patients exactly as we want to be treated when we inevitably suffer from an illness.

Personalized care is at the heart of good medical care. It is an indispensable ingredient for optimal clinical outcomes because each patient is unique, as an individual and as a patient, and requires customized treatment.

If 10 patients with depression walk into a psychiatrist’s office on any given day, each will be different and should be treated accordingly. Their symptoms may be similar thematically but they differ widely in presentation and content. Their medical and psychiatric histories and social, educational, religious, ethnic, socioeconomic, and attitudinal diversity can be stunning in complexity and disparity. Just as patients’ symptoms can be similar yet different, so can their response to a specific antidepressant or psychotherapy. Their clinical and functional outcomes will vary widely in degree and valence. Every psychiatrist expects (and enjoys) the richness of patient backgrounds and manages each individually.

Given these individual differences among our psychiatric patients, why are practitioners being barraged by various entities to abandon the traditional medical approach to their patients? Why is there a push to transform personalized clinical care to an assembly-line system, where patients are defined by their disease and are managed like “human widgets” as though they can be “processed” in an identical, protocolized, mechanical manner? This is completely antithetical to the magnificent personal approach inherent in the classic and highly effective doctor-patient relationship.

There is nothing wrong with treating patients based on up-to-date practice guidelines and evidence-based principles of clinical effectiveness. The issue is whether clinical decisions should be made by the physician, one patient at a time, rather than imposing the dreaded “cookie-cutter” approach of protocols or algorithms on a population of patients whose only commonality is a DSM-IV-TR diagnosis. The not-so-hidden agenda of the business-oriented managed care systems is to lower costs, not to provide the best personalized medical care. Who came up with the absurd notion that there is such a thing as “an average patient” who would respond to a prepackaged, economically efficient “average treatment”? That is a serious disservice to the spectrum of patients suffering from psychiatric illnesses and an insult to skilled, compassionate psychiatrists who can provide customized care to each patient.

It is certainly paradoxical that at a time when personalized medicine is advocated as “best practice” in medical care, managed care health systems are propagating and implementing a contrarian movement of homogenizing treatment into rigid protocols with a preset, algorithmic approach. These competing messages create a confusing state of cognitive dissonance, especially for trainees, as to how clinicians should deliver medical care for their patients.

It is well known that a large proportion of psychiatric disorders (>80%) have no evidence-based, FDA-approved treatments, and no practice guidelines, protocols, or standards of care.¹ This is where psychiatrists have to use more art than science—including the necessary, but often maligned, off-label treatments—to help reduce their patients’ suffering. In these situations, the physician-patient relationship simply cannot be superseded by any prepackaged protocol, and physicians should decide what is best for their patient.

So let physicians unite behind what makes medicine such a noble profession: combining the best available scientific knowledge with experience and well-honed clinical judgment to deliver customized care, one patient at a time. We must treat our patients exactly as we want to be treated when we inevitably suffer from an illness.

Discuss this article at http://currentpsychiatry.blogspot.com/2010/08/adolescents-who-self-harm.html#comments

Josh, age 16, gets poor grades in school and occasionally smokes marijuana and abuses inhalants. After his girlfriend breaks up with him, he cuts his wrist with a hunting knife. While bleeding profusely, Josh calls his mother at work, who calls 911. The cut is deep and requires sutures. Josh says he did not try to kill himself; he only wanted to carve his girlfriend’s initials into his wrist to show his love for her.

When treating teenagers with self-harming thoughts and behavior, it may be difficult to distinguish suicide attempts from self-injury without intent to die. Understanding adolescent self-harm, suicide risk assessment, and treatment options guides clinicians to appropriate interventions. Recognizing the need for aggressive treatment—including psychiatric hospitalization—is essential to keeping self-harming teenagers safe.

Suicidal vs nonsuicidal self-harm

Suicidal behavior involves intent to end one’s life and includes ideation (thoughts) and actions (non fatal or fatal attempts).¹ Nonsuicidal self-injury (NSSI) involves socially unacceptable, self-inflicted harm to one’s body without intent to die.²

Suicide is the third leading cause of death among youths age 12 to 19, claiming almost 2,000 lives each year.³ Nearly 1 in 5 (17%) U.S. high school students has suicidal thoughts each year, and almost 1 in 10 (8%) attempts suicide.⁴

Studies report a 13% to 23% lifetime prevalence of NSSI.⁵ These behaviors often begin between age 13 to 15.⁶ Cutting and hitting are the most common forms of NSSI; other methods include burning, scratching, and interfering with wound healing. Most teens who harm themselves without suicidal intent report that they feel little or no pain during the act.⁵ Unlike suicide attempts, NSSI can be viewed as a means to stay alive. Many adolescents injure themselves to cope with overwhelming feelings that can produce suicidal thoughts. Self-injury may distract the adolescent from painful emotions, reduce tensions, or penetrate numbness.⁷

Teenagers who hurt themselves but do not intend to die are at high risk for suicide and suicide attempts. Adolescents who engage in NSSI are more likely to experience suicidal behaviors, and vice versa.⁸ In a large study, 70% of adolescents who engaged in NSSI had made at least 1 suicide attempt and 55% made multiple suicide attempts.² Current suicidal ideation is a risk factor for suicide, and a past suicide attempt is the strongest predictor of future suicidal behavior.⁹

Risk factors for suicidal behavior and NSSI overlap (Table 1)^2,5,6,10,11 and include:

• depression
• substance use
• anxiety
• impulsive aggression
• history of childhood trauma.

Many teens who engage in NSSI report depression.² A history of psychiatric illness—especially depression—increases the likelihood of adolescent suicide.⁸ A study comparing adolescents who engaged in NSSI with those who attempted suicide found that both groups reported similar levels of suicidal ideation and depressive symptoms.⁶ However, adolescents with a history of NSSI and attempted suicide reported higher levels of suicidal ideation and fewer reasons for living than those who attempted suicide but have no history of NSSI.¹²

Factors that protect against suicidal behavior include:

• a good parent-child relationship
• strong cultural or religious values
• an intact family
• a sense of connection with peer group and community.¹³

No studies have determined protective factors for NSSI.

Table 1

Characteristics of teens who harm themselves

CASE CONTINUED: ‘No point in living’

As Josh becomes less guarded, he says that he sees “no point in living” without his girlfriend. He thought the only way to feel better was to “get high,” but this left him feeling even more despondent and anxious. He wrote a suicide note, but after cutting himself he was unsure he wanted to die. Josh says that when he feels depressed he can’t talk to his parents because “they wouldn’t understand and don’t care.”

Assessing self-harming adolescents

Distinguishing between suicidal behaviors and NSSI can be challenging (Table 2). Identifying risk factors for adolescent suicidal behavior must be coupled with a thorough psychiatric evaluation. If possible, interview the adolescent alone and obtain collateral information from parents, family members, teachers, caseworkers, probation officers, and others as needed. Also examine family interactions because conflicts and communication problems could undermine the teenager’s safety.

Consider using standardized measures of suicidal intentions such as the Scale for Suicidal Ideation (SSI).¹⁴ Although the SSI was developed for adults, a large case-control study validated the scale’s use in adolescent psychiatric outpatients and students.¹⁵ In addition to assessing subjective reports of suicidal intent, the SSI also takes into account objective indicators of increased risk, such as planning an attempt, hiding details from others, and making preparations for death.

Questions to ask. After a self-harm incident, it may be helpful to begin the psychiatric interview with a general question such as, “What happened that led you to hurt yourself?” This does not categorize the act as suicidal and allows the adolescent to describe it in his or her own words. Shea¹⁶ suggests normalizing the act by assuming that self-harm occurred, rather than making the patient admit to it. For example, an interviewer might say “Many people I know who are hurting inside also try to hurt their body; how often do you do that?”

Inquire about suicidal intent in a few ways. For example, first ask, “Do you ever wish you were dead?” and follow up with, “Would you ever do anything to try to make yourself dead?” Asking about suicidal thoughts does not increase suicidal thoughts or behavior.^10,17

Reviewing thoughts and feelings leading up to a self-harm act can help identify triggers, coping difficulties, and issues to address in treatment. This behavioral analysis can be completed using the mnemonic ABC:

• Antecedents (situations or stressors leading to self-harming thoughts or actions)
• Behavior characteristics (frequency, intensity, and duration of self-harming)
• Consequences (eg, emotional relief, care and attention from others).

The results of this analysis could suggest treatment strategies, such as cognitive restructuring or techniques for decreasing feelings of distress.

The Risk of Suicide Questionnaire, which is designed for adolescents, asks:¹⁸

• Are you here because you tried to hurt yourself?
• In the past week, have you been having thoughts about killing yourself?
• Have you ever tried to hurt yourself in the past?
• Has something very stressful happened to you in the past few weeks?

Research has yet to determine whether this simple, rapid screen accurately identifies the need for psychiatric hospitalization or risk for suicidal outcomes. Although clinician- and self-administered suicide questionnaires may be useful for screening large populations, they are not a substitute for a thorough clinical assessment.

Inpatient or outpatient? When evaluating self-harming adolescents, first determine if they are in imminent danger of suicide and if more intensive services, such as hospitalization, are needed to maintain safety. Inpatient psychiatric services are appropriate for adolescents with suicidal thoughts or self-harm behaviors in addition to acute psychiatric disorders, significant substance abuse, serious medical issues, poor social supports, or inability to be managed safely as an outpatient.¹⁹ See the Table for a list of additional factors to consider.

Consent for treatment may be required because many self-harming adolescents do not present with life-threatening symptoms. Laws regarding consent vary among states. In some jurisdictions, patients age ≥15 can consent to mental health treatment without parents’ knowledge or consent. If an adolescent is in imminent danger and cannot voluntarily consent to treatment, physicians can initiate mental health “holds.” Some states allow registered nurses, psychologists, licensed social workers, and others to initiate mental health holds.

Table 2

Strategies for assessing adolescent self-harm

CASE CONTINUED: Inpatient treatment

After the interview Josh says he still feels that “there is no point in living” and he cannot develop an adequate safety plan with his family. He is hospitalized to maintain safety, improve his coping skills and communication with his family, and mobilize safety plans, social supports, and follow-up care.

Maintaining safety

Psychosocial treatments for suicidal behaviors and NSSI are similar because with both, the priority is to help the patient maintain safety. This may include:

• developing a collaborative safety plan with family
• increasing monitoring
• removing access to firearms or other lethal means
• helping the adolescent to develop alternate, safer coping methods.

Many clinicians rely on no-harm contracts or agreements; however, there is no evidence that they are effective.²⁰ The American Psychiatric Association recommends against using no-harm contracts with patients who are new, in an emergency setting, using substances, agitated, psychotic, or impulsive.²¹ Instead, clinicians, adolescents, and families can discuss specific steps the patient can take to remain safe. This collaborative plan should identify situations likely to trigger self-harming impulses; adaptive ways the teenager can cope, such as taking a nap or jogging; methods for communicating distress to family members and other helpers; and places to go for help, such as an emergency room. These safety plans should draw on the patient’s internal and external resources.

CASE CONTINUED: Strengthening relationships

While in the hospital, Josh finds it helpful to use a 0-to-10 scale to measure his distress and let his family know the intensity of his feelings. He identifies situations when he felt like hurting himself, such as being humiliated in math class. Josh learns about cognitive distortions—such as “they don’t care” and “there is no point in living”—and discusses methods for managing his feelings if he encounters further disappointments. His parents become more attentive when Josh explains his feelings, which allows the family to develop a collaborative safety plan. Josh decides to strengthen friendships he had been neglecting and agrees to attend a substance abuse treatment program.

Psychosocial treatment

In addition to maintaining safety, treatment goals for self-harming adolescents include:

• managing underlying psychiatric disorders
• identifying triggers for self-injurious acts
• improving family relationships
• developing better communication and coping skills.

Improving affective language skills, acquiring frustration tolerance, and learning alternatives to self-injury are key to strengthening coping abilities. Address problem-solving skills because self-harming adolescents often lack these abilities.²²

Treatment of self-harming adolescents often consists of cognitive-behavioral therapy (CBT)²³ or dialectical behavior therapy (DBT).²⁴ CBT involves examining cognitive distortions or otherwise unhealthy beliefs about oneself, others, and life in general, focusing specifically on thoughts the patient has immediately before engaging in self-harm. DBT also integrates emotion regulation training and mindfulness. A review of 28 studies found these therapies effectively reduced self-harm behaviors in adults.²⁵ However, few studies have examined these therapies’ efficacy in self-harming adolescents.

Pharmacotherapy

Psychopharmacology should focus on treating underlying psychiatric disorders. No medications are specifically effective for treating suicidal thoughts, suicidal behaviors, or NSSI. Some evidence suggests that antidepressants may trigger suicidal thoughts in a small proportion of youth,²⁶ but the benefits of antidepressants outweigh the risk of suicidal thoughts.²⁷ When prescribing antidepressants, inform patients and their parents of possible adverse reactions and monitor the patient regularly.²⁸

Take precautions when prescribing medication for self-harming adolescents. For example, benzodiazepines may cause disinhibition, and larger quantities of medication could be lethal in an overdose. If possible, arrange for parents or guardians to monitor medication use.

What to document

Good documentation is especially helpful when an adolescent requires involuntary commitment or is discharged home. Involuntary commitment is based on legal interpretation of 3 circumstances—danger to self, danger to others, or gravely disabled—in which safety concerns may override an individual’s civil rights. If involuntary commitment is needed, a physician must clarify how the youth meets ≥1 of these criteria. If the adolescent is discharged, document that the patient is not an imminent danger to self or others and why you made this determination. Also note that follow-up services and a safety plan are in place, a parent will monitor safety issues and remove firearms and other lethal means from the home, and acute conflicts have been resolved. Other details, such as using the patient’s words to describe reasons for living, can be helpful.

Related Resources

• National Alliance for the Mentally Ill. National Helpline. 800-950-6264.
• American Foundation for Suicide Prevention. www.afsp.org.
• American Association of Suicidology. www.suicidology.org.
• National Suicide Prevention Lifeline. 800-273-TALK (8255).

Acknowledgements

The authors wish to thank students Scott Schubert from Regis University, Denver, CO, and Emily Peterson from Beloit College, Beloit, WI, for their help in preparing the manuscript.

Disclosure

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

Table

Hospitalization or home? Acute crisis planning for self-harming youth

Discuss this article at http://currentpsychiatry.blogspot.com/2010/08/adolescents-who-self-harm.html#comments

Josh, age 16, gets poor grades in school and occasionally smokes marijuana and abuses inhalants. After his girlfriend breaks up with him, he cuts his wrist with a hunting knife. While bleeding profusely, Josh calls his mother at work, who calls 911. The cut is deep and requires sutures. Josh says he did not try to kill himself; he only wanted to carve his girlfriend’s initials into his wrist to show his love for her.

When treating teenagers with self-harming thoughts and behavior, it may be difficult to distinguish suicide attempts from self-injury without intent to die. Understanding adolescent self-harm, suicide risk assessment, and treatment options guides clinicians to appropriate interventions. Recognizing the need for aggressive treatment—including psychiatric hospitalization—is essential to keeping self-harming teenagers safe.

Suicidal vs nonsuicidal self-harm

Suicidal behavior involves intent to end one’s life and includes ideation (thoughts) and actions (non fatal or fatal attempts).¹ Nonsuicidal self-injury (NSSI) involves socially unacceptable, self-inflicted harm to one’s body without intent to die.²

Suicide is the third leading cause of death among youths age 12 to 19, claiming almost 2,000 lives each year.³ Nearly 1 in 5 (17%) U.S. high school students has suicidal thoughts each year, and almost 1 in 10 (8%) attempts suicide.⁴

Studies report a 13% to 23% lifetime prevalence of NSSI.⁵ These behaviors often begin between age 13 to 15.⁶ Cutting and hitting are the most common forms of NSSI; other methods include burning, scratching, and interfering with wound healing. Most teens who harm themselves without suicidal intent report that they feel little or no pain during the act.⁵ Unlike suicide attempts, NSSI can be viewed as a means to stay alive. Many adolescents injure themselves to cope with overwhelming feelings that can produce suicidal thoughts. Self-injury may distract the adolescent from painful emotions, reduce tensions, or penetrate numbness.⁷

Teenagers who hurt themselves but do not intend to die are at high risk for suicide and suicide attempts. Adolescents who engage in NSSI are more likely to experience suicidal behaviors, and vice versa.⁸ In a large study, 70% of adolescents who engaged in NSSI had made at least 1 suicide attempt and 55% made multiple suicide attempts.² Current suicidal ideation is a risk factor for suicide, and a past suicide attempt is the strongest predictor of future suicidal behavior.⁹

Risk factors for suicidal behavior and NSSI overlap (Table 1)^2,5,6,10,11 and include:

• depression
• substance use
• anxiety
• impulsive aggression
• history of childhood trauma.

Many teens who engage in NSSI report depression.² A history of psychiatric illness—especially depression—increases the likelihood of adolescent suicide.⁸ A study comparing adolescents who engaged in NSSI with those who attempted suicide found that both groups reported similar levels of suicidal ideation and depressive symptoms.⁶ However, adolescents with a history of NSSI and attempted suicide reported higher levels of suicidal ideation and fewer reasons for living than those who attempted suicide but have no history of NSSI.¹²

Factors that protect against suicidal behavior include:

• a good parent-child relationship
• strong cultural or religious values
• an intact family
• a sense of connection with peer group and community.¹³

No studies have determined protective factors for NSSI.

Table 1

Characteristics of teens who harm themselves

CASE CONTINUED: ‘No point in living’

As Josh becomes less guarded, he says that he sees “no point in living” without his girlfriend. He thought the only way to feel better was to “get high,” but this left him feeling even more despondent and anxious. He wrote a suicide note, but after cutting himself he was unsure he wanted to die. Josh says that when he feels depressed he can’t talk to his parents because “they wouldn’t understand and don’t care.”

Assessing self-harming adolescents

Distinguishing between suicidal behaviors and NSSI can be challenging (Table 2). Identifying risk factors for adolescent suicidal behavior must be coupled with a thorough psychiatric evaluation. If possible, interview the adolescent alone and obtain collateral information from parents, family members, teachers, caseworkers, probation officers, and others as needed. Also examine family interactions because conflicts and communication problems could undermine the teenager’s safety.

Consider using standardized measures of suicidal intentions such as the Scale for Suicidal Ideation (SSI).¹⁴ Although the SSI was developed for adults, a large case-control study validated the scale’s use in adolescent psychiatric outpatients and students.¹⁵ In addition to assessing subjective reports of suicidal intent, the SSI also takes into account objective indicators of increased risk, such as planning an attempt, hiding details from others, and making preparations for death.

Questions to ask. After a self-harm incident, it may be helpful to begin the psychiatric interview with a general question such as, “What happened that led you to hurt yourself?” This does not categorize the act as suicidal and allows the adolescent to describe it in his or her own words. Shea¹⁶ suggests normalizing the act by assuming that self-harm occurred, rather than making the patient admit to it. For example, an interviewer might say “Many people I know who are hurting inside also try to hurt their body; how often do you do that?”

Inquire about suicidal intent in a few ways. For example, first ask, “Do you ever wish you were dead?” and follow up with, “Would you ever do anything to try to make yourself dead?” Asking about suicidal thoughts does not increase suicidal thoughts or behavior.^10,17

Reviewing thoughts and feelings leading up to a self-harm act can help identify triggers, coping difficulties, and issues to address in treatment. This behavioral analysis can be completed using the mnemonic ABC:

• Antecedents (situations or stressors leading to self-harming thoughts or actions)
• Behavior characteristics (frequency, intensity, and duration of self-harming)
• Consequences (eg, emotional relief, care and attention from others).

The results of this analysis could suggest treatment strategies, such as cognitive restructuring or techniques for decreasing feelings of distress.

The Risk of Suicide Questionnaire, which is designed for adolescents, asks:¹⁸

• Are you here because you tried to hurt yourself?
• In the past week, have you been having thoughts about killing yourself?
• Have you ever tried to hurt yourself in the past?
• Has something very stressful happened to you in the past few weeks?

Research has yet to determine whether this simple, rapid screen accurately identifies the need for psychiatric hospitalization or risk for suicidal outcomes. Although clinician- and self-administered suicide questionnaires may be useful for screening large populations, they are not a substitute for a thorough clinical assessment.

Inpatient or outpatient? When evaluating self-harming adolescents, first determine if they are in imminent danger of suicide and if more intensive services, such as hospitalization, are needed to maintain safety. Inpatient psychiatric services are appropriate for adolescents with suicidal thoughts or self-harm behaviors in addition to acute psychiatric disorders, significant substance abuse, serious medical issues, poor social supports, or inability to be managed safely as an outpatient.¹⁹ See the Table for a list of additional factors to consider.

Consent for treatment may be required because many self-harming adolescents do not present with life-threatening symptoms. Laws regarding consent vary among states. In some jurisdictions, patients age ≥15 can consent to mental health treatment without parents’ knowledge or consent. If an adolescent is in imminent danger and cannot voluntarily consent to treatment, physicians can initiate mental health “holds.” Some states allow registered nurses, psychologists, licensed social workers, and others to initiate mental health holds.

Table 2

Strategies for assessing adolescent self-harm

CASE CONTINUED: Inpatient treatment

After the interview Josh says he still feels that “there is no point in living” and he cannot develop an adequate safety plan with his family. He is hospitalized to maintain safety, improve his coping skills and communication with his family, and mobilize safety plans, social supports, and follow-up care.

Maintaining safety

Psychosocial treatments for suicidal behaviors and NSSI are similar because with both, the priority is to help the patient maintain safety. This may include:

• developing a collaborative safety plan with family
• increasing monitoring
• removing access to firearms or other lethal means
• helping the adolescent to develop alternate, safer coping methods.

Many clinicians rely on no-harm contracts or agreements; however, there is no evidence that they are effective.²⁰ The American Psychiatric Association recommends against using no-harm contracts with patients who are new, in an emergency setting, using substances, agitated, psychotic, or impulsive.²¹ Instead, clinicians, adolescents, and families can discuss specific steps the patient can take to remain safe. This collaborative plan should identify situations likely to trigger self-harming impulses; adaptive ways the teenager can cope, such as taking a nap or jogging; methods for communicating distress to family members and other helpers; and places to go for help, such as an emergency room. These safety plans should draw on the patient’s internal and external resources.

CASE CONTINUED: Strengthening relationships

While in the hospital, Josh finds it helpful to use a 0-to-10 scale to measure his distress and let his family know the intensity of his feelings. He identifies situations when he felt like hurting himself, such as being humiliated in math class. Josh learns about cognitive distortions—such as “they don’t care” and “there is no point in living”—and discusses methods for managing his feelings if he encounters further disappointments. His parents become more attentive when Josh explains his feelings, which allows the family to develop a collaborative safety plan. Josh decides to strengthen friendships he had been neglecting and agrees to attend a substance abuse treatment program.

Psychosocial treatment

In addition to maintaining safety, treatment goals for self-harming adolescents include:

• managing underlying psychiatric disorders
• identifying triggers for self-injurious acts
• improving family relationships
• developing better communication and coping skills.

Improving affective language skills, acquiring frustration tolerance, and learning alternatives to self-injury are key to strengthening coping abilities. Address problem-solving skills because self-harming adolescents often lack these abilities.²²

Treatment of self-harming adolescents often consists of cognitive-behavioral therapy (CBT)²³ or dialectical behavior therapy (DBT).²⁴ CBT involves examining cognitive distortions or otherwise unhealthy beliefs about oneself, others, and life in general, focusing specifically on thoughts the patient has immediately before engaging in self-harm. DBT also integrates emotion regulation training and mindfulness. A review of 28 studies found these therapies effectively reduced self-harm behaviors in adults.²⁵ However, few studies have examined these therapies’ efficacy in self-harming adolescents.

Pharmacotherapy

Psychopharmacology should focus on treating underlying psychiatric disorders. No medications are specifically effective for treating suicidal thoughts, suicidal behaviors, or NSSI. Some evidence suggests that antidepressants may trigger suicidal thoughts in a small proportion of youth,²⁶ but the benefits of antidepressants outweigh the risk of suicidal thoughts.²⁷ When prescribing antidepressants, inform patients and their parents of possible adverse reactions and monitor the patient regularly.²⁸

Take precautions when prescribing medication for self-harming adolescents. For example, benzodiazepines may cause disinhibition, and larger quantities of medication could be lethal in an overdose. If possible, arrange for parents or guardians to monitor medication use.

What to document

Good documentation is especially helpful when an adolescent requires involuntary commitment or is discharged home. Involuntary commitment is based on legal interpretation of 3 circumstances—danger to self, danger to others, or gravely disabled—in which safety concerns may override an individual’s civil rights. If involuntary commitment is needed, a physician must clarify how the youth meets ≥1 of these criteria. If the adolescent is discharged, document that the patient is not an imminent danger to self or others and why you made this determination. Also note that follow-up services and a safety plan are in place, a parent will monitor safety issues and remove firearms and other lethal means from the home, and acute conflicts have been resolved. Other details, such as using the patient’s words to describe reasons for living, can be helpful.

Related Resources

• National Alliance for the Mentally Ill. National Helpline. 800-950-6264.
• American Foundation for Suicide Prevention. www.afsp.org.
• American Association of Suicidology. www.suicidology.org.
• National Suicide Prevention Lifeline. 800-273-TALK (8255).

Acknowledgements

The authors wish to thank students Scott Schubert from Regis University, Denver, CO, and Emily Peterson from Beloit College, Beloit, WI, for their help in preparing the manuscript.

Disclosure

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

Table

Hospitalization or home? Acute crisis planning for self-harming youth

Discuss this article at http://currentpsychiatry.blogspot.com/2010/08/adolescents-who-self-harm.html#comments

Josh, age 16, gets poor grades in school and occasionally smokes marijuana and abuses inhalants. After his girlfriend breaks up with him, he cuts his wrist with a hunting knife. While bleeding profusely, Josh calls his mother at work, who calls 911. The cut is deep and requires sutures. Josh says he did not try to kill himself; he only wanted to carve his girlfriend’s initials into his wrist to show his love for her.

When treating teenagers with self-harming thoughts and behavior, it may be difficult to distinguish suicide attempts from self-injury without intent to die. Understanding adolescent self-harm, suicide risk assessment, and treatment options guides clinicians to appropriate interventions. Recognizing the need for aggressive treatment—including psychiatric hospitalization—is essential to keeping self-harming teenagers safe.

Suicidal vs nonsuicidal self-harm

Suicidal behavior involves intent to end one’s life and includes ideation (thoughts) and actions (non fatal or fatal attempts).¹ Nonsuicidal self-injury (NSSI) involves socially unacceptable, self-inflicted harm to one’s body without intent to die.²

Suicide is the third leading cause of death among youths age 12 to 19, claiming almost 2,000 lives each year.³ Nearly 1 in 5 (17%) U.S. high school students has suicidal thoughts each year, and almost 1 in 10 (8%) attempts suicide.⁴

Studies report a 13% to 23% lifetime prevalence of NSSI.⁵ These behaviors often begin between age 13 to 15.⁶ Cutting and hitting are the most common forms of NSSI; other methods include burning, scratching, and interfering with wound healing. Most teens who harm themselves without suicidal intent report that they feel little or no pain during the act.⁵ Unlike suicide attempts, NSSI can be viewed as a means to stay alive. Many adolescents injure themselves to cope with overwhelming feelings that can produce suicidal thoughts. Self-injury may distract the adolescent from painful emotions, reduce tensions, or penetrate numbness.⁷

Teenagers who hurt themselves but do not intend to die are at high risk for suicide and suicide attempts. Adolescents who engage in NSSI are more likely to experience suicidal behaviors, and vice versa.⁸ In a large study, 70% of adolescents who engaged in NSSI had made at least 1 suicide attempt and 55% made multiple suicide attempts.² Current suicidal ideation is a risk factor for suicide, and a past suicide attempt is the strongest predictor of future suicidal behavior.⁹

Risk factors for suicidal behavior and NSSI overlap (Table 1)^2,5,6,10,11 and include:

• depression
• substance use
• anxiety
• impulsive aggression
• history of childhood trauma.

Many teens who engage in NSSI report depression.² A history of psychiatric illness—especially depression—increases the likelihood of adolescent suicide.⁸ A study comparing adolescents who engaged in NSSI with those who attempted suicide found that both groups reported similar levels of suicidal ideation and depressive symptoms.⁶ However, adolescents with a history of NSSI and attempted suicide reported higher levels of suicidal ideation and fewer reasons for living than those who attempted suicide but have no history of NSSI.¹²

Factors that protect against suicidal behavior include:

• a good parent-child relationship
• strong cultural or religious values
• an intact family
• a sense of connection with peer group and community.¹³

No studies have determined protective factors for NSSI.

Table 1

Characteristics of teens who harm themselves

CASE CONTINUED: ‘No point in living’

As Josh becomes less guarded, he says that he sees “no point in living” without his girlfriend. He thought the only way to feel better was to “get high,” but this left him feeling even more despondent and anxious. He wrote a suicide note, but after cutting himself he was unsure he wanted to die. Josh says that when he feels depressed he can’t talk to his parents because “they wouldn’t understand and don’t care.”

Assessing self-harming adolescents

Distinguishing between suicidal behaviors and NSSI can be challenging (Table 2). Identifying risk factors for adolescent suicidal behavior must be coupled with a thorough psychiatric evaluation. If possible, interview the adolescent alone and obtain collateral information from parents, family members, teachers, caseworkers, probation officers, and others as needed. Also examine family interactions because conflicts and communication problems could undermine the teenager’s safety.

Consider using standardized measures of suicidal intentions such as the Scale for Suicidal Ideation (SSI).¹⁴ Although the SSI was developed for adults, a large case-control study validated the scale’s use in adolescent psychiatric outpatients and students.¹⁵ In addition to assessing subjective reports of suicidal intent, the SSI also takes into account objective indicators of increased risk, such as planning an attempt, hiding details from others, and making preparations for death.

Questions to ask. After a self-harm incident, it may be helpful to begin the psychiatric interview with a general question such as, “What happened that led you to hurt yourself?” This does not categorize the act as suicidal and allows the adolescent to describe it in his or her own words. Shea¹⁶ suggests normalizing the act by assuming that self-harm occurred, rather than making the patient admit to it. For example, an interviewer might say “Many people I know who are hurting inside also try to hurt their body; how often do you do that?”

Inquire about suicidal intent in a few ways. For example, first ask, “Do you ever wish you were dead?” and follow up with, “Would you ever do anything to try to make yourself dead?” Asking about suicidal thoughts does not increase suicidal thoughts or behavior.^10,17

Reviewing thoughts and feelings leading up to a self-harm act can help identify triggers, coping difficulties, and issues to address in treatment. This behavioral analysis can be completed using the mnemonic ABC:

• Antecedents (situations or stressors leading to self-harming thoughts or actions)
• Behavior characteristics (frequency, intensity, and duration of self-harming)
• Consequences (eg, emotional relief, care and attention from others).

The results of this analysis could suggest treatment strategies, such as cognitive restructuring or techniques for decreasing feelings of distress.

The Risk of Suicide Questionnaire, which is designed for adolescents, asks:¹⁸

• Are you here because you tried to hurt yourself?
• In the past week, have you been having thoughts about killing yourself?
• Have you ever tried to hurt yourself in the past?
• Has something very stressful happened to you in the past few weeks?

Research has yet to determine whether this simple, rapid screen accurately identifies the need for psychiatric hospitalization or risk for suicidal outcomes. Although clinician- and self-administered suicide questionnaires may be useful for screening large populations, they are not a substitute for a thorough clinical assessment.

Inpatient or outpatient? When evaluating self-harming adolescents, first determine if they are in imminent danger of suicide and if more intensive services, such as hospitalization, are needed to maintain safety. Inpatient psychiatric services are appropriate for adolescents with suicidal thoughts or self-harm behaviors in addition to acute psychiatric disorders, significant substance abuse, serious medical issues, poor social supports, or inability to be managed safely as an outpatient.¹⁹ See the Table for a list of additional factors to consider.

Consent for treatment may be required because many self-harming adolescents do not present with life-threatening symptoms. Laws regarding consent vary among states. In some jurisdictions, patients age ≥15 can consent to mental health treatment without parents’ knowledge or consent. If an adolescent is in imminent danger and cannot voluntarily consent to treatment, physicians can initiate mental health “holds.” Some states allow registered nurses, psychologists, licensed social workers, and others to initiate mental health holds.

Table 2

Strategies for assessing adolescent self-harm

CASE CONTINUED: Inpatient treatment

After the interview Josh says he still feels that “there is no point in living” and he cannot develop an adequate safety plan with his family. He is hospitalized to maintain safety, improve his coping skills and communication with his family, and mobilize safety plans, social supports, and follow-up care.

Maintaining safety

Psychosocial treatments for suicidal behaviors and NSSI are similar because with both, the priority is to help the patient maintain safety. This may include:

• developing a collaborative safety plan with family
• increasing monitoring
• removing access to firearms or other lethal means
• helping the adolescent to develop alternate, safer coping methods.

Many clinicians rely on no-harm contracts or agreements; however, there is no evidence that they are effective.²⁰ The American Psychiatric Association recommends against using no-harm contracts with patients who are new, in an emergency setting, using substances, agitated, psychotic, or impulsive.²¹ Instead, clinicians, adolescents, and families can discuss specific steps the patient can take to remain safe. This collaborative plan should identify situations likely to trigger self-harming impulses; adaptive ways the teenager can cope, such as taking a nap or jogging; methods for communicating distress to family members and other helpers; and places to go for help, such as an emergency room. These safety plans should draw on the patient’s internal and external resources.

CASE CONTINUED: Strengthening relationships

While in the hospital, Josh finds it helpful to use a 0-to-10 scale to measure his distress and let his family know the intensity of his feelings. He identifies situations when he felt like hurting himself, such as being humiliated in math class. Josh learns about cognitive distortions—such as “they don’t care” and “there is no point in living”—and discusses methods for managing his feelings if he encounters further disappointments. His parents become more attentive when Josh explains his feelings, which allows the family to develop a collaborative safety plan. Josh decides to strengthen friendships he had been neglecting and agrees to attend a substance abuse treatment program.

Psychosocial treatment

In addition to maintaining safety, treatment goals for self-harming adolescents include:

• managing underlying psychiatric disorders
• identifying triggers for self-injurious acts
• improving family relationships
• developing better communication and coping skills.

Improving affective language skills, acquiring frustration tolerance, and learning alternatives to self-injury are key to strengthening coping abilities. Address problem-solving skills because self-harming adolescents often lack these abilities.²²

Treatment of self-harming adolescents often consists of cognitive-behavioral therapy (CBT)²³ or dialectical behavior therapy (DBT).²⁴ CBT involves examining cognitive distortions or otherwise unhealthy beliefs about oneself, others, and life in general, focusing specifically on thoughts the patient has immediately before engaging in self-harm. DBT also integrates emotion regulation training and mindfulness. A review of 28 studies found these therapies effectively reduced self-harm behaviors in adults.²⁵ However, few studies have examined these therapies’ efficacy in self-harming adolescents.

Pharmacotherapy

Psychopharmacology should focus on treating underlying psychiatric disorders. No medications are specifically effective for treating suicidal thoughts, suicidal behaviors, or NSSI. Some evidence suggests that antidepressants may trigger suicidal thoughts in a small proportion of youth,²⁶ but the benefits of antidepressants outweigh the risk of suicidal thoughts.²⁷ When prescribing antidepressants, inform patients and their parents of possible adverse reactions and monitor the patient regularly.²⁸

Take precautions when prescribing medication for self-harming adolescents. For example, benzodiazepines may cause disinhibition, and larger quantities of medication could be lethal in an overdose. If possible, arrange for parents or guardians to monitor medication use.

What to document

Good documentation is especially helpful when an adolescent requires involuntary commitment or is discharged home. Involuntary commitment is based on legal interpretation of 3 circumstances—danger to self, danger to others, or gravely disabled—in which safety concerns may override an individual’s civil rights. If involuntary commitment is needed, a physician must clarify how the youth meets ≥1 of these criteria. If the adolescent is discharged, document that the patient is not an imminent danger to self or others and why you made this determination. Also note that follow-up services and a safety plan are in place, a parent will monitor safety issues and remove firearms and other lethal means from the home, and acute conflicts have been resolved. Other details, such as using the patient’s words to describe reasons for living, can be helpful.

Related Resources

• National Alliance for the Mentally Ill. National Helpline. 800-950-6264.
• American Foundation for Suicide Prevention. www.afsp.org.
• American Association of Suicidology. www.suicidology.org.
• National Suicide Prevention Lifeline. 800-273-TALK (8255).

Acknowledgements

The authors wish to thank students Scott Schubert from Regis University, Denver, CO, and Emily Peterson from Beloit College, Beloit, WI, for their help in preparing the manuscript.

Disclosure

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

Table

Hospitalization or home? Acute crisis planning for self-harming youth

After testing positive for cocaine on a recent court-mandated urine drug screen, Mr. M, age 49, is referred by his parole officer for psychiatric and substance abuse treatment. Mr. M has bipolar I disorder and alcohol, cocaine, and opioid dependence. He says he has been hospitalized or incarcerated at least once each year for the past 22 years. Mr. M has seen numerous psychiatrists as an outpatient, but rarely for more than 2 to 3 months and has not taken any psychotropics for more than 5 months.

Approximately 1 week before his recent urine drug screen, Mr. M became euphoric, stopped sleeping for several days, and spent $2,000 on cocaine and “escorts.” He reports that each day he smokes 30 cigarettes and drinks 1 pint of liquor and prefers to use cocaine and opiates by IV injection. Several years ago Mr. M was diagnosed with hepatitis C virus (HCV) but received no further workup or treatment. Although he denies manic or psychotic symptoms, Mr. M is observed speaking to unseen others in the waiting room and has difficulty remaining still during his interview. His chief concern is insomnia, stating, “Doc, I just need something to help me sleep.”

The high prevalence of substance use disorders (SUDs) in persons with bipolar disorder (BD) is well documented.^1,2 Up to 60% of bipolar patients develop an SUD at some point in their lives.^3-5 Alcohol use disorders are particularly common among BD patients, with a lifetime prevalence of roughly 50%.^2-5 Recent epidemiologic data indicate that 38% of persons with bipolar I disorder and 19% of those with bipolar II disorder meet criteria for alcohol dependence.⁵ Comorbid SUDs in patients with BD are associated with:

• poor treatment compliance
• longer and more frequent mood episodes
• more mixed episodes
• more hospitalizations
• more frequent suicide attempts.^1,2

The impact of co-occurring SUDs on suicidality is particularly high among those with bipolar I disorder.⁶

Frequently referred to as “dual diagnosis” conditions, co-occurring BD and SUDs may be more accurately envisioned as multi-morbid, rather than comorbid, illnesses. Data from the Stanley Foundation Bipolar Network suggest that 42% of BD patients have a lifetime history of ≥2 comorbid axis I disorders.⁷ Rates of generalized anxiety disorder, panic disorder, and posttraumatic stress disorder are particularly high in BD patients with co-occurring SUDs.^8,9 In addition, the presence of 1 SUD may mark the presence of other SUDs; for example, alcohol dependence is strongly associated with polysubstance abuse, especially in females with BD.⁸ Furthermore, medical comorbidities that impact treatment decisions also are highly prevalent in BD patients with comorbid SUDs.¹⁰ In particular, HCV rates are higher in persons with BD compared with the general population,¹¹ and are >5 times as likely in bipolar patients with co-occurring SUDs.¹²

Unfortunately, limited treatment research guides clinical management of comorbid BD and SUDs.^2,13,14 Clinical trials of medications for BD traditionally have excluded patients with SUDs, and persons with serious mental illness usually are ineligible for SUD treatment studies.¹³ Furthermore, the few randomized controlled trials (RCTs) conducted in persons with both illnesses have been constrained by relatively small sample sizes and low retention rates. In the absence of a definitive consensus for optimal treatment, this article outlines general clinical considerations and an integrated approach to assessing and managing this complex patient population.

Birds of a feather

Multiple hypotheses try to account for the high rate of SUDs in patients with BD (Table 1), but none fully explain the complex interaction observed clinically.^14,15 Substance dependence and BD are chronic remitting/relapsing disorders with heterogeneous presentations and highly variable natural histories. As with SUDs, BD may go undiagnosed and untreated for years, coming to clinical attention only after substantial disease progression.¹⁶

The fluctuating illness course in BD and SUDs makes diagnosis and treatment difficult. Symptomatic periods often are interrupted by spontaneous remissions and longer—although usually temporary—periods of perceived control. Both BD and SUDs may be associated with profound mood instability, increased impulsivity, altered responsiveness to reward, and impaired executive function.¹⁷ Finally, the high degree of heritability in BD¹⁸ and many SUDs¹⁹ may make treatment engagement more difficult if either disorder is present in multiple family members because of:

• potential for greater clinical severity
• reduced psychosocial resources
• altered familial behavioral norms that may impede the patient’s recognition of illness.

Denial of illness is a critical symptom that may fluctuate with disease course in both disorders. Persons with BD or SUDs may be least likely to recognize that they are ill during periods of highest symptom severity. Accordingly, treatment adherence in patients with either disorder may be limited at baseline and decline further when the 2 illnesses co-occur.²⁰

Involvement in the criminal justice system and medical comorbidities, particularly HCV, also complicate diagnosis and treatment of BD patients with SUDs. For more information about these topics, see Box1 and Box2.

Table 1

Why is substance abuse so prevalent among bipolar patients?

Integrated clinical management

Assessment. Although not intended to be comprehensive, suggestions for routine assessment of patients with suspected SUDs and/or BD are listed in Table 2. Because clinicians may encounter dual diagnosis patients in general psychiatric clinics or specialty (addiction or mood disorder) clinics, it is useful to obtain a thorough substance use history in all patients with known or suspected BD as well as a thorough history of hypomania/mania and depression in all patients with addictive disorders. BD diagnoses by self-report or chart history in patients with SUDs should be considered cautiously because BD often is overdiagnosed in persons engaged in active substance abuse or experiencing withdrawal.²¹ If past or present mood symptoms and substance use have co-occurred, further focused assessment of mood symptoms before alcohol and drug use or during extended periods of abstinence are necessary to make the diagnosis of bipolar disorder with confidence. Family history of SUDs and/or BD are neither necessary nor sufficient for either diagnosis; however, collateral information from family or significant others could help make the diagnosis and may identify aids and obstacles to treatment planning and engagement.

When a patient’s clinical history strongly supports the diagnoses of BD and co-occurring SUDs, more detailed inquiry is warranted. Determining the patient’s age at onset of each disorder may have prognostic value because onset of mania before SUDs developed, especially in adolescence, may predict a more severe course of both illnesses.²² A complete alcohol use history should include routine questioning about past withdrawal. Previous withdrawal seizures in an actively drinking BD patient may tip the balance toward adding an anticonvulsant for mood stabilization. A thorough SUD history should elicit information about polysubstance abuse or dependence and include screening for injection drug use and other risk factors for HCV and human immunodeficiency virus (HIV), such as hypersexuality during manic or hypomanic episodes. Document the date of the last screening for HCV/ HIV in BD patients at high risk of infection. The U.S. Centers for Disease Control and Prevention recommends that all patients at high risk for HIV consider voluntary screening at least annually.²³

Assess your patient’s historical and ongoing alcohol and other drug abuse at the initial visit, and continue to monitor substance use at all subsequent visits, especially in patients with HCV. When feasible, order urine drug screening and laboratory testing for alcohol use biomarkers such as carbohydrate-deficient transferrin and gamma-glutamyltransferase to supplement self-report data, especially in patients with poor insight or low motivation. Assess suicidal ideation and any changes in suicide risk factors at every visit.

Treatment. No biologic therapies have been FDA-approved for treating patients with co-occurring BD and SUDs. Comorbid SUDs in BD patients—as well as rapid cycling and mixed mood episodes, both of which are more common in patients with comorbid SUDs—predict poor response to lithium.¹⁷ However, the evidence base for optimal pharmacotherapy remains extremely limited. Published double-blind, placebo-controlled RCTs in persons with BD and co-morbid SUDs are limited to only 1 trial each of lithium, carbamazepine, quetiapine, and naltrexone, and 2 comparisons of lithium plus divalproex vs lithium alone (Table 3).^24-29

Salloum et al²⁴ reported that bipolar I disorder patients with alcohol dependence who received divalproex plus lithium as maintenance treatment had fewer heavy drinking days and fewer drinks per heavy drinking day than those receiving lithium plus placebo. However, the addition of divalproex did not improve manic or depressive symptoms, and depression remission rates remained relatively low in both groups. A recent 6-month study comparing lithium vs lithium plus divalproex in patients with SUDs and rapid-cycling BD found no additional benefit of divalproex over lithium monotherapy in retention, mood, or substance use outcomes.²⁸ However, modest evidence that anticonvulsants such as valproic acid and carbamazepine may help treat acute alcohol withdrawal² could support their preferential use as mood stabilizers over lithium in actively drinking BD patients.

Research underscores the difficulty in keeping dual diagnosis patients in treatment. Salloum et al²⁴ reported that only one-third of randomized subjects completed the 24-week study. In the Kemp study,²⁸ 79% of 149 recruited subjects failed to complete the lead-in stabilization phase. Of the 31 remaining subjects who were randomized to lithium or lithium/divalproex combination, only 8 (26% of those randomized, 5% of those recruited) completed the 6-month trial.

Substance abuse is associated with significantly decreased treatment adherence in persons with BD²⁰ and may affect medication choice. For example, caution may be warranted in the use of lamotrigine in substance-abusing patients with poor adherence because re-titration from the starting dose is recommended if the medicine has been missed for a consecutive period exceeding 5 half-lives of the drug.³⁰

Notable progress has been made in developing psychosocial treatments for comorbid SUDs and BD. Integrated group therapy has been designed to address the 2 disorders simultaneously by emphasizing the relationship between the disorders and highlighting similarities in cognitive and behavioral change that promote recovery in both.³¹ In a recent well-controlled RCT, this approach reduced alcohol and other drug use to approximately one-half the levels observed in those who received only group drug counseling.³¹

Research suggests that an integrated approach that encompasses psychiatric, medical, psychosocial, and legal dimensions simultaneously may be most effective. For patients such as Mr. M, this would include aggressively treating mood symptoms while employing motivational interviewing techniques to improve engagement in substance dependence treatment. If possible, involving family members, parole officials, housing agencies, and other public assistance workers in the treatment plan may increase treatment adherence and reduce loss of contact during illness exacerbations. Stabilization of substance use and psychiatric morbidity should be accompanied by timely evaluation of HCV and other medical comorbidities in order to improve long-term prognosis.

Table 2

Assessing patients you suspect have comorbid BD and SUDs

Table 3

Pharmacotherapy for bipolar disorder and co-occurring SUDs

Related Resources

• International Society for Bipolar Disorders. www.isbd.org.
• National Institute on Alcohol Abuse and Alcoholism. www.niaaa.nih.gov.
• National Institute on Drug Abuse. www.nida.nih.gov.

Drug Brand Names

• Carbamazepine • Tegretol
• Divalproex/valproic acid • Depakote
• Lamotrigine • Lamictal
• Lithium • Lithobid
• Naltrexone • ReVia
• Quetiapine • Seroquel

Disclosure

Dr. Tolliver receives research grant funding from Forest Laboratories and the National Institute on Alcohol Abuse and Alcoholism. Neither source influenced the content or submission of this manuscript.

Box 1

Box 2

After testing positive for cocaine on a recent court-mandated urine drug screen, Mr. M, age 49, is referred by his parole officer for psychiatric and substance abuse treatment. Mr. M has bipolar I disorder and alcohol, cocaine, and opioid dependence. He says he has been hospitalized or incarcerated at least once each year for the past 22 years. Mr. M has seen numerous psychiatrists as an outpatient, but rarely for more than 2 to 3 months and has not taken any psychotropics for more than 5 months.

Approximately 1 week before his recent urine drug screen, Mr. M became euphoric, stopped sleeping for several days, and spent $2,000 on cocaine and “escorts.” He reports that each day he smokes 30 cigarettes and drinks 1 pint of liquor and prefers to use cocaine and opiates by IV injection. Several years ago Mr. M was diagnosed with hepatitis C virus (HCV) but received no further workup or treatment. Although he denies manic or psychotic symptoms, Mr. M is observed speaking to unseen others in the waiting room and has difficulty remaining still during his interview. His chief concern is insomnia, stating, “Doc, I just need something to help me sleep.”

The high prevalence of substance use disorders (SUDs) in persons with bipolar disorder (BD) is well documented.^1,2 Up to 60% of bipolar patients develop an SUD at some point in their lives.^3-5 Alcohol use disorders are particularly common among BD patients, with a lifetime prevalence of roughly 50%.^2-5 Recent epidemiologic data indicate that 38% of persons with bipolar I disorder and 19% of those with bipolar II disorder meet criteria for alcohol dependence.⁵ Comorbid SUDs in patients with BD are associated with:

• poor treatment compliance
• longer and more frequent mood episodes
• more mixed episodes
• more hospitalizations
• more frequent suicide attempts.^1,2

The impact of co-occurring SUDs on suicidality is particularly high among those with bipolar I disorder.⁶

Frequently referred to as “dual diagnosis” conditions, co-occurring BD and SUDs may be more accurately envisioned as multi-morbid, rather than comorbid, illnesses. Data from the Stanley Foundation Bipolar Network suggest that 42% of BD patients have a lifetime history of ≥2 comorbid axis I disorders.⁷ Rates of generalized anxiety disorder, panic disorder, and posttraumatic stress disorder are particularly high in BD patients with co-occurring SUDs.^8,9 In addition, the presence of 1 SUD may mark the presence of other SUDs; for example, alcohol dependence is strongly associated with polysubstance abuse, especially in females with BD.⁸ Furthermore, medical comorbidities that impact treatment decisions also are highly prevalent in BD patients with comorbid SUDs.¹⁰ In particular, HCV rates are higher in persons with BD compared with the general population,¹¹ and are >5 times as likely in bipolar patients with co-occurring SUDs.¹²

Unfortunately, limited treatment research guides clinical management of comorbid BD and SUDs.^2,13,14 Clinical trials of medications for BD traditionally have excluded patients with SUDs, and persons with serious mental illness usually are ineligible for SUD treatment studies.¹³ Furthermore, the few randomized controlled trials (RCTs) conducted in persons with both illnesses have been constrained by relatively small sample sizes and low retention rates. In the absence of a definitive consensus for optimal treatment, this article outlines general clinical considerations and an integrated approach to assessing and managing this complex patient population.

Birds of a feather

Multiple hypotheses try to account for the high rate of SUDs in patients with BD (Table 1), but none fully explain the complex interaction observed clinically.^14,15 Substance dependence and BD are chronic remitting/relapsing disorders with heterogeneous presentations and highly variable natural histories. As with SUDs, BD may go undiagnosed and untreated for years, coming to clinical attention only after substantial disease progression.¹⁶

The fluctuating illness course in BD and SUDs makes diagnosis and treatment difficult. Symptomatic periods often are interrupted by spontaneous remissions and longer—although usually temporary—periods of perceived control. Both BD and SUDs may be associated with profound mood instability, increased impulsivity, altered responsiveness to reward, and impaired executive function.¹⁷ Finally, the high degree of heritability in BD¹⁸ and many SUDs¹⁹ may make treatment engagement more difficult if either disorder is present in multiple family members because of:

• potential for greater clinical severity
• reduced psychosocial resources
• altered familial behavioral norms that may impede the patient’s recognition of illness.

Denial of illness is a critical symptom that may fluctuate with disease course in both disorders. Persons with BD or SUDs may be least likely to recognize that they are ill during periods of highest symptom severity. Accordingly, treatment adherence in patients with either disorder may be limited at baseline and decline further when the 2 illnesses co-occur.²⁰

Involvement in the criminal justice system and medical comorbidities, particularly HCV, also complicate diagnosis and treatment of BD patients with SUDs. For more information about these topics, see Box1 and Box2.

Table 1

Why is substance abuse so prevalent among bipolar patients?

Integrated clinical management

Assessment. Although not intended to be comprehensive, suggestions for routine assessment of patients with suspected SUDs and/or BD are listed in Table 2. Because clinicians may encounter dual diagnosis patients in general psychiatric clinics or specialty (addiction or mood disorder) clinics, it is useful to obtain a thorough substance use history in all patients with known or suspected BD as well as a thorough history of hypomania/mania and depression in all patients with addictive disorders. BD diagnoses by self-report or chart history in patients with SUDs should be considered cautiously because BD often is overdiagnosed in persons engaged in active substance abuse or experiencing withdrawal.²¹ If past or present mood symptoms and substance use have co-occurred, further focused assessment of mood symptoms before alcohol and drug use or during extended periods of abstinence are necessary to make the diagnosis of bipolar disorder with confidence. Family history of SUDs and/or BD are neither necessary nor sufficient for either diagnosis; however, collateral information from family or significant others could help make the diagnosis and may identify aids and obstacles to treatment planning and engagement.

When a patient’s clinical history strongly supports the diagnoses of BD and co-occurring SUDs, more detailed inquiry is warranted. Determining the patient’s age at onset of each disorder may have prognostic value because onset of mania before SUDs developed, especially in adolescence, may predict a more severe course of both illnesses.²² A complete alcohol use history should include routine questioning about past withdrawal. Previous withdrawal seizures in an actively drinking BD patient may tip the balance toward adding an anticonvulsant for mood stabilization. A thorough SUD history should elicit information about polysubstance abuse or dependence and include screening for injection drug use and other risk factors for HCV and human immunodeficiency virus (HIV), such as hypersexuality during manic or hypomanic episodes. Document the date of the last screening for HCV/ HIV in BD patients at high risk of infection. The U.S. Centers for Disease Control and Prevention recommends that all patients at high risk for HIV consider voluntary screening at least annually.²³

Assess your patient’s historical and ongoing alcohol and other drug abuse at the initial visit, and continue to monitor substance use at all subsequent visits, especially in patients with HCV. When feasible, order urine drug screening and laboratory testing for alcohol use biomarkers such as carbohydrate-deficient transferrin and gamma-glutamyltransferase to supplement self-report data, especially in patients with poor insight or low motivation. Assess suicidal ideation and any changes in suicide risk factors at every visit.

Treatment. No biologic therapies have been FDA-approved for treating patients with co-occurring BD and SUDs. Comorbid SUDs in BD patients—as well as rapid cycling and mixed mood episodes, both of which are more common in patients with comorbid SUDs—predict poor response to lithium.¹⁷ However, the evidence base for optimal pharmacotherapy remains extremely limited. Published double-blind, placebo-controlled RCTs in persons with BD and co-morbid SUDs are limited to only 1 trial each of lithium, carbamazepine, quetiapine, and naltrexone, and 2 comparisons of lithium plus divalproex vs lithium alone (Table 3).^24-29

Salloum et al²⁴ reported that bipolar I disorder patients with alcohol dependence who received divalproex plus lithium as maintenance treatment had fewer heavy drinking days and fewer drinks per heavy drinking day than those receiving lithium plus placebo. However, the addition of divalproex did not improve manic or depressive symptoms, and depression remission rates remained relatively low in both groups. A recent 6-month study comparing lithium vs lithium plus divalproex in patients with SUDs and rapid-cycling BD found no additional benefit of divalproex over lithium monotherapy in retention, mood, or substance use outcomes.²⁸ However, modest evidence that anticonvulsants such as valproic acid and carbamazepine may help treat acute alcohol withdrawal² could support their preferential use as mood stabilizers over lithium in actively drinking BD patients.

Research underscores the difficulty in keeping dual diagnosis patients in treatment. Salloum et al²⁴ reported that only one-third of randomized subjects completed the 24-week study. In the Kemp study,²⁸ 79% of 149 recruited subjects failed to complete the lead-in stabilization phase. Of the 31 remaining subjects who were randomized to lithium or lithium/divalproex combination, only 8 (26% of those randomized, 5% of those recruited) completed the 6-month trial.

Substance abuse is associated with significantly decreased treatment adherence in persons with BD²⁰ and may affect medication choice. For example, caution may be warranted in the use of lamotrigine in substance-abusing patients with poor adherence because re-titration from the starting dose is recommended if the medicine has been missed for a consecutive period exceeding 5 half-lives of the drug.³⁰

Notable progress has been made in developing psychosocial treatments for comorbid SUDs and BD. Integrated group therapy has been designed to address the 2 disorders simultaneously by emphasizing the relationship between the disorders and highlighting similarities in cognitive and behavioral change that promote recovery in both.³¹ In a recent well-controlled RCT, this approach reduced alcohol and other drug use to approximately one-half the levels observed in those who received only group drug counseling.³¹

Research suggests that an integrated approach that encompasses psychiatric, medical, psychosocial, and legal dimensions simultaneously may be most effective. For patients such as Mr. M, this would include aggressively treating mood symptoms while employing motivational interviewing techniques to improve engagement in substance dependence treatment. If possible, involving family members, parole officials, housing agencies, and other public assistance workers in the treatment plan may increase treatment adherence and reduce loss of contact during illness exacerbations. Stabilization of substance use and psychiatric morbidity should be accompanied by timely evaluation of HCV and other medical comorbidities in order to improve long-term prognosis.

Table 2

Assessing patients you suspect have comorbid BD and SUDs

Table 3

Pharmacotherapy for bipolar disorder and co-occurring SUDs

Related Resources

• International Society for Bipolar Disorders. www.isbd.org.
• National Institute on Alcohol Abuse and Alcoholism. www.niaaa.nih.gov.
• National Institute on Drug Abuse. www.nida.nih.gov.

Drug Brand Names

• Carbamazepine • Tegretol
• Divalproex/valproic acid • Depakote
• Lamotrigine • Lamictal
• Lithium • Lithobid
• Naltrexone • ReVia
• Quetiapine • Seroquel

Disclosure

Dr. Tolliver receives research grant funding from Forest Laboratories and the National Institute on Alcohol Abuse and Alcoholism. Neither source influenced the content or submission of this manuscript.

Box 1

Box 2

After testing positive for cocaine on a recent court-mandated urine drug screen, Mr. M, age 49, is referred by his parole officer for psychiatric and substance abuse treatment. Mr. M has bipolar I disorder and alcohol, cocaine, and opioid dependence. He says he has been hospitalized or incarcerated at least once each year for the past 22 years. Mr. M has seen numerous psychiatrists as an outpatient, but rarely for more than 2 to 3 months and has not taken any psychotropics for more than 5 months.

Approximately 1 week before his recent urine drug screen, Mr. M became euphoric, stopped sleeping for several days, and spent $2,000 on cocaine and “escorts.” He reports that each day he smokes 30 cigarettes and drinks 1 pint of liquor and prefers to use cocaine and opiates by IV injection. Several years ago Mr. M was diagnosed with hepatitis C virus (HCV) but received no further workup or treatment. Although he denies manic or psychotic symptoms, Mr. M is observed speaking to unseen others in the waiting room and has difficulty remaining still during his interview. His chief concern is insomnia, stating, “Doc, I just need something to help me sleep.”

The high prevalence of substance use disorders (SUDs) in persons with bipolar disorder (BD) is well documented.^1,2 Up to 60% of bipolar patients develop an SUD at some point in their lives.^3-5 Alcohol use disorders are particularly common among BD patients, with a lifetime prevalence of roughly 50%.^2-5 Recent epidemiologic data indicate that 38% of persons with bipolar I disorder and 19% of those with bipolar II disorder meet criteria for alcohol dependence.⁵ Comorbid SUDs in patients with BD are associated with:

• poor treatment compliance
• longer and more frequent mood episodes
• more mixed episodes
• more hospitalizations
• more frequent suicide attempts.^1,2

The impact of co-occurring SUDs on suicidality is particularly high among those with bipolar I disorder.⁶

Frequently referred to as “dual diagnosis” conditions, co-occurring BD and SUDs may be more accurately envisioned as multi-morbid, rather than comorbid, illnesses. Data from the Stanley Foundation Bipolar Network suggest that 42% of BD patients have a lifetime history of ≥2 comorbid axis I disorders.⁷ Rates of generalized anxiety disorder, panic disorder, and posttraumatic stress disorder are particularly high in BD patients with co-occurring SUDs.^8,9 In addition, the presence of 1 SUD may mark the presence of other SUDs; for example, alcohol dependence is strongly associated with polysubstance abuse, especially in females with BD.⁸ Furthermore, medical comorbidities that impact treatment decisions also are highly prevalent in BD patients with comorbid SUDs.¹⁰ In particular, HCV rates are higher in persons with BD compared with the general population,¹¹ and are >5 times as likely in bipolar patients with co-occurring SUDs.¹²

Unfortunately, limited treatment research guides clinical management of comorbid BD and SUDs.^2,13,14 Clinical trials of medications for BD traditionally have excluded patients with SUDs, and persons with serious mental illness usually are ineligible for SUD treatment studies.¹³ Furthermore, the few randomized controlled trials (RCTs) conducted in persons with both illnesses have been constrained by relatively small sample sizes and low retention rates. In the absence of a definitive consensus for optimal treatment, this article outlines general clinical considerations and an integrated approach to assessing and managing this complex patient population.

Birds of a feather

Multiple hypotheses try to account for the high rate of SUDs in patients with BD (Table 1), but none fully explain the complex interaction observed clinically.^14,15 Substance dependence and BD are chronic remitting/relapsing disorders with heterogeneous presentations and highly variable natural histories. As with SUDs, BD may go undiagnosed and untreated for years, coming to clinical attention only after substantial disease progression.¹⁶

The fluctuating illness course in BD and SUDs makes diagnosis and treatment difficult. Symptomatic periods often are interrupted by spontaneous remissions and longer—although usually temporary—periods of perceived control. Both BD and SUDs may be associated with profound mood instability, increased impulsivity, altered responsiveness to reward, and impaired executive function.¹⁷ Finally, the high degree of heritability in BD¹⁸ and many SUDs¹⁹ may make treatment engagement more difficult if either disorder is present in multiple family members because of:

• potential for greater clinical severity
• reduced psychosocial resources
• altered familial behavioral norms that may impede the patient’s recognition of illness.

Denial of illness is a critical symptom that may fluctuate with disease course in both disorders. Persons with BD or SUDs may be least likely to recognize that they are ill during periods of highest symptom severity. Accordingly, treatment adherence in patients with either disorder may be limited at baseline and decline further when the 2 illnesses co-occur.²⁰

Involvement in the criminal justice system and medical comorbidities, particularly HCV, also complicate diagnosis and treatment of BD patients with SUDs. For more information about these topics, see Box1 and Box2.

Table 1

Why is substance abuse so prevalent among bipolar patients?

Integrated clinical management

Assessment. Although not intended to be comprehensive, suggestions for routine assessment of patients with suspected SUDs and/or BD are listed in Table 2. Because clinicians may encounter dual diagnosis patients in general psychiatric clinics or specialty (addiction or mood disorder) clinics, it is useful to obtain a thorough substance use history in all patients with known or suspected BD as well as a thorough history of hypomania/mania and depression in all patients with addictive disorders. BD diagnoses by self-report or chart history in patients with SUDs should be considered cautiously because BD often is overdiagnosed in persons engaged in active substance abuse or experiencing withdrawal.²¹ If past or present mood symptoms and substance use have co-occurred, further focused assessment of mood symptoms before alcohol and drug use or during extended periods of abstinence are necessary to make the diagnosis of bipolar disorder with confidence. Family history of SUDs and/or BD are neither necessary nor sufficient for either diagnosis; however, collateral information from family or significant others could help make the diagnosis and may identify aids and obstacles to treatment planning and engagement.

When a patient’s clinical history strongly supports the diagnoses of BD and co-occurring SUDs, more detailed inquiry is warranted. Determining the patient’s age at onset of each disorder may have prognostic value because onset of mania before SUDs developed, especially in adolescence, may predict a more severe course of both illnesses.²² A complete alcohol use history should include routine questioning about past withdrawal. Previous withdrawal seizures in an actively drinking BD patient may tip the balance toward adding an anticonvulsant for mood stabilization. A thorough SUD history should elicit information about polysubstance abuse or dependence and include screening for injection drug use and other risk factors for HCV and human immunodeficiency virus (HIV), such as hypersexuality during manic or hypomanic episodes. Document the date of the last screening for HCV/ HIV in BD patients at high risk of infection. The U.S. Centers for Disease Control and Prevention recommends that all patients at high risk for HIV consider voluntary screening at least annually.²³

Assess your patient’s historical and ongoing alcohol and other drug abuse at the initial visit, and continue to monitor substance use at all subsequent visits, especially in patients with HCV. When feasible, order urine drug screening and laboratory testing for alcohol use biomarkers such as carbohydrate-deficient transferrin and gamma-glutamyltransferase to supplement self-report data, especially in patients with poor insight or low motivation. Assess suicidal ideation and any changes in suicide risk factors at every visit.

Treatment. No biologic therapies have been FDA-approved for treating patients with co-occurring BD and SUDs. Comorbid SUDs in BD patients—as well as rapid cycling and mixed mood episodes, both of which are more common in patients with comorbid SUDs—predict poor response to lithium.¹⁷ However, the evidence base for optimal pharmacotherapy remains extremely limited. Published double-blind, placebo-controlled RCTs in persons with BD and co-morbid SUDs are limited to only 1 trial each of lithium, carbamazepine, quetiapine, and naltrexone, and 2 comparisons of lithium plus divalproex vs lithium alone (Table 3).^24-29

Salloum et al²⁴ reported that bipolar I disorder patients with alcohol dependence who received divalproex plus lithium as maintenance treatment had fewer heavy drinking days and fewer drinks per heavy drinking day than those receiving lithium plus placebo. However, the addition of divalproex did not improve manic or depressive symptoms, and depression remission rates remained relatively low in both groups. A recent 6-month study comparing lithium vs lithium plus divalproex in patients with SUDs and rapid-cycling BD found no additional benefit of divalproex over lithium monotherapy in retention, mood, or substance use outcomes.²⁸ However, modest evidence that anticonvulsants such as valproic acid and carbamazepine may help treat acute alcohol withdrawal² could support their preferential use as mood stabilizers over lithium in actively drinking BD patients.

Research underscores the difficulty in keeping dual diagnosis patients in treatment. Salloum et al²⁴ reported that only one-third of randomized subjects completed the 24-week study. In the Kemp study,²⁸ 79% of 149 recruited subjects failed to complete the lead-in stabilization phase. Of the 31 remaining subjects who were randomized to lithium or lithium/divalproex combination, only 8 (26% of those randomized, 5% of those recruited) completed the 6-month trial.

Substance abuse is associated with significantly decreased treatment adherence in persons with BD²⁰ and may affect medication choice. For example, caution may be warranted in the use of lamotrigine in substance-abusing patients with poor adherence because re-titration from the starting dose is recommended if the medicine has been missed for a consecutive period exceeding 5 half-lives of the drug.³⁰

Notable progress has been made in developing psychosocial treatments for comorbid SUDs and BD. Integrated group therapy has been designed to address the 2 disorders simultaneously by emphasizing the relationship between the disorders and highlighting similarities in cognitive and behavioral change that promote recovery in both.³¹ In a recent well-controlled RCT, this approach reduced alcohol and other drug use to approximately one-half the levels observed in those who received only group drug counseling.³¹

Research suggests that an integrated approach that encompasses psychiatric, medical, psychosocial, and legal dimensions simultaneously may be most effective. For patients such as Mr. M, this would include aggressively treating mood symptoms while employing motivational interviewing techniques to improve engagement in substance dependence treatment. If possible, involving family members, parole officials, housing agencies, and other public assistance workers in the treatment plan may increase treatment adherence and reduce loss of contact during illness exacerbations. Stabilization of substance use and psychiatric morbidity should be accompanied by timely evaluation of HCV and other medical comorbidities in order to improve long-term prognosis.

Table 2

Assessing patients you suspect have comorbid BD and SUDs

Table 3

Pharmacotherapy for bipolar disorder and co-occurring SUDs

Related Resources

• International Society for Bipolar Disorders. www.isbd.org.
• National Institute on Alcohol Abuse and Alcoholism. www.niaaa.nih.gov.
• National Institute on Drug Abuse. www.nida.nih.gov.

Drug Brand Names

• Carbamazepine • Tegretol
• Divalproex/valproic acid • Depakote
• Lamotrigine • Lamictal
• Lithium • Lithobid
• Naltrexone • ReVia
• Quetiapine • Seroquel

Disclosure

Dr. Tolliver receives research grant funding from Forest Laboratories and the National Institute on Alcohol Abuse and Alcoholism. Neither source influenced the content or submission of this manuscript.

Box 1

Box 2

Discuss this article at http://currentpsychiatry.blogspot.com/2010/08/depression-in-older-adults.html#comments

Depression in older adults (age ≥65) can devastate their quality of life and increase the likelihood of institutionalization because of behavioral problems.¹ Depression is a primary risk factor for suicide, and suicide rates are highest among those age ≥65, especially among white males.² The burden of geriatric depression can extend to caregivers.¹ Prompt recognition and treatment of depression could help minimize morbidity and reduce suffering in older adults and their caregivers.

Although geriatric depression varies in severity and presentation, common categories include:

• major depressive disorder (MDD)
• vascular depression
• dysthymia
• depression in the context of dementias, psychosis, bipolar disorder, and executive dysfunction.

Diagnoses in this population generally correspond with DSM-IV-TR criteria, but geriatric depression has distinct clinical manifestations.^1,2 Compared with younger depressed patients, older adults are less likely to endorse depressed mood and more likely to report a lack of emotions.^1,2 Older patients report feelings of irritability and fearfulness more often than sadness.^1,2 Mood symptoms tend to be transient, reoccur frequently, and display either a diurnal pattern or multiple fluctuations in a single day.^1,2 Other common presentations include loss of interest in usual activities, lack of motivation, social withdrawal, and decline in activities of daily living.^1,2

Summary of recommendations

Age-specific recommendations for assessing and treating geriatric depression can be generated in part from evidence-based reviews, meta-analyses,³ and geriatric expert consensus guidelines.⁴ Such guidelines and recommendations often do not take into account the marked heterogeneity of medical, cognitive, and overall functioning in patients age ≥65, however, because they are based on studies of younger populations and patients with complicated issues often are excluded from studies. The recommendations in this article are based largely on findings from a National Institutes of Health (NIH)-sponsored project by Alexopoulos et al to develop consensus guidelines for managing geriatric depression and expert opinion from clinicians who treat geriatric patients.⁴

During your initial clinical evaluation, confirm the diagnosis and type, duration, and severity of depression. Seek to understand the biopsychosocial context of each patient’s presentation. Carefully consider your patient’s suicide risk. Hospitalization may be required if he or she is at high risk for suicide or has complex medical and social circumstances that cannot be managed adequately in an outpatient setting.⁵

Unipolar major depression

For unipolar, nonpsychotic geriatric depression, the NIH-Alexopoulos et al guidelines emphasize a combination of antidepressants and psychotherapy (Algorithm 1).⁴ Selective serotonin reuptake inhibitors (SSRIs) and venlafaxine are first-line options.^4,6,7 Tricyclic antidepressants (TCAs), bupropion, and mirtazapine are alternatives.⁴ Among SSRIs, citalopram, escitalopram, and sertraline are preferred initial antidepressants. Fluoxetine is used less frequently.⁴ Paroxetine also is less commonly used because of its anticholinergic effects and because the drug inhibits cytochrome P4502D6,² which metabolizes several medications commonly prescribed for older adults. Among TCAs, nortriptyline is preferred.⁴ Studies have shown that duloxetine improves depression and is safe and well-tolerated in older adults with recurrent MDD.⁸ Electroconvulsive therapy (ECT) is an option for treating severe or treatment-resistant unipolar major depression.⁹

For unipolar depression with psychotic symptoms, guidelines recommend a combination of an antidepressant and an antipsychotic or ECT.⁴ Atypical antipsychotics are preferred over typical antipsychotics⁴; risperidone, olanzapine, and quetiapine are most frequently used.⁴ Clinical data on aripiprazole and ziprasidone in older adults are limited. Many geriatric experts recommend continuing an antipsychotic for 6 months after symptom remission, then gradually tapering the dose.⁴

During acute illness, administer an anti-depressant for 6 to 12 weeks at the individually determined dose required to achieve symptom remission.⁶ For an older adult experiencing a first lifetime episode of major depression, continue antidepressant treatment for 1 year after remission.⁴ If your patient has had 2 lifetime episodes of major depression, continue the antidepressant at the same dose used to achieve remission for at least 3 years. For patients who have had ≥3 episodes of depression or whose index episode was particularly severe or involved significant suicidal thoughts or behaviors, continue maintenance treatment indefinitely.

Algorithm 1: Treatment for unipolar depression in geriatric patients

ECT: electroconvulsive therapy

Bipolar depression

Mood stabilizers such as lithium or valproate—as monotherapy or in combination with an antidepressant—are recommended to treat bipolar depression without psychotic symptoms in older adults (Algorithm 2).¹⁰ For bipolar depression with psychotic symptoms, a combination of a mood stabilizer and an atypical antipsychotic or ECT is recommended.¹⁰

Older adults’ increased sensitivity to side effects and reduced ability to tolerate lithium may limit its use and may prompt you to consider atypical antipsychotics as alternatives to other mood stabilizers. Although quetiapine and fluoxetineolanzapine combination are well studied in younger patients,^11,12 there is a lack of data to support their clinical effectiveness and tolerability in older adults. Among antidepressants, SSRIs or bupropion are preferred over TCAs to prevent a switch to mania.¹⁰ Lamotrigine is an effective maintenance treatment for bipolar depressive episodes in older adults.¹³

Although optimal mood stabilizer and antidepressant dosing for this population has not been adequately assessed, pharmacotherapy that has been effective generally should be continued without modification for at least 6 to 12 months.¹⁰ After the patient achieves remission, gradually discontinue antidepressants while maintaining the mood stabilizer.¹⁰

Algorithm 2: Bipolar depression: Options for combination therapy

ECT: electroconvulsive therapy; SSRIs: selective serotonin reuptake inhibitors

Depression in dementia

Managing depression in dementia patients is similar to treatment in older adults without dementia,^5,14 although pharmacologic agents must be carefully selected because of increased risk of side effects (Algorithm 3). American Psychiatric Association practice guidelines recommend considering antidepressants for depressed patients with dementia even if their mood disturbances do not meet DSM-IV-TR criteria for MDD.⁵

SSRIs’ lower side effect profile make them the preferred treatment; the selective serotonin-norepinephrine reuptake inhibitor (SNRI) venlafaxine is a second-line option.^4,14 Avoid TCAs and other agents with anticholinergic side effects because of potential cardiovascular complications and cognitive side effects, unless SSRIs or SNRIs are ineffective or contraindicated.¹⁴ Recently clinicians have been reluctant to use antipsychotics in patients with dementia, because of the FDA’s “black-box” warning regarding the increased mortality risk associated with their use in this population.

When using ECT to treat depression in patients with dementia, the treatment protocol often is modified to twice-a-week, unilateral stimulus because of these patients’ increased risk of delirium.¹⁴ The safety of ECT to treat depression in patients with dementia has not been adequately assessed.¹⁴

Algorithm 3: Treating comorbid depression and dementia

ECT: electroconvulsive therapy; SNRI: selective serotoninnorepinephrine reuptake inhibitor; SSRI: selective serotonin reuptake inhibitor

Vascular depression

The “vascular depression hypothesis” proposes that accumulation of subcortical white matter hyperintensities can disrupt frontostriatal pathways, resulting in depressive symptoms.¹⁵ This hypothesis is supported by the confluence of depression and vascular risk factors.¹⁵ Sertraline, citalopram, nortriptyline,¹⁶ and trazodone¹⁵ have been shown to reduce depressive symptoms after a stroke.

Minor depression and dysthymia

Although the efficacy of antidepressants in minor depression—depression that does not meet criteria for MDD—is not well established, expert consensus guidelines recommend SSRIs and psychotherapy, separately or in combination, for minor depression and dysthymia in older adults (Algorithm 4).⁴ Depression in executive dysfunction responds poorly to SSRI treatment²; however, behaviorally oriented psychotherapeutic interventions such as problem-solving therapy (PST) show promise.²

Algorithm 4: Minor depression: SSRIs plus psychotherapy

SSRIs: selective serotonin reuptake inhibitors

Comorbid medical conditions

When an older adult has a medical problem that likely contributes to depression—such as hypothyroidism—treat the condition and prescribe antidepressants simultaneously.² However, if the medical problem likely causes depression—such as substance withdrawal—treat the condition first and prescribe antidepressants only if mood symptoms persist.²

Refractory depression

If your patient does not respond to an antidepressant trial of adequate dosage and duration, first make sure he or she is taking it correctly (Algorithm 5). After ruling out poor adherence, screen for comorbid psychiatric or medical conditions or psychosocial stressors and reassess the principal diagnosis.⁵

If these steps don’t address your patient’s depressive symptoms, expert consensus guidelines suggest switching to a different antidepressant:⁴

• If you first prescribed an SSRI, consider venlafaxine XR or bupropion SR.^4,17
• If your patient initially received a TCA or bupropion, an SSRI or venlafaxine XR would be appropriate.⁴
• If venlafaxine XR was the first antidepressant, a SSRI is recommended.⁴

If your patient experienced a partial response but not full remission with the initial antidepressant, consider adding a second antidepressant or an augmenting agent:⁴

• If your patient first received an SSRI, adding bupropion, lithium, or nortriptyline is recommended.
• If the initial antidepressant was a TCA or bupropion, consider adding lithium or an SSRI.
• Augmenting venlafaxine XR with lithium is recommended.⁴

The National Institutes of Mental Health-sponsored Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study of treatment-resistant depression in mixed-age groups reported that patients who do not attain remission with an initial SSRI may respond to switching to bupropion SR or venlafaxine XR.¹⁷ Augmenting an SSRI with bupropion SR has been shown to be effective.¹⁸ In addition, consider mirtazapine augmentation,¹⁹ especially if your patient experiences insomnia or anorexia. A combination of mirtazapine and venlafaxine have better efficacy and tolerability compared with the monoamine oxidase inhibitor tranylcypromine.¹⁹ Some studies have shown augmenting SSRIs with buspirone in patients with severe depression is efficacious and safe in younger adults,²⁰ but this practice is not well studied in older patients.

Algorithm 5: Treatment-resistant geriatric depression: Partial vs no response

SNRI: selective serotonin-norepinephrine reuptake inhibitor; SSRI: selective serotonin reuptake inhibitor; TCA: tricyclic antidepressant

Nonpharmacologic treatments

ECT is an important therapeutic intervention because of its safety, efficacy, and faster clinical response.^6,7,9,21 Consider ECT for older adults with severe or psychotic major depression, acute suicidality, catatonia, or severe malnutrition caused by refusal to eat. Patients who remain significantly symptomatic after multiple medication trials, do not tolerate medications well, or have comorbid medical conditions that preclude antidepressant use also are potential candidates for ECT.^5,22

ECT can be administered to many older depressed adults with relatively low complication rates. Pretreatment clinical and laboratory evaluations and consultation with medical colleagues may minimize the risk of adverse effects, including cardiovascular instability, delirium, and falls.⁹ Anterograde memory loss—a common concern for clinicians and patients—usually is temporary and can be reduced by modifying the ECT administration parameters, such as switching from bilateral to unilateral stimulus and spacing treatments.⁹ Use caution when considering ECT for patients with cardiovascular or neurologic conditions—such as myocardial infarction or cerebrovascular accident within 6 months of treatment—that may increase the risk of adverse effects. Some pharmacologic agents, such as benzodiazepines and anticonvulsant mood stabilizers, may decrease ECT’s efficacy by inhibiting seizure.²²

Depressive relapse after ECT is a major clinical concern.²¹ Continuation ECT— within the first 6 months of remission— aims to prevent relapse of the same episode, whereas maintenance ECT—beyond the first 6 months—helps avert occurrence of new episodes.^4,21 Relapse and recurrence also can be prevented with continuation or maintenance pharmacotherapy,^4,21 which should be initiated immediately after the index course of ECT.²¹ Typically, ECT continuation/maintenance treatments are provided weekly, then gradually spaced out to once a month based on the minimum frequency that is effective for an individual patient.²¹

Psychotherapy for geriatric depression generally is effective.²³ One-half of older patients prefer psychotherapy over pharmacotherapy.²⁴ Efficacious psychotherapies include behavioral therapy, cognitive-behavioral therapy (CBT), PST, brief dynamic therapy, interpersonal therapy, supportive therapy, and reminiscence therapy.²³ CBT has the most empiric support for treating geriatric depression.^5,6

Psychotherapy alone is appropriate for mild-to-moderate depression, although severe depression requires adding medication.²⁵ The combination of pharmacotherapy and psychotherapy appears to be more effective than either intervention alone in preventing recurrent major depression, especially when a specific psychosocial stressor has been identified.^5,6 CBT, interpersonal therapy, and family-focused therapy enhance pharmacotherapy outcomes in bipolar disorder.¹³

The Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD) study found that in mixed-age patients, pharmacotherapy plus psychotherapy is more beneficial than medication alone in stabilizing bipolar depression.²⁶ For older adults with executive dysfunction, research suggests that PST is more effective than other psychotherapies.²⁷ Psychosocial interventions—such as psychoeducation for the family and caregivers, family counseling, and participation in senior citizen centers and services—are strongly recommended for many patients.⁴

Related Resources

• Blazer DG, Steffens DC, Koenig HG. Mood disorders. In: Blazer DG, Steffens DC, eds. The American Psychiatric Publishing textbook of geriatric psychiatry. 4th ed. Arlington, VA: American Psychiatric Publishing, Inc.; 2009:275-300.
• American Association for Geriatric Psychiatry. www.aagponline.org.

Drug Brand Names

• Aripiprazole • Abilify
• Bupropion • Wellbutrin, Zyban
• Buspirone • Buspar
• Citalopram • Celexa
• Duloxetine • Cymbalta
• Escitalopram • Lexapro
• Fluoxetine • Prozac
• Fluoxetine-olanzapine • Symbyax
• Lamotrigine • Lamictal
• Lithium • Eskalith, Lithobid
• Mirtazapine • Remeron
• Nortriptyline • Aventyl, Pamelor
• Olanzapine • Zyprexa
• Paroxetine • Paxil
• Quetiapine • Seroquel
• Risperidone • Risperdal
• Sertraline • Zoloft
• Tranylcypromine • Parnate
• Trazodone • Desyrel
• Valproate • Depakote
• Venlafaxine • Effexor
• Ziprasidone • Geodon

Disclosure

The authors report no financial relationship with the manufacturer of any product mentioned in this article or with manufacturers of competing products.

Discuss this article at http://currentpsychiatry.blogspot.com/2010/08/depression-in-older-adults.html#comments

Depression in older adults (age ≥65) can devastate their quality of life and increase the likelihood of institutionalization because of behavioral problems.¹ Depression is a primary risk factor for suicide, and suicide rates are highest among those age ≥65, especially among white males.² The burden of geriatric depression can extend to caregivers.¹ Prompt recognition and treatment of depression could help minimize morbidity and reduce suffering in older adults and their caregivers.

Although geriatric depression varies in severity and presentation, common categories include:

• major depressive disorder (MDD)
• vascular depression
• dysthymia
• depression in the context of dementias, psychosis, bipolar disorder, and executive dysfunction.

Diagnoses in this population generally correspond with DSM-IV-TR criteria, but geriatric depression has distinct clinical manifestations.^1,2 Compared with younger depressed patients, older adults are less likely to endorse depressed mood and more likely to report a lack of emotions.^1,2 Older patients report feelings of irritability and fearfulness more often than sadness.^1,2 Mood symptoms tend to be transient, reoccur frequently, and display either a diurnal pattern or multiple fluctuations in a single day.^1,2 Other common presentations include loss of interest in usual activities, lack of motivation, social withdrawal, and decline in activities of daily living.^1,2

Summary of recommendations

Age-specific recommendations for assessing and treating geriatric depression can be generated in part from evidence-based reviews, meta-analyses,³ and geriatric expert consensus guidelines.⁴ Such guidelines and recommendations often do not take into account the marked heterogeneity of medical, cognitive, and overall functioning in patients age ≥65, however, because they are based on studies of younger populations and patients with complicated issues often are excluded from studies. The recommendations in this article are based largely on findings from a National Institutes of Health (NIH)-sponsored project by Alexopoulos et al to develop consensus guidelines for managing geriatric depression and expert opinion from clinicians who treat geriatric patients.⁴

During your initial clinical evaluation, confirm the diagnosis and type, duration, and severity of depression. Seek to understand the biopsychosocial context of each patient’s presentation. Carefully consider your patient’s suicide risk. Hospitalization may be required if he or she is at high risk for suicide or has complex medical and social circumstances that cannot be managed adequately in an outpatient setting.⁵

Unipolar major depression

For unipolar, nonpsychotic geriatric depression, the NIH-Alexopoulos et al guidelines emphasize a combination of antidepressants and psychotherapy (Algorithm 1).⁴ Selective serotonin reuptake inhibitors (SSRIs) and venlafaxine are first-line options.^4,6,7 Tricyclic antidepressants (TCAs), bupropion, and mirtazapine are alternatives.⁴ Among SSRIs, citalopram, escitalopram, and sertraline are preferred initial antidepressants. Fluoxetine is used less frequently.⁴ Paroxetine also is less commonly used because of its anticholinergic effects and because the drug inhibits cytochrome P4502D6,² which metabolizes several medications commonly prescribed for older adults. Among TCAs, nortriptyline is preferred.⁴ Studies have shown that duloxetine improves depression and is safe and well-tolerated in older adults with recurrent MDD.⁸ Electroconvulsive therapy (ECT) is an option for treating severe or treatment-resistant unipolar major depression.⁹

For unipolar depression with psychotic symptoms, guidelines recommend a combination of an antidepressant and an antipsychotic or ECT.⁴ Atypical antipsychotics are preferred over typical antipsychotics⁴; risperidone, olanzapine, and quetiapine are most frequently used.⁴ Clinical data on aripiprazole and ziprasidone in older adults are limited. Many geriatric experts recommend continuing an antipsychotic for 6 months after symptom remission, then gradually tapering the dose.⁴

During acute illness, administer an anti-depressant for 6 to 12 weeks at the individually determined dose required to achieve symptom remission.⁶ For an older adult experiencing a first lifetime episode of major depression, continue antidepressant treatment for 1 year after remission.⁴ If your patient has had 2 lifetime episodes of major depression, continue the antidepressant at the same dose used to achieve remission for at least 3 years. For patients who have had ≥3 episodes of depression or whose index episode was particularly severe or involved significant suicidal thoughts or behaviors, continue maintenance treatment indefinitely.

Algorithm 1: Treatment for unipolar depression in geriatric patients

ECT: electroconvulsive therapy

Bipolar depression

Mood stabilizers such as lithium or valproate—as monotherapy or in combination with an antidepressant—are recommended to treat bipolar depression without psychotic symptoms in older adults (Algorithm 2).¹⁰ For bipolar depression with psychotic symptoms, a combination of a mood stabilizer and an atypical antipsychotic or ECT is recommended.¹⁰

Older adults’ increased sensitivity to side effects and reduced ability to tolerate lithium may limit its use and may prompt you to consider atypical antipsychotics as alternatives to other mood stabilizers. Although quetiapine and fluoxetineolanzapine combination are well studied in younger patients,^11,12 there is a lack of data to support their clinical effectiveness and tolerability in older adults. Among antidepressants, SSRIs or bupropion are preferred over TCAs to prevent a switch to mania.¹⁰ Lamotrigine is an effective maintenance treatment for bipolar depressive episodes in older adults.¹³

Although optimal mood stabilizer and antidepressant dosing for this population has not been adequately assessed, pharmacotherapy that has been effective generally should be continued without modification for at least 6 to 12 months.¹⁰ After the patient achieves remission, gradually discontinue antidepressants while maintaining the mood stabilizer.¹⁰

Algorithm 2: Bipolar depression: Options for combination therapy

ECT: electroconvulsive therapy; SSRIs: selective serotonin reuptake inhibitors

Depression in dementia

Managing depression in dementia patients is similar to treatment in older adults without dementia,^5,14 although pharmacologic agents must be carefully selected because of increased risk of side effects (Algorithm 3). American Psychiatric Association practice guidelines recommend considering antidepressants for depressed patients with dementia even if their mood disturbances do not meet DSM-IV-TR criteria for MDD.⁵

SSRIs’ lower side effect profile make them the preferred treatment; the selective serotonin-norepinephrine reuptake inhibitor (SNRI) venlafaxine is a second-line option.^4,14 Avoid TCAs and other agents with anticholinergic side effects because of potential cardiovascular complications and cognitive side effects, unless SSRIs or SNRIs are ineffective or contraindicated.¹⁴ Recently clinicians have been reluctant to use antipsychotics in patients with dementia, because of the FDA’s “black-box” warning regarding the increased mortality risk associated with their use in this population.

When using ECT to treat depression in patients with dementia, the treatment protocol often is modified to twice-a-week, unilateral stimulus because of these patients’ increased risk of delirium.¹⁴ The safety of ECT to treat depression in patients with dementia has not been adequately assessed.¹⁴

Algorithm 3: Treating comorbid depression and dementia

ECT: electroconvulsive therapy; SNRI: selective serotoninnorepinephrine reuptake inhibitor; SSRI: selective serotonin reuptake inhibitor

Vascular depression

The “vascular depression hypothesis” proposes that accumulation of subcortical white matter hyperintensities can disrupt frontostriatal pathways, resulting in depressive symptoms.¹⁵ This hypothesis is supported by the confluence of depression and vascular risk factors.¹⁵ Sertraline, citalopram, nortriptyline,¹⁶ and trazodone¹⁵ have been shown to reduce depressive symptoms after a stroke.

Minor depression and dysthymia

Although the efficacy of antidepressants in minor depression—depression that does not meet criteria for MDD—is not well established, expert consensus guidelines recommend SSRIs and psychotherapy, separately or in combination, for minor depression and dysthymia in older adults (Algorithm 4).⁴ Depression in executive dysfunction responds poorly to SSRI treatment²; however, behaviorally oriented psychotherapeutic interventions such as problem-solving therapy (PST) show promise.²

Algorithm 4: Minor depression: SSRIs plus psychotherapy

SSRIs: selective serotonin reuptake inhibitors

Comorbid medical conditions

When an older adult has a medical problem that likely contributes to depression—such as hypothyroidism—treat the condition and prescribe antidepressants simultaneously.² However, if the medical problem likely causes depression—such as substance withdrawal—treat the condition first and prescribe antidepressants only if mood symptoms persist.²

Refractory depression

If your patient does not respond to an antidepressant trial of adequate dosage and duration, first make sure he or she is taking it correctly (Algorithm 5). After ruling out poor adherence, screen for comorbid psychiatric or medical conditions or psychosocial stressors and reassess the principal diagnosis.⁵

If these steps don’t address your patient’s depressive symptoms, expert consensus guidelines suggest switching to a different antidepressant:⁴

• If you first prescribed an SSRI, consider venlafaxine XR or bupropion SR.^4,17
• If your patient initially received a TCA or bupropion, an SSRI or venlafaxine XR would be appropriate.⁴
• If venlafaxine XR was the first antidepressant, a SSRI is recommended.⁴

If your patient experienced a partial response but not full remission with the initial antidepressant, consider adding a second antidepressant or an augmenting agent:⁴

• If your patient first received an SSRI, adding bupropion, lithium, or nortriptyline is recommended.
• If the initial antidepressant was a TCA or bupropion, consider adding lithium or an SSRI.
• Augmenting venlafaxine XR with lithium is recommended.⁴

The National Institutes of Mental Health-sponsored Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study of treatment-resistant depression in mixed-age groups reported that patients who do not attain remission with an initial SSRI may respond to switching to bupropion SR or venlafaxine XR.¹⁷ Augmenting an SSRI with bupropion SR has been shown to be effective.¹⁸ In addition, consider mirtazapine augmentation,¹⁹ especially if your patient experiences insomnia or anorexia. A combination of mirtazapine and venlafaxine have better efficacy and tolerability compared with the monoamine oxidase inhibitor tranylcypromine.¹⁹ Some studies have shown augmenting SSRIs with buspirone in patients with severe depression is efficacious and safe in younger adults,²⁰ but this practice is not well studied in older patients.

Algorithm 5: Treatment-resistant geriatric depression: Partial vs no response

SNRI: selective serotonin-norepinephrine reuptake inhibitor; SSRI: selective serotonin reuptake inhibitor; TCA: tricyclic antidepressant

Nonpharmacologic treatments

ECT is an important therapeutic intervention because of its safety, efficacy, and faster clinical response.^6,7,9,21 Consider ECT for older adults with severe or psychotic major depression, acute suicidality, catatonia, or severe malnutrition caused by refusal to eat. Patients who remain significantly symptomatic after multiple medication trials, do not tolerate medications well, or have comorbid medical conditions that preclude antidepressant use also are potential candidates for ECT.^5,22

ECT can be administered to many older depressed adults with relatively low complication rates. Pretreatment clinical and laboratory evaluations and consultation with medical colleagues may minimize the risk of adverse effects, including cardiovascular instability, delirium, and falls.⁹ Anterograde memory loss—a common concern for clinicians and patients—usually is temporary and can be reduced by modifying the ECT administration parameters, such as switching from bilateral to unilateral stimulus and spacing treatments.⁹ Use caution when considering ECT for patients with cardiovascular or neurologic conditions—such as myocardial infarction or cerebrovascular accident within 6 months of treatment—that may increase the risk of adverse effects. Some pharmacologic agents, such as benzodiazepines and anticonvulsant mood stabilizers, may decrease ECT’s efficacy by inhibiting seizure.²²

Depressive relapse after ECT is a major clinical concern.²¹ Continuation ECT— within the first 6 months of remission— aims to prevent relapse of the same episode, whereas maintenance ECT—beyond the first 6 months—helps avert occurrence of new episodes.^4,21 Relapse and recurrence also can be prevented with continuation or maintenance pharmacotherapy,^4,21 which should be initiated immediately after the index course of ECT.²¹ Typically, ECT continuation/maintenance treatments are provided weekly, then gradually spaced out to once a month based on the minimum frequency that is effective for an individual patient.²¹

Psychotherapy for geriatric depression generally is effective.²³ One-half of older patients prefer psychotherapy over pharmacotherapy.²⁴ Efficacious psychotherapies include behavioral therapy, cognitive-behavioral therapy (CBT), PST, brief dynamic therapy, interpersonal therapy, supportive therapy, and reminiscence therapy.²³ CBT has the most empiric support for treating geriatric depression.^5,6

Psychotherapy alone is appropriate for mild-to-moderate depression, although severe depression requires adding medication.²⁵ The combination of pharmacotherapy and psychotherapy appears to be more effective than either intervention alone in preventing recurrent major depression, especially when a specific psychosocial stressor has been identified.^5,6 CBT, interpersonal therapy, and family-focused therapy enhance pharmacotherapy outcomes in bipolar disorder.¹³

The Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD) study found that in mixed-age patients, pharmacotherapy plus psychotherapy is more beneficial than medication alone in stabilizing bipolar depression.²⁶ For older adults with executive dysfunction, research suggests that PST is more effective than other psychotherapies.²⁷ Psychosocial interventions—such as psychoeducation for the family and caregivers, family counseling, and participation in senior citizen centers and services—are strongly recommended for many patients.⁴

Related Resources

• Blazer DG, Steffens DC, Koenig HG. Mood disorders. In: Blazer DG, Steffens DC, eds. The American Psychiatric Publishing textbook of geriatric psychiatry. 4th ed. Arlington, VA: American Psychiatric Publishing, Inc.; 2009:275-300.
• American Association for Geriatric Psychiatry. www.aagponline.org.

Drug Brand Names

• Aripiprazole • Abilify
• Bupropion • Wellbutrin, Zyban
• Buspirone • Buspar
• Citalopram • Celexa
• Duloxetine • Cymbalta
• Escitalopram • Lexapro
• Fluoxetine • Prozac
• Fluoxetine-olanzapine • Symbyax
• Lamotrigine • Lamictal
• Lithium • Eskalith, Lithobid
• Mirtazapine • Remeron
• Nortriptyline • Aventyl, Pamelor
• Olanzapine • Zyprexa
• Paroxetine • Paxil
• Quetiapine • Seroquel
• Risperidone • Risperdal
• Sertraline • Zoloft
• Tranylcypromine • Parnate
• Trazodone • Desyrel
• Valproate • Depakote
• Venlafaxine • Effexor
• Ziprasidone • Geodon

Disclosure

The authors report no financial relationship with the manufacturer of any product mentioned in this article or with manufacturers of competing products.

Discuss this article at http://currentpsychiatry.blogspot.com/2010/08/depression-in-older-adults.html#comments

Depression in older adults (age ≥65) can devastate their quality of life and increase the likelihood of institutionalization because of behavioral problems.¹ Depression is a primary risk factor for suicide, and suicide rates are highest among those age ≥65, especially among white males.² The burden of geriatric depression can extend to caregivers.¹ Prompt recognition and treatment of depression could help minimize morbidity and reduce suffering in older adults and their caregivers.

Although geriatric depression varies in severity and presentation, common categories include:

• major depressive disorder (MDD)
• vascular depression
• dysthymia
• depression in the context of dementias, psychosis, bipolar disorder, and executive dysfunction.

Diagnoses in this population generally correspond with DSM-IV-TR criteria, but geriatric depression has distinct clinical manifestations.^1,2 Compared with younger depressed patients, older adults are less likely to endorse depressed mood and more likely to report a lack of emotions.^1,2 Older patients report feelings of irritability and fearfulness more often than sadness.^1,2 Mood symptoms tend to be transient, reoccur frequently, and display either a diurnal pattern or multiple fluctuations in a single day.^1,2 Other common presentations include loss of interest in usual activities, lack of motivation, social withdrawal, and decline in activities of daily living.^1,2

Summary of recommendations

Age-specific recommendations for assessing and treating geriatric depression can be generated in part from evidence-based reviews, meta-analyses,³ and geriatric expert consensus guidelines.⁴ Such guidelines and recommendations often do not take into account the marked heterogeneity of medical, cognitive, and overall functioning in patients age ≥65, however, because they are based on studies of younger populations and patients with complicated issues often are excluded from studies. The recommendations in this article are based largely on findings from a National Institutes of Health (NIH)-sponsored project by Alexopoulos et al to develop consensus guidelines for managing geriatric depression and expert opinion from clinicians who treat geriatric patients.⁴

During your initial clinical evaluation, confirm the diagnosis and type, duration, and severity of depression. Seek to understand the biopsychosocial context of each patient’s presentation. Carefully consider your patient’s suicide risk. Hospitalization may be required if he or she is at high risk for suicide or has complex medical and social circumstances that cannot be managed adequately in an outpatient setting.⁵

Unipolar major depression

For unipolar, nonpsychotic geriatric depression, the NIH-Alexopoulos et al guidelines emphasize a combination of antidepressants and psychotherapy (Algorithm 1).⁴ Selective serotonin reuptake inhibitors (SSRIs) and venlafaxine are first-line options.^4,6,7 Tricyclic antidepressants (TCAs), bupropion, and mirtazapine are alternatives.⁴ Among SSRIs, citalopram, escitalopram, and sertraline are preferred initial antidepressants. Fluoxetine is used less frequently.⁴ Paroxetine also is less commonly used because of its anticholinergic effects and because the drug inhibits cytochrome P4502D6,² which metabolizes several medications commonly prescribed for older adults. Among TCAs, nortriptyline is preferred.⁴ Studies have shown that duloxetine improves depression and is safe and well-tolerated in older adults with recurrent MDD.⁸ Electroconvulsive therapy (ECT) is an option for treating severe or treatment-resistant unipolar major depression.⁹

For unipolar depression with psychotic symptoms, guidelines recommend a combination of an antidepressant and an antipsychotic or ECT.⁴ Atypical antipsychotics are preferred over typical antipsychotics⁴; risperidone, olanzapine, and quetiapine are most frequently used.⁴ Clinical data on aripiprazole and ziprasidone in older adults are limited. Many geriatric experts recommend continuing an antipsychotic for 6 months after symptom remission, then gradually tapering the dose.⁴

During acute illness, administer an anti-depressant for 6 to 12 weeks at the individually determined dose required to achieve symptom remission.⁶ For an older adult experiencing a first lifetime episode of major depression, continue antidepressant treatment for 1 year after remission.⁴ If your patient has had 2 lifetime episodes of major depression, continue the antidepressant at the same dose used to achieve remission for at least 3 years. For patients who have had ≥3 episodes of depression or whose index episode was particularly severe or involved significant suicidal thoughts or behaviors, continue maintenance treatment indefinitely.

Algorithm 1: Treatment for unipolar depression in geriatric patients

ECT: electroconvulsive therapy

Bipolar depression

Mood stabilizers such as lithium or valproate—as monotherapy or in combination with an antidepressant—are recommended to treat bipolar depression without psychotic symptoms in older adults (Algorithm 2).¹⁰ For bipolar depression with psychotic symptoms, a combination of a mood stabilizer and an atypical antipsychotic or ECT is recommended.¹⁰

Older adults’ increased sensitivity to side effects and reduced ability to tolerate lithium may limit its use and may prompt you to consider atypical antipsychotics as alternatives to other mood stabilizers. Although quetiapine and fluoxetineolanzapine combination are well studied in younger patients,^11,12 there is a lack of data to support their clinical effectiveness and tolerability in older adults. Among antidepressants, SSRIs or bupropion are preferred over TCAs to prevent a switch to mania.¹⁰ Lamotrigine is an effective maintenance treatment for bipolar depressive episodes in older adults.¹³

Although optimal mood stabilizer and antidepressant dosing for this population has not been adequately assessed, pharmacotherapy that has been effective generally should be continued without modification for at least 6 to 12 months.¹⁰ After the patient achieves remission, gradually discontinue antidepressants while maintaining the mood stabilizer.¹⁰

Algorithm 2: Bipolar depression: Options for combination therapy

ECT: electroconvulsive therapy; SSRIs: selective serotonin reuptake inhibitors

Depression in dementia

Managing depression in dementia patients is similar to treatment in older adults without dementia,^5,14 although pharmacologic agents must be carefully selected because of increased risk of side effects (Algorithm 3). American Psychiatric Association practice guidelines recommend considering antidepressants for depressed patients with dementia even if their mood disturbances do not meet DSM-IV-TR criteria for MDD.⁵

SSRIs’ lower side effect profile make them the preferred treatment; the selective serotonin-norepinephrine reuptake inhibitor (SNRI) venlafaxine is a second-line option.^4,14 Avoid TCAs and other agents with anticholinergic side effects because of potential cardiovascular complications and cognitive side effects, unless SSRIs or SNRIs are ineffective or contraindicated.¹⁴ Recently clinicians have been reluctant to use antipsychotics in patients with dementia, because of the FDA’s “black-box” warning regarding the increased mortality risk associated with their use in this population.

When using ECT to treat depression in patients with dementia, the treatment protocol often is modified to twice-a-week, unilateral stimulus because of these patients’ increased risk of delirium.¹⁴ The safety of ECT to treat depression in patients with dementia has not been adequately assessed.¹⁴

Algorithm 3: Treating comorbid depression and dementia

ECT: electroconvulsive therapy; SNRI: selective serotoninnorepinephrine reuptake inhibitor; SSRI: selective serotonin reuptake inhibitor

Vascular depression

The “vascular depression hypothesis” proposes that accumulation of subcortical white matter hyperintensities can disrupt frontostriatal pathways, resulting in depressive symptoms.¹⁵ This hypothesis is supported by the confluence of depression and vascular risk factors.¹⁵ Sertraline, citalopram, nortriptyline,¹⁶ and trazodone¹⁵ have been shown to reduce depressive symptoms after a stroke.

Minor depression and dysthymia

Although the efficacy of antidepressants in minor depression—depression that does not meet criteria for MDD—is not well established, expert consensus guidelines recommend SSRIs and psychotherapy, separately or in combination, for minor depression and dysthymia in older adults (Algorithm 4).⁴ Depression in executive dysfunction responds poorly to SSRI treatment²; however, behaviorally oriented psychotherapeutic interventions such as problem-solving therapy (PST) show promise.²

Algorithm 4: Minor depression: SSRIs plus psychotherapy

SSRIs: selective serotonin reuptake inhibitors

Comorbid medical conditions

When an older adult has a medical problem that likely contributes to depression—such as hypothyroidism—treat the condition and prescribe antidepressants simultaneously.² However, if the medical problem likely causes depression—such as substance withdrawal—treat the condition first and prescribe antidepressants only if mood symptoms persist.²

Refractory depression

If your patient does not respond to an antidepressant trial of adequate dosage and duration, first make sure he or she is taking it correctly (Algorithm 5). After ruling out poor adherence, screen for comorbid psychiatric or medical conditions or psychosocial stressors and reassess the principal diagnosis.⁵

If these steps don’t address your patient’s depressive symptoms, expert consensus guidelines suggest switching to a different antidepressant:⁴

• If you first prescribed an SSRI, consider venlafaxine XR or bupropion SR.^4,17
• If your patient initially received a TCA or bupropion, an SSRI or venlafaxine XR would be appropriate.⁴
• If venlafaxine XR was the first antidepressant, a SSRI is recommended.⁴

If your patient experienced a partial response but not full remission with the initial antidepressant, consider adding a second antidepressant or an augmenting agent:⁴

• If your patient first received an SSRI, adding bupropion, lithium, or nortriptyline is recommended.
• If the initial antidepressant was a TCA or bupropion, consider adding lithium or an SSRI.
• Augmenting venlafaxine XR with lithium is recommended.⁴

The National Institutes of Mental Health-sponsored Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study of treatment-resistant depression in mixed-age groups reported that patients who do not attain remission with an initial SSRI may respond to switching to bupropion SR or venlafaxine XR.¹⁷ Augmenting an SSRI with bupropion SR has been shown to be effective.¹⁸ In addition, consider mirtazapine augmentation,¹⁹ especially if your patient experiences insomnia or anorexia. A combination of mirtazapine and venlafaxine have better efficacy and tolerability compared with the monoamine oxidase inhibitor tranylcypromine.¹⁹ Some studies have shown augmenting SSRIs with buspirone in patients with severe depression is efficacious and safe in younger adults,²⁰ but this practice is not well studied in older patients.

Algorithm 5: Treatment-resistant geriatric depression: Partial vs no response

SNRI: selective serotonin-norepinephrine reuptake inhibitor; SSRI: selective serotonin reuptake inhibitor; TCA: tricyclic antidepressant

Nonpharmacologic treatments

ECT is an important therapeutic intervention because of its safety, efficacy, and faster clinical response.^6,7,9,21 Consider ECT for older adults with severe or psychotic major depression, acute suicidality, catatonia, or severe malnutrition caused by refusal to eat. Patients who remain significantly symptomatic after multiple medication trials, do not tolerate medications well, or have comorbid medical conditions that preclude antidepressant use also are potential candidates for ECT.^5,22

ECT can be administered to many older depressed adults with relatively low complication rates. Pretreatment clinical and laboratory evaluations and consultation with medical colleagues may minimize the risk of adverse effects, including cardiovascular instability, delirium, and falls.⁹ Anterograde memory loss—a common concern for clinicians and patients—usually is temporary and can be reduced by modifying the ECT administration parameters, such as switching from bilateral to unilateral stimulus and spacing treatments.⁹ Use caution when considering ECT for patients with cardiovascular or neurologic conditions—such as myocardial infarction or cerebrovascular accident within 6 months of treatment—that may increase the risk of adverse effects. Some pharmacologic agents, such as benzodiazepines and anticonvulsant mood stabilizers, may decrease ECT’s efficacy by inhibiting seizure.²²

Depressive relapse after ECT is a major clinical concern.²¹ Continuation ECT— within the first 6 months of remission— aims to prevent relapse of the same episode, whereas maintenance ECT—beyond the first 6 months—helps avert occurrence of new episodes.^4,21 Relapse and recurrence also can be prevented with continuation or maintenance pharmacotherapy,^4,21 which should be initiated immediately after the index course of ECT.²¹ Typically, ECT continuation/maintenance treatments are provided weekly, then gradually spaced out to once a month based on the minimum frequency that is effective for an individual patient.²¹

Psychotherapy for geriatric depression generally is effective.²³ One-half of older patients prefer psychotherapy over pharmacotherapy.²⁴ Efficacious psychotherapies include behavioral therapy, cognitive-behavioral therapy (CBT), PST, brief dynamic therapy, interpersonal therapy, supportive therapy, and reminiscence therapy.²³ CBT has the most empiric support for treating geriatric depression.^5,6

Psychotherapy alone is appropriate for mild-to-moderate depression, although severe depression requires adding medication.²⁵ The combination of pharmacotherapy and psychotherapy appears to be more effective than either intervention alone in preventing recurrent major depression, especially when a specific psychosocial stressor has been identified.^5,6 CBT, interpersonal therapy, and family-focused therapy enhance pharmacotherapy outcomes in bipolar disorder.¹³

The Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD) study found that in mixed-age patients, pharmacotherapy plus psychotherapy is more beneficial than medication alone in stabilizing bipolar depression.²⁶ For older adults with executive dysfunction, research suggests that PST is more effective than other psychotherapies.²⁷ Psychosocial interventions—such as psychoeducation for the family and caregivers, family counseling, and participation in senior citizen centers and services—are strongly recommended for many patients.⁴

Related Resources

• Blazer DG, Steffens DC, Koenig HG. Mood disorders. In: Blazer DG, Steffens DC, eds. The American Psychiatric Publishing textbook of geriatric psychiatry. 4th ed. Arlington, VA: American Psychiatric Publishing, Inc.; 2009:275-300.
• American Association for Geriatric Psychiatry. www.aagponline.org.

Drug Brand Names

• Aripiprazole • Abilify
• Bupropion • Wellbutrin, Zyban
• Buspirone • Buspar
• Citalopram • Celexa
• Duloxetine • Cymbalta
• Escitalopram • Lexapro
• Fluoxetine • Prozac
• Fluoxetine-olanzapine • Symbyax
• Lamotrigine • Lamictal
• Lithium • Eskalith, Lithobid
• Mirtazapine • Remeron
• Nortriptyline • Aventyl, Pamelor
• Olanzapine • Zyprexa
• Paroxetine • Paxil
• Quetiapine • Seroquel
• Risperidone • Risperdal
• Sertraline • Zoloft
• Tranylcypromine • Parnate
• Trazodone • Desyrel
• Valproate • Depakote
• Venlafaxine • Effexor
• Ziprasidone • Geodon

Disclosure

The authors report no financial relationship with the manufacturer of any product mentioned in this article or with manufacturers of competing products.

HISTORY: Visual disturbances

Ms. G, age 30, has schizoaffective disorder and presents with worsening auditory hallucinations, paranoid delusions, and thought broadcasting and insertion, which is suspected to be related to a change from olanzapine to aripiprazole. She complains of visual disturbances—a hallucination of flashing lights—that she describes as occurring “all the time, like salt-and-pepper dancing, and sometimes splotches.” Her visual symptoms are worse when she closes her eyes and in dimly lit environments, but occur in daylight as well. Her vision is otherwise unimpaired.

The visual disturbances started with Ms. G’s first psychotic break at age 23 and have persisted continuously. She reports that at first, the spots were more intense, like an incessant strobe light. These symptoms are aggravated by sleep deprivation and anger and improve only with topiramate, 100 mg/d, which was prescribed a year earlier after a neurologic consultation but discontinued because of cognitive side effects.

Ms. G also complains of dull, aching, continuous headaches in the center of her forehead, which do not wake her, are not worse in the morning, and do not vary with the intensity of her visual symptoms. The headaches are incorporated into her system of delusions; she believes they “drain the electricity from her head” and make her feel better. She denies a history of migraines. Ms. G has a history of experiencing well-formed visual hallucinations, such as the devil, but has not had them for several years.

She had 5 to 10 episodes over the last 2 years that she describes as seizures—“an earthquake, a huge whooshing noise,” accompanied by heart palpitations that occur with stress, in public, in bright lights, or whenever she feels vulnerable. These episodes are not associated with loss of consciousness, incontinence, amnesia, or post-ictal confusion and she can stop them with relaxation techniques such as deep breathing and reminding herself to calm down. The last episode was 2 months ago and seemed unrelated to her visual phenomena. She also complains of occasional paresthesias in her extremities but denies dizziness, presyncope, or blurry or double vision.

Ms. G is obese and has no history of head injury or birth defects. Her family history is negative for epilepsy; however, her mother has bipolar disorder and father has schizophrenia. Her medications are phentermine, 37.5 mg/d for short-term management of obesity, oxcarbazepine, 600 mg/d for supposed partial seizures, sertraline, 100 mg/d for depression, aripiprazole, 10 mg/d, and nizatidine, 150 mg/d, for gastroesophageal reflux disease. She does not use alcohol, cigarettes, or illicit drugs. Ms. G obtained a master’s degree in health communications and works as a research assistant at a local medical school.

The authors’ observations

Visual hallucinations are not distortions of reality but original false perceptions that co-occur with real perceptions.¹ They vary from simple, elemental hallucinations involving flashes of light or geometric figures to more complex elaborations such as seeing crawling insects or a flock of angels.² Hallucinations can originate anywhere in the visual system from the lens to the visual cortex and can be vascular, toxic, electrophysiologic, structural, or neurochemical manifestations of neurologic, psychiatric, or ophthalmologic disorders. Depending on the hallucination’s etiology, treatment varies from emergent ophthalmologic examination and retinal surgery to antipsychotic therapy to simple observation. Photopsia is a perception of flashing lights, and its etiology varies; causes can be categorized as emergent or nonemergent.

Emergent causes of photopsia typically have a sudden onset, often can be diagnosed from associated symptoms or medical history, and must be recognized quickly to prevent permanent vision loss. Posterior vitreous detachment and retinal tearing can manifest as flashes of light, floaters, or blurred vision in the peripheral visual field that are worse in dim illumination and localizable by the patient. Retinal detachment can cause photopsia, floaters, or a “curtain” or “shadow” moving over the visual field, which may be accompanied by central or peripheral vision loss.³ Severe myopia is a risk factor for retinal detachment.

New-onset photopsia can suggest cytomegaloviral retinitis in an immune-compromised patient, and ocular-vascular crisis in a patient with sickle cell disease. New-onset or recurrent photopsia with bilateral blurred vision, ataxia, vertigo, dysarthria, or drop attacks may be caused by vertebrobasilar artery insufficiency (VBAI). Photopsia has been reported as the presenting symptom in carotid artery dissection.⁴ Photopsia correlated with eye movement implicates the retina or optic nerve. All of these etiologies except for VBAI should be referred to an ophthalmologist for indirect ophthalmoscopy and evaluation.

Nonemergent causes of photopsia include:

• drug toxicity
• migraine headache
• thyroid ophthalmopathy
• arteriovenous malformations
• seizures
• psychiatric disorders.

Digoxin, cocaine, paclitaxel, and clomiphene may have ocular side effects, including photopsia. The first symptoms of digitalis toxicity often are visual and include photopsia, yellow or green discoloration of the visual field, halos, and the appearance of frost over objects.⁵ Signs of cocaine intoxication can include visual hallucinations such as photopsia, shadows, moving objects, and insects crawling and co-occurring euphoria, hypervigilance, impaired judgment, and autonomic changes such as tachycardia, pupillary dilation, hypertension, and nausea.⁶ High-dose paclitaxel infusion has been reported to cause photopsia.⁷ Clomiphene can cause flickering lights and shimmering that persists after discontinuing the medication.⁸

Migraine with aura, which includes photopsia 39% of the time,⁹ often involves an expanding central visual disturbance or scotoma, and usually is confined to 1 visual field but may involve both. The aura typically lasts 10 to 20 minutes and often is followed by headache. Patients often report a history of episodic stereotyped auras and headache. Ocular examination is normal.

Ophthalmic migraine, also known as ocular or retinal migraine, is thought to be caused by transient vasospasm of the choroidal or retinal arteries and can be precipitated by postural changes, exercise, and oral contraceptives. It manifests as a gradual visual disturbance in a mosaic scotomata pattern that enlarges, producing total unilateral visual loss lasting from minutes to an hour, and—misleadingly—can be associated with minimal or no headache. Ocular examination is normal, and a personal or family history of migraine confirms the diagnosis.

Seizure activity in the occipital cortex and adjacent association areas can produce static light and stars. In a prospective study of 18 patients with occipital epilepsy, visual seizures lasted from a few seconds to 3 minutes—although rarely 20 to 150 minutes—and occurred in multiple clusters a day or week.¹⁰ All but 2 patients had secondary generalized tonic-clonic convulsions. Occipital seizures often are misdiagnosed as the visual aura of migraine, particularly when elementary visual hallucinations are followed by post-ictal headache and vomiting.¹¹ Lights with a color or a spherical shape suggest occipital epilepsy. Consider ordering an electroencephalogram (EEG) if the clinical diagnosis is unclear.

EXAMINATION: No obvious cause

During a mental status exam, Ms. G is pleasant, cooperative, alert, and oriented to person, place, and time. Her speech is fluent, her affect is full, and she describes her mood as a “sensitive, overwhelmed state.” Her thoughts occasionally are tangential, and she perseverates on guilty, embarrassed, and paranoid thoughts. She has auditory hallucinations and experiences photopsia during the interview, but denies other visual hallucinations. She shows no deficits in attention, concentration, memory, language, calculations, visuospatial abilities, or general fund of knowledge.

Her physical exam shows pupils that are equal, round, and reactive to light with normal extraocular movements, intact visual fields, 20/20 visual acuity in both eyes with glasses, and sharp optic disc margins with no retinal abnormalities apparent on funduscopic exam. The rest of her physical and neurologic exam is normal. Her complete blood count, electrolytes, kidney, liver function tests, urinalysis, and serum toxicology screens are normal. She has no history of immunodeficiency.

An MRI performed a year earlier showed mild diffuse congestion of a left maxillary sinus but no other intracranial abnormalities, and a waking EEG with sphenoidal electrodes was normal.

The authors’ observations

Evaluation of patients with photopsia includes taking a history and performing an eye examination including visual acuity testing, visual field testing, and funduscopic examination. The appearance, location, and duration of photopsia may help narrow the differential diagnosis and identify emergent cases (Table 1 and Table 2). Seizure activity associated with flashing lights suggests occipital epilepsy. A history of hypertension, diabetes, or polymyalgia rheumatica points to vascular causes, and hyperthyroidism suggests that flashing lights may be phosphenes of Graves’ disease. Myopia or eye trauma could indicate vitreous traction. Prescription medications and illicit drug use may point to a toxic etiology.¹²

Table 1
Diagnoses suggested by photopsia phenomena

Table 2
Photopsia differential diagnosis: Look for co-occurring symptoms

OUTCOME: Diagnosis of exclusion

Ms. G’s visual disturbance resembles vitreous detachment or classic migraine; however, the onset is not sudden, her funduscopic exam is normal, and she has no discrete episodes with concomitant headache. Ms. G’s improvement with topiramate is inconsistent with an ophthalmic origin because such photopsia should not improve with medication. It is consistent with recurrent seizures or status epilepticus of the occipital cortex; however, she did not experience multiple discrete episodes a day or generalized seizures experienced by most occipital seizure patients, and her EEG and clinical history are not consistent with seizures. Because we cannot rule out occipital lobe epilepsy, we refer Ms. G for repeat EEG and routine ophthalmologic exam, both of which are normal.

Perceptual disturbances are common in schizoaffective disorder and schizophrenia,^13-15 which we consider the most likely etiology of Ms. G’s photopsia, given her normal neurologic and ophthalmologic examinations. We switch her from aripiprazole to ziprasidone, 40 mg/d, discontinue phentermine, and taper and discontinue oxcarbazepine.

Related resources

• Ophthalmology Web. www.ophthalmologyweb.com.
• Aleman A, Larøi F. Hallucinations: the science of idiosyncratic perception. Washington, DC: American Psychological Association; 2008.

Drug Brand Names

• Aripiprazole • Abilify
• Clomiphene • Clomid, Serophene
• Digoxin • Lanoxin
• Nizatidine • Axid
• Olanzapine • Zyprexa
• Oxcarbazepine • Trileptal
• Paclitaxel • Onxol, Taxol
• Phentermine • Adipex-P, Inoamin
• Sertraline • Zoloft
• Topiramate • Topamax
• Ziprasidone • Geodon

Disclosure

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

HISTORY: Visual disturbances

Ms. G, age 30, has schizoaffective disorder and presents with worsening auditory hallucinations, paranoid delusions, and thought broadcasting and insertion, which is suspected to be related to a change from olanzapine to aripiprazole. She complains of visual disturbances—a hallucination of flashing lights—that she describes as occurring “all the time, like salt-and-pepper dancing, and sometimes splotches.” Her visual symptoms are worse when she closes her eyes and in dimly lit environments, but occur in daylight as well. Her vision is otherwise unimpaired.

The visual disturbances started with Ms. G’s first psychotic break at age 23 and have persisted continuously. She reports that at first, the spots were more intense, like an incessant strobe light. These symptoms are aggravated by sleep deprivation and anger and improve only with topiramate, 100 mg/d, which was prescribed a year earlier after a neurologic consultation but discontinued because of cognitive side effects.

Ms. G also complains of dull, aching, continuous headaches in the center of her forehead, which do not wake her, are not worse in the morning, and do not vary with the intensity of her visual symptoms. The headaches are incorporated into her system of delusions; she believes they “drain the electricity from her head” and make her feel better. She denies a history of migraines. Ms. G has a history of experiencing well-formed visual hallucinations, such as the devil, but has not had them for several years.

She had 5 to 10 episodes over the last 2 years that she describes as seizures—“an earthquake, a huge whooshing noise,” accompanied by heart palpitations that occur with stress, in public, in bright lights, or whenever she feels vulnerable. These episodes are not associated with loss of consciousness, incontinence, amnesia, or post-ictal confusion and she can stop them with relaxation techniques such as deep breathing and reminding herself to calm down. The last episode was 2 months ago and seemed unrelated to her visual phenomena. She also complains of occasional paresthesias in her extremities but denies dizziness, presyncope, or blurry or double vision.

Ms. G is obese and has no history of head injury or birth defects. Her family history is negative for epilepsy; however, her mother has bipolar disorder and father has schizophrenia. Her medications are phentermine, 37.5 mg/d for short-term management of obesity, oxcarbazepine, 600 mg/d for supposed partial seizures, sertraline, 100 mg/d for depression, aripiprazole, 10 mg/d, and nizatidine, 150 mg/d, for gastroesophageal reflux disease. She does not use alcohol, cigarettes, or illicit drugs. Ms. G obtained a master’s degree in health communications and works as a research assistant at a local medical school.

The authors’ observations

Visual hallucinations are not distortions of reality but original false perceptions that co-occur with real perceptions.¹ They vary from simple, elemental hallucinations involving flashes of light or geometric figures to more complex elaborations such as seeing crawling insects or a flock of angels.² Hallucinations can originate anywhere in the visual system from the lens to the visual cortex and can be vascular, toxic, electrophysiologic, structural, or neurochemical manifestations of neurologic, psychiatric, or ophthalmologic disorders. Depending on the hallucination’s etiology, treatment varies from emergent ophthalmologic examination and retinal surgery to antipsychotic therapy to simple observation. Photopsia is a perception of flashing lights, and its etiology varies; causes can be categorized as emergent or nonemergent.

Emergent causes of photopsia typically have a sudden onset, often can be diagnosed from associated symptoms or medical history, and must be recognized quickly to prevent permanent vision loss. Posterior vitreous detachment and retinal tearing can manifest as flashes of light, floaters, or blurred vision in the peripheral visual field that are worse in dim illumination and localizable by the patient. Retinal detachment can cause photopsia, floaters, or a “curtain” or “shadow” moving over the visual field, which may be accompanied by central or peripheral vision loss.³ Severe myopia is a risk factor for retinal detachment.

New-onset photopsia can suggest cytomegaloviral retinitis in an immune-compromised patient, and ocular-vascular crisis in a patient with sickle cell disease. New-onset or recurrent photopsia with bilateral blurred vision, ataxia, vertigo, dysarthria, or drop attacks may be caused by vertebrobasilar artery insufficiency (VBAI). Photopsia has been reported as the presenting symptom in carotid artery dissection.⁴ Photopsia correlated with eye movement implicates the retina or optic nerve. All of these etiologies except for VBAI should be referred to an ophthalmologist for indirect ophthalmoscopy and evaluation.

Nonemergent causes of photopsia include:

• drug toxicity
• migraine headache
• thyroid ophthalmopathy
• arteriovenous malformations
• seizures
• psychiatric disorders.

Digoxin, cocaine, paclitaxel, and clomiphene may have ocular side effects, including photopsia. The first symptoms of digitalis toxicity often are visual and include photopsia, yellow or green discoloration of the visual field, halos, and the appearance of frost over objects.⁵ Signs of cocaine intoxication can include visual hallucinations such as photopsia, shadows, moving objects, and insects crawling and co-occurring euphoria, hypervigilance, impaired judgment, and autonomic changes such as tachycardia, pupillary dilation, hypertension, and nausea.⁶ High-dose paclitaxel infusion has been reported to cause photopsia.⁷ Clomiphene can cause flickering lights and shimmering that persists after discontinuing the medication.⁸

Migraine with aura, which includes photopsia 39% of the time,⁹ often involves an expanding central visual disturbance or scotoma, and usually is confined to 1 visual field but may involve both. The aura typically lasts 10 to 20 minutes and often is followed by headache. Patients often report a history of episodic stereotyped auras and headache. Ocular examination is normal.

Ophthalmic migraine, also known as ocular or retinal migraine, is thought to be caused by transient vasospasm of the choroidal or retinal arteries and can be precipitated by postural changes, exercise, and oral contraceptives. It manifests as a gradual visual disturbance in a mosaic scotomata pattern that enlarges, producing total unilateral visual loss lasting from minutes to an hour, and—misleadingly—can be associated with minimal or no headache. Ocular examination is normal, and a personal or family history of migraine confirms the diagnosis.

Seizure activity in the occipital cortex and adjacent association areas can produce static light and stars. In a prospective study of 18 patients with occipital epilepsy, visual seizures lasted from a few seconds to 3 minutes—although rarely 20 to 150 minutes—and occurred in multiple clusters a day or week.¹⁰ All but 2 patients had secondary generalized tonic-clonic convulsions. Occipital seizures often are misdiagnosed as the visual aura of migraine, particularly when elementary visual hallucinations are followed by post-ictal headache and vomiting.¹¹ Lights with a color or a spherical shape suggest occipital epilepsy. Consider ordering an electroencephalogram (EEG) if the clinical diagnosis is unclear.

EXAMINATION: No obvious cause

During a mental status exam, Ms. G is pleasant, cooperative, alert, and oriented to person, place, and time. Her speech is fluent, her affect is full, and she describes her mood as a “sensitive, overwhelmed state.” Her thoughts occasionally are tangential, and she perseverates on guilty, embarrassed, and paranoid thoughts. She has auditory hallucinations and experiences photopsia during the interview, but denies other visual hallucinations. She shows no deficits in attention, concentration, memory, language, calculations, visuospatial abilities, or general fund of knowledge.

Her physical exam shows pupils that are equal, round, and reactive to light with normal extraocular movements, intact visual fields, 20/20 visual acuity in both eyes with glasses, and sharp optic disc margins with no retinal abnormalities apparent on funduscopic exam. The rest of her physical and neurologic exam is normal. Her complete blood count, electrolytes, kidney, liver function tests, urinalysis, and serum toxicology screens are normal. She has no history of immunodeficiency.

An MRI performed a year earlier showed mild diffuse congestion of a left maxillary sinus but no other intracranial abnormalities, and a waking EEG with sphenoidal electrodes was normal.

The authors’ observations

Evaluation of patients with photopsia includes taking a history and performing an eye examination including visual acuity testing, visual field testing, and funduscopic examination. The appearance, location, and duration of photopsia may help narrow the differential diagnosis and identify emergent cases (Table 1 and Table 2). Seizure activity associated with flashing lights suggests occipital epilepsy. A history of hypertension, diabetes, or polymyalgia rheumatica points to vascular causes, and hyperthyroidism suggests that flashing lights may be phosphenes of Graves’ disease. Myopia or eye trauma could indicate vitreous traction. Prescription medications and illicit drug use may point to a toxic etiology.¹²

Table 1
Diagnoses suggested by photopsia phenomena

Table 2
Photopsia differential diagnosis: Look for co-occurring symptoms

OUTCOME: Diagnosis of exclusion

Ms. G’s visual disturbance resembles vitreous detachment or classic migraine; however, the onset is not sudden, her funduscopic exam is normal, and she has no discrete episodes with concomitant headache. Ms. G’s improvement with topiramate is inconsistent with an ophthalmic origin because such photopsia should not improve with medication. It is consistent with recurrent seizures or status epilepticus of the occipital cortex; however, she did not experience multiple discrete episodes a day or generalized seizures experienced by most occipital seizure patients, and her EEG and clinical history are not consistent with seizures. Because we cannot rule out occipital lobe epilepsy, we refer Ms. G for repeat EEG and routine ophthalmologic exam, both of which are normal.

Perceptual disturbances are common in schizoaffective disorder and schizophrenia,^13-15 which we consider the most likely etiology of Ms. G’s photopsia, given her normal neurologic and ophthalmologic examinations. We switch her from aripiprazole to ziprasidone, 40 mg/d, discontinue phentermine, and taper and discontinue oxcarbazepine.

Related resources

• Ophthalmology Web. www.ophthalmologyweb.com.
• Aleman A, Larøi F. Hallucinations: the science of idiosyncratic perception. Washington, DC: American Psychological Association; 2008.

Drug Brand Names

• Aripiprazole • Abilify
• Clomiphene • Clomid, Serophene
• Digoxin • Lanoxin
• Nizatidine • Axid
• Olanzapine • Zyprexa
• Oxcarbazepine • Trileptal
• Paclitaxel • Onxol, Taxol
• Phentermine • Adipex-P, Inoamin
• Sertraline • Zoloft
• Topiramate • Topamax
• Ziprasidone • Geodon

Disclosure

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

HISTORY: Visual disturbances

Ms. G, age 30, has schizoaffective disorder and presents with worsening auditory hallucinations, paranoid delusions, and thought broadcasting and insertion, which is suspected to be related to a change from olanzapine to aripiprazole. She complains of visual disturbances—a hallucination of flashing lights—that she describes as occurring “all the time, like salt-and-pepper dancing, and sometimes splotches.” Her visual symptoms are worse when she closes her eyes and in dimly lit environments, but occur in daylight as well. Her vision is otherwise unimpaired.

The visual disturbances started with Ms. G’s first psychotic break at age 23 and have persisted continuously. She reports that at first, the spots were more intense, like an incessant strobe light. These symptoms are aggravated by sleep deprivation and anger and improve only with topiramate, 100 mg/d, which was prescribed a year earlier after a neurologic consultation but discontinued because of cognitive side effects.

Ms. G also complains of dull, aching, continuous headaches in the center of her forehead, which do not wake her, are not worse in the morning, and do not vary with the intensity of her visual symptoms. The headaches are incorporated into her system of delusions; she believes they “drain the electricity from her head” and make her feel better. She denies a history of migraines. Ms. G has a history of experiencing well-formed visual hallucinations, such as the devil, but has not had them for several years.

She had 5 to 10 episodes over the last 2 years that she describes as seizures—“an earthquake, a huge whooshing noise,” accompanied by heart palpitations that occur with stress, in public, in bright lights, or whenever she feels vulnerable. These episodes are not associated with loss of consciousness, incontinence, amnesia, or post-ictal confusion and she can stop them with relaxation techniques such as deep breathing and reminding herself to calm down. The last episode was 2 months ago and seemed unrelated to her visual phenomena. She also complains of occasional paresthesias in her extremities but denies dizziness, presyncope, or blurry or double vision.

Ms. G is obese and has no history of head injury or birth defects. Her family history is negative for epilepsy; however, her mother has bipolar disorder and father has schizophrenia. Her medications are phentermine, 37.5 mg/d for short-term management of obesity, oxcarbazepine, 600 mg/d for supposed partial seizures, sertraline, 100 mg/d for depression, aripiprazole, 10 mg/d, and nizatidine, 150 mg/d, for gastroesophageal reflux disease. She does not use alcohol, cigarettes, or illicit drugs. Ms. G obtained a master’s degree in health communications and works as a research assistant at a local medical school.

The authors’ observations

Visual hallucinations are not distortions of reality but original false perceptions that co-occur with real perceptions.¹ They vary from simple, elemental hallucinations involving flashes of light or geometric figures to more complex elaborations such as seeing crawling insects or a flock of angels.² Hallucinations can originate anywhere in the visual system from the lens to the visual cortex and can be vascular, toxic, electrophysiologic, structural, or neurochemical manifestations of neurologic, psychiatric, or ophthalmologic disorders. Depending on the hallucination’s etiology, treatment varies from emergent ophthalmologic examination and retinal surgery to antipsychotic therapy to simple observation. Photopsia is a perception of flashing lights, and its etiology varies; causes can be categorized as emergent or nonemergent.

Emergent causes of photopsia typically have a sudden onset, often can be diagnosed from associated symptoms or medical history, and must be recognized quickly to prevent permanent vision loss. Posterior vitreous detachment and retinal tearing can manifest as flashes of light, floaters, or blurred vision in the peripheral visual field that are worse in dim illumination and localizable by the patient. Retinal detachment can cause photopsia, floaters, or a “curtain” or “shadow” moving over the visual field, which may be accompanied by central or peripheral vision loss.³ Severe myopia is a risk factor for retinal detachment.

New-onset photopsia can suggest cytomegaloviral retinitis in an immune-compromised patient, and ocular-vascular crisis in a patient with sickle cell disease. New-onset or recurrent photopsia with bilateral blurred vision, ataxia, vertigo, dysarthria, or drop attacks may be caused by vertebrobasilar artery insufficiency (VBAI). Photopsia has been reported as the presenting symptom in carotid artery dissection.⁴ Photopsia correlated with eye movement implicates the retina or optic nerve. All of these etiologies except for VBAI should be referred to an ophthalmologist for indirect ophthalmoscopy and evaluation.

Nonemergent causes of photopsia include:

• drug toxicity
• migraine headache
• thyroid ophthalmopathy
• arteriovenous malformations
• seizures
• psychiatric disorders.

Digoxin, cocaine, paclitaxel, and clomiphene may have ocular side effects, including photopsia. The first symptoms of digitalis toxicity often are visual and include photopsia, yellow or green discoloration of the visual field, halos, and the appearance of frost over objects.⁵ Signs of cocaine intoxication can include visual hallucinations such as photopsia, shadows, moving objects, and insects crawling and co-occurring euphoria, hypervigilance, impaired judgment, and autonomic changes such as tachycardia, pupillary dilation, hypertension, and nausea.⁶ High-dose paclitaxel infusion has been reported to cause photopsia.⁷ Clomiphene can cause flickering lights and shimmering that persists after discontinuing the medication.⁸

Migraine with aura, which includes photopsia 39% of the time,⁹ often involves an expanding central visual disturbance or scotoma, and usually is confined to 1 visual field but may involve both. The aura typically lasts 10 to 20 minutes and often is followed by headache. Patients often report a history of episodic stereotyped auras and headache. Ocular examination is normal.

Ophthalmic migraine, also known as ocular or retinal migraine, is thought to be caused by transient vasospasm of the choroidal or retinal arteries and can be precipitated by postural changes, exercise, and oral contraceptives. It manifests as a gradual visual disturbance in a mosaic scotomata pattern that enlarges, producing total unilateral visual loss lasting from minutes to an hour, and—misleadingly—can be associated with minimal or no headache. Ocular examination is normal, and a personal or family history of migraine confirms the diagnosis.

Seizure activity in the occipital cortex and adjacent association areas can produce static light and stars. In a prospective study of 18 patients with occipital epilepsy, visual seizures lasted from a few seconds to 3 minutes—although rarely 20 to 150 minutes—and occurred in multiple clusters a day or week.¹⁰ All but 2 patients had secondary generalized tonic-clonic convulsions. Occipital seizures often are misdiagnosed as the visual aura of migraine, particularly when elementary visual hallucinations are followed by post-ictal headache and vomiting.¹¹ Lights with a color or a spherical shape suggest occipital epilepsy. Consider ordering an electroencephalogram (EEG) if the clinical diagnosis is unclear.

EXAMINATION: No obvious cause

During a mental status exam, Ms. G is pleasant, cooperative, alert, and oriented to person, place, and time. Her speech is fluent, her affect is full, and she describes her mood as a “sensitive, overwhelmed state.” Her thoughts occasionally are tangential, and she perseverates on guilty, embarrassed, and paranoid thoughts. She has auditory hallucinations and experiences photopsia during the interview, but denies other visual hallucinations. She shows no deficits in attention, concentration, memory, language, calculations, visuospatial abilities, or general fund of knowledge.

Her physical exam shows pupils that are equal, round, and reactive to light with normal extraocular movements, intact visual fields, 20/20 visual acuity in both eyes with glasses, and sharp optic disc margins with no retinal abnormalities apparent on funduscopic exam. The rest of her physical and neurologic exam is normal. Her complete blood count, electrolytes, kidney, liver function tests, urinalysis, and serum toxicology screens are normal. She has no history of immunodeficiency.

An MRI performed a year earlier showed mild diffuse congestion of a left maxillary sinus but no other intracranial abnormalities, and a waking EEG with sphenoidal electrodes was normal.

The authors’ observations

Evaluation of patients with photopsia includes taking a history and performing an eye examination including visual acuity testing, visual field testing, and funduscopic examination. The appearance, location, and duration of photopsia may help narrow the differential diagnosis and identify emergent cases (Table 1 and Table 2). Seizure activity associated with flashing lights suggests occipital epilepsy. A history of hypertension, diabetes, or polymyalgia rheumatica points to vascular causes, and hyperthyroidism suggests that flashing lights may be phosphenes of Graves’ disease. Myopia or eye trauma could indicate vitreous traction. Prescription medications and illicit drug use may point to a toxic etiology.¹²

Table 1
Diagnoses suggested by photopsia phenomena

Table 2
Photopsia differential diagnosis: Look for co-occurring symptoms

OUTCOME: Diagnosis of exclusion

Ms. G’s visual disturbance resembles vitreous detachment or classic migraine; however, the onset is not sudden, her funduscopic exam is normal, and she has no discrete episodes with concomitant headache. Ms. G’s improvement with topiramate is inconsistent with an ophthalmic origin because such photopsia should not improve with medication. It is consistent with recurrent seizures or status epilepticus of the occipital cortex; however, she did not experience multiple discrete episodes a day or generalized seizures experienced by most occipital seizure patients, and her EEG and clinical history are not consistent with seizures. Because we cannot rule out occipital lobe epilepsy, we refer Ms. G for repeat EEG and routine ophthalmologic exam, both of which are normal.

Perceptual disturbances are common in schizoaffective disorder and schizophrenia,^13-15 which we consider the most likely etiology of Ms. G’s photopsia, given her normal neurologic and ophthalmologic examinations. We switch her from aripiprazole to ziprasidone, 40 mg/d, discontinue phentermine, and taper and discontinue oxcarbazepine.

Related resources

• Ophthalmology Web. www.ophthalmologyweb.com.
• Aleman A, Larøi F. Hallucinations: the science of idiosyncratic perception. Washington, DC: American Psychological Association; 2008.

Drug Brand Names

• Aripiprazole • Abilify
• Clomiphene • Clomid, Serophene
• Digoxin • Lanoxin
• Nizatidine • Axid
• Olanzapine • Zyprexa
• Oxcarbazepine • Trileptal
• Paclitaxel • Onxol, Taxol
• Phentermine • Adipex-P, Inoamin
• Sertraline • Zoloft
• Topiramate • Topamax
• Ziprasidone • Geodon

Disclosure

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