Current Psychiatry

The FDA, with its complex and challenging mission, is constantly caught between Congress’ scrutiny and the public’s demands. It does not need nagging about doing more, but I’d like to suggest how the FDA could stimulate psychiatric drug discovery and encourage continuing medical education (CME). I propose 2 creative ideas as a long-time researcher with thorough knowledge of controlled clinical trials and an educator with extensive CME involvement.

Reward innovation in drug discovery

Like all psychiatric clinicians, I am painfully aware that many DSM-IVTR disorders have no FDA-approved medications. Within each of the existing classes of approved psychotropics—antipsychotics, antidepressants, and mood stabilizers—drugs tend to have similar mechanisms of action. Further, most available psychotropics are of limited effectiveness, and hardly any major therapeutic breakthroughs in psychiatry have been achieved in the past 50 years. Pharmaceutical companies seem content to settle for the safety of developing “me too” drugs and yet more formulations of existing agents.

Incentives can be powerful motivators for individuals and organizations to excel. So, to spur innovation by the pharmaceutical industry, I urge the FDA to extend the patent lives of breakthrough drugs (only those with completely new mechanisms of action) from the current 17 years to 25 years. I believe this very lucrative “carrot” will motivate every drug company to mobilize its resources and invest heavily in financially risky but innovative research and development.

Excellence deserves to be differentially rewarded, and outstanding drug discovery should be no exception. Promising an 8-year patent “bonus” extension could generate a tsunami of first-in-class drugs and cures for diseases that today’s medications treat inadequately or not at all. Yes, the pharmaceutical companies would make higher profits, but the ultimate winners would be patients whose lives are improved and extended by these novel medications.

Support CME collaboratively

A second idea occurred to me after I read the Josiah Macy, Jr. Foundation’s November 2007 report recommending the unthinkable: that CME be completely divorced from the financial life-blood of pharmaceutical sponsorship! More than 60% of financial support for accredited CME activities in medicine comes from pharmaceutical and medical device companies ($1.45 billion of a total $2.4 billion in 2006).¹

The Macy Foundation’s report acknowledges that “abrupt cessation of all such support would impose unacceptable hardship” on many professional organizations and institutions, and it proposes a 5-year phase-out period.¹ This ban on commercial support would include grand rounds and symposia organized by medical schools, which provide almost no funds for CME in their limited budgets.

How can the FDA help? I propose that the FDA support CME nationwide by pooling pharmaceutical company contributions in a not-for-profit independent fund and appoint the Accreditation Council for Continuing Medical Education (ACCME) as overseer. The ACCME would evaluate applications from all specialties and allocate funds to support meritorious CME programs.

For each year that a pharmaceutical company contributes to the CME fund, the FDA would grant a 3-month patent extension on all of the company’s drugs. This extension could yield drug companies more than $1 billion in additional sales for blockbusters such as atypical antipsychotics.

CME could be adequately funded, but without today’s perceived tarnish—whether deserved or not—of pharmaceutical influence. Pharmaceutical companies would no longer directly sponsor CME programs but would continue to be important philanthropic partners. Contributions from dozens of companies would fund the vital activities by which physicians and nurses keep pace with advances in the diagnosis and treatment of disease.

Your comments?

My suggestions are intended to start a dialogue about financial support for two critical needs in psychiatry: research to develop new psychiatric medications and support for continuing education. To advance, we must break with stale models and exploit reasonable solutions. I invite you to send your comments to me at [email protected]

The FDA, with its complex and challenging mission, is constantly caught between Congress’ scrutiny and the public’s demands. It does not need nagging about doing more, but I’d like to suggest how the FDA could stimulate psychiatric drug discovery and encourage continuing medical education (CME). I propose 2 creative ideas as a long-time researcher with thorough knowledge of controlled clinical trials and an educator with extensive CME involvement.

Reward innovation in drug discovery

Like all psychiatric clinicians, I am painfully aware that many DSM-IVTR disorders have no FDA-approved medications. Within each of the existing classes of approved psychotropics—antipsychotics, antidepressants, and mood stabilizers—drugs tend to have similar mechanisms of action. Further, most available psychotropics are of limited effectiveness, and hardly any major therapeutic breakthroughs in psychiatry have been achieved in the past 50 years. Pharmaceutical companies seem content to settle for the safety of developing “me too” drugs and yet more formulations of existing agents.

Incentives can be powerful motivators for individuals and organizations to excel. So, to spur innovation by the pharmaceutical industry, I urge the FDA to extend the patent lives of breakthrough drugs (only those with completely new mechanisms of action) from the current 17 years to 25 years. I believe this very lucrative “carrot” will motivate every drug company to mobilize its resources and invest heavily in financially risky but innovative research and development.

Excellence deserves to be differentially rewarded, and outstanding drug discovery should be no exception. Promising an 8-year patent “bonus” extension could generate a tsunami of first-in-class drugs and cures for diseases that today’s medications treat inadequately or not at all. Yes, the pharmaceutical companies would make higher profits, but the ultimate winners would be patients whose lives are improved and extended by these novel medications.

Support CME collaboratively

A second idea occurred to me after I read the Josiah Macy, Jr. Foundation’s November 2007 report recommending the unthinkable: that CME be completely divorced from the financial life-blood of pharmaceutical sponsorship! More than 60% of financial support for accredited CME activities in medicine comes from pharmaceutical and medical device companies ($1.45 billion of a total $2.4 billion in 2006).¹

The Macy Foundation’s report acknowledges that “abrupt cessation of all such support would impose unacceptable hardship” on many professional organizations and institutions, and it proposes a 5-year phase-out period.¹ This ban on commercial support would include grand rounds and symposia organized by medical schools, which provide almost no funds for CME in their limited budgets.

How can the FDA help? I propose that the FDA support CME nationwide by pooling pharmaceutical company contributions in a not-for-profit independent fund and appoint the Accreditation Council for Continuing Medical Education (ACCME) as overseer. The ACCME would evaluate applications from all specialties and allocate funds to support meritorious CME programs.

For each year that a pharmaceutical company contributes to the CME fund, the FDA would grant a 3-month patent extension on all of the company’s drugs. This extension could yield drug companies more than $1 billion in additional sales for blockbusters such as atypical antipsychotics.

CME could be adequately funded, but without today’s perceived tarnish—whether deserved or not—of pharmaceutical influence. Pharmaceutical companies would no longer directly sponsor CME programs but would continue to be important philanthropic partners. Contributions from dozens of companies would fund the vital activities by which physicians and nurses keep pace with advances in the diagnosis and treatment of disease.

Your comments?

My suggestions are intended to start a dialogue about financial support for two critical needs in psychiatry: research to develop new psychiatric medications and support for continuing education. To advance, we must break with stale models and exploit reasonable solutions. I invite you to send your comments to me at [email protected]

The FDA, with its complex and challenging mission, is constantly caught between Congress’ scrutiny and the public’s demands. It does not need nagging about doing more, but I’d like to suggest how the FDA could stimulate psychiatric drug discovery and encourage continuing medical education (CME). I propose 2 creative ideas as a long-time researcher with thorough knowledge of controlled clinical trials and an educator with extensive CME involvement.

Reward innovation in drug discovery

Like all psychiatric clinicians, I am painfully aware that many DSM-IVTR disorders have no FDA-approved medications. Within each of the existing classes of approved psychotropics—antipsychotics, antidepressants, and mood stabilizers—drugs tend to have similar mechanisms of action. Further, most available psychotropics are of limited effectiveness, and hardly any major therapeutic breakthroughs in psychiatry have been achieved in the past 50 years. Pharmaceutical companies seem content to settle for the safety of developing “me too” drugs and yet more formulations of existing agents.

Incentives can be powerful motivators for individuals and organizations to excel. So, to spur innovation by the pharmaceutical industry, I urge the FDA to extend the patent lives of breakthrough drugs (only those with completely new mechanisms of action) from the current 17 years to 25 years. I believe this very lucrative “carrot” will motivate every drug company to mobilize its resources and invest heavily in financially risky but innovative research and development.

Excellence deserves to be differentially rewarded, and outstanding drug discovery should be no exception. Promising an 8-year patent “bonus” extension could generate a tsunami of first-in-class drugs and cures for diseases that today’s medications treat inadequately or not at all. Yes, the pharmaceutical companies would make higher profits, but the ultimate winners would be patients whose lives are improved and extended by these novel medications.

Support CME collaboratively

A second idea occurred to me after I read the Josiah Macy, Jr. Foundation’s November 2007 report recommending the unthinkable: that CME be completely divorced from the financial life-blood of pharmaceutical sponsorship! More than 60% of financial support for accredited CME activities in medicine comes from pharmaceutical and medical device companies ($1.45 billion of a total $2.4 billion in 2006).¹

The Macy Foundation’s report acknowledges that “abrupt cessation of all such support would impose unacceptable hardship” on many professional organizations and institutions, and it proposes a 5-year phase-out period.¹ This ban on commercial support would include grand rounds and symposia organized by medical schools, which provide almost no funds for CME in their limited budgets.

How can the FDA help? I propose that the FDA support CME nationwide by pooling pharmaceutical company contributions in a not-for-profit independent fund and appoint the Accreditation Council for Continuing Medical Education (ACCME) as overseer. The ACCME would evaluate applications from all specialties and allocate funds to support meritorious CME programs.

For each year that a pharmaceutical company contributes to the CME fund, the FDA would grant a 3-month patent extension on all of the company’s drugs. This extension could yield drug companies more than $1 billion in additional sales for blockbusters such as atypical antipsychotics.

CME could be adequately funded, but without today’s perceived tarnish—whether deserved or not—of pharmaceutical influence. Pharmaceutical companies would no longer directly sponsor CME programs but would continue to be important philanthropic partners. Contributions from dozens of companies would fund the vital activities by which physicians and nurses keep pace with advances in the diagnosis and treatment of disease.

Your comments?

My suggestions are intended to start a dialogue about financial support for two critical needs in psychiatry: research to develop new psychiatric medications and support for continuing education. To advance, we must break with stale models and exploit reasonable solutions. I invite you to send your comments to me at [email protected]

CASE: She’s not herself

Mrs. M, age 74, is brought to the ER by her husband after he finds her lying on their bedroom floor, incoherent and extremely drowsy. He reports that his wife, who suffers chronic arthritic back and joint pain, might have overdosed on pain medications.

According to her husband, Mrs. M has been taking combination oxycodone/acetaminophen and transdermal fentanyl at unknown dosages, but he is unsure when she started using these medications or if she is taking others. Serum toxicology screening shows twice the normal values for opioids and benzodiazepines; other laboratory results are normal.

Mrs. M is medically stable but her mental status is altered. She is oblivious to time, place and person, speaks to no one, and seems lost in her own world. The hospital’s medical service admits Mrs. M for stabilization and to determine whether the overdose was intentional.

Two days later, we evaluate Mrs. M’s mental status at the attending physician’s request. She appears confused and cannot answer our questions. Her husband tells us she was “doing fine” until approximately 4 months ago, when she started becoming increasingly forgetful and lethargic. He says she has been forgetting routine chores such as paying bills and grocery shopping. Recently, she has been getting lost during her evening walk; neighbors often help her find her way home.

The authors’ observations

Two opioid medications—oxycodone/acetaminophen combination and transdermal fentanyl—are commonly used to manage moderate or severe pain from any type of chronic arthritis.

• Oxycodone, a semisynthetic opioid analgesic indicated for moderate to moderately severe pain, is used when nondrug measures and nonnarcotic medications do not control the pain.
  
• Transdermal fentanyl, a potent analgesic indicated for persistent moderate to severe chronic pain, typically is prescribed to patients who tolerate oral oxycodone, 30 mg/d; morphine, 60 mg; hydromorphone, 8 mg; or an equianalgesic dosage of another opioid for ≥1 week.
  

Mrs. M also was taking a benzodiazepine, but which medication—and why she was taking it—were unclear. She had no psychiatric diagnosis, and her husband could not recall her medication history.

We also cannot explain Mrs. M’s negative cognitive and behavioral changes. Opioid overuse and onset of dementia-related cognitive decline are possibilities.

TRANSFER Why is she confused?

Based on information from the pharmacy department, doctors at the medical unit restart oxycodone/acetaminophen, 7.5/325 mg tid, and transdermal fentanyl, 25 mcg/hr every 3 days. After discussing how to treat Mrs. M, the psychiatric and medical services transfer her to the geriatric psychiatric inpatient unit 3 days after admission.

We visit Mrs. M hours after her transfer. She seems lethargic but not confused, although Mini-Mental State Examination (MMSE) score of 15 suggests moderate cognitive impairment. Vitamin B₁₂ and thyroid levels, erythrocyte sedimentation rate, and syphilis test results are normal, allowing us to rule out organic causes for her dementia. Brain MRI shows no neurologic damage. On a scale of 1 to 5 with 5 being most severe, Mrs. M scores her pain as 2 (mild) and her sedation as 3 (moderate).

With Mrs. M’s permission, we call her primary care physician for collateral information. The physician tells us Mrs. M has suffered severe joint pain for 2 years. Nonnarcotic medications and treatments—including counseling, support groups, massage, yoga, exercise, biofeedback, relaxation therapy, and physical therapy—were ineffective.

Approximately 10 months ago, the physician started oxycodone/acetaminophen at 2.5/325 mg bid and titrated it over 6 weeks to 7.5/325 mg tid for Mrs. M’s persistent joint pain. Four months ago, with her pain still severe, the physician added transdermal fentanyl, 25 mcg/hr every 3 days, after which the patient reported mild improvement.

One month after starting the fentanyl patch, Mrs. M complained of sudden forgetfulness, low energy, poor concentration, and increased sleep. The physician suspected depression with possible comorbid anxiety and prescribed sertraline, 50 mg/d, and alprazolam, 0.5 mg bid. Mrs. M stopped sertraline after 3 days because it was causing diarrhea but kept taking alprazolam.

Mrs. M saw her primary care physician once after starting alprazolam and sertraline but missed her most recent appointment last month. The physician says he inadvertently approved at least 1 premature request for an alprazolam refill.

Six days after admission, Mrs. M’s sedation, cognitive impairment, and lethargy persist. She reports no mood and anxiety problems, and we have not restarted alprazolam.

The authors’ observations

The fentanyl patch most likely began to diminish Mrs. M’s alertness soon after she started using it. The doctor, however, mistook cognitive slowing for new-onset depression or anxiety. Depressive symptoms can imitate dementia, but Mrs. M’s severe sedation and denial of depressive symptoms suggest a medication side effect.

The primary care physician’s reconstruction of Mrs. M’s history explained her positive benzodiazepine reading, and her use of the short-acting benzodiazepine alprazolam could account for her sudden-onset paranoia and cognitive decline (Box). Benzodiazepines can cause behavioral side effects such as disinhibition, agitation, or paranoia, and patients age ≥65 are at increased risk for these side effects.⁴ In particular, benzodiazepines with half-lives ≥6 to 8 hours such as clonazepam and oxazepam can cause short-term memory impairment, confusion, and delirium.^5-7

Box

When prescribing benzodiazepines (especially in older patients) watch for signs of overuse or abuse, such as early requests for refills, unkempt appearance, excessive sleepiness, or agitation (Table 1).

Warning signs of opioid, benzodiazepine overuse

The authors’ observations

Persistent chronic pain in the elderly can diminish health and quality of life, resulting in depression, social isolation, immobility, and sleep disturbance.

Managing an older patient’s pain can be challenging (Table 2). Opioids are effective painkillers, but even at relatively low dosages they can diminish function and cognition and increase risk of delirium. Also, patients’ ability to tolerate different opioids at different dosages varies widely.

Mrs. M’s opioid regimen was intolerable and numerous other treatments did not alleviate her pain. At this point, replacing fentanyl with another opioid was our best option.⁸

We decided to try methadone, which is indicated for moderate to severe pain that does not respond to nonnarcotic treatments. Methadone often is used for chronic pain associated with arthritis or malignancy.

Methadone is less sedating, more tolerable, and carries a lower risk of cognitive side effects than other opioids. Methadone also is fast- and long-acting—its analgesic effects begin within 30 minutes to 1 hour of oral administration⁹ and last approximately 12 hours, thus reducing the risk of breakthrough pain. Methadone also:

• has no active metabolites, which decreases the risk of hepatic side effects
  
• offers a high volume of distribution, thus allowing clinical effect with minimal dosing.
  

• watch for symptoms such as oversedation, memory and concentration problems, and sudden changes in personality
  
• call you to clarify if these symptoms are methadone side effects.
  

Watch for other potential side effects of methadone, such as constipation, sedation, breakthrough pain, sexual dysfunction, decreased immunity, respiratory depression, or prolonged corrected QT intervals.

Patients usually tolerate an immediate switch from transdermal fentanyl to methadone, but a sudden switch from high-dose fentanyl can reduce methadone’s effectiveness. Starting methadone at a high dosage to compensate for loss of effectiveness could increase side effect risk. If the fentanyl dosage exceeds 100 mcg/hr, taper by 25 mcg weekly. Simultaneously start methadone at a low dosage and titrate by 5 to 10 mg weekly as needed.

Table 2

Chronic pain management in the elderly: Dos and don’ts

TREATMENT Medication change

We stop transdermal fentanyl and start oral methadone, 5 mg bid, while continuing oxycodone/acetaminophen at the previous dosage.

Two days later, Mrs. M is much more alert. Since admission 1 week ago, her sedation rating has improved from 3 (mildly sedated) to 4 (almost fully alert). She rates her pain as mild and reports no breakthrough pain or other side effects from methadone. Her MMSE score has improved to 24—suggesting close to normal cognition—and she is much more interactive with staff and family.

Eight days after we start methadone, we stop oxycodone/acetaminophen and increase methadone to 10 mg bid to further improve cognition and alertness and to see if 1 pain medication is suffcient. Two days later, we discharge Mrs. M. She is fully alert, feels little or no joint pain, and is tolerating the methadone increase.

At outpatient follow-up 4 weeks later, Mrs. M remains pain-free and her MMSE score is 29, suggesting normal cognition. Over 8 months, we continue to see her monthly and then bi-monthly, after which we refer her to her primary care physician.

Related resources

• American Association for Geriatric Psychiatry. www.aagponline.org.
  
• American Pain Society. www.ampainsoc.org.
  

• Alprazolam • Xanax
  
• Clonazepam • Klonopin
  
• Fentanyl (transdermal) • Duragesic
  
• Hydromorphone • Dilaudid
  
• Meperidine • Demerol
  
• Methadone • Dolophine
  
• Oxazepam • Serax
  
• Oxycodone • OxyContin, Roxicodone
  
• Oxycodone/acetaminophen • Percocet
  
• Propoxyphene • Darvon
  
• Sertraline • Zoloft
  

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

CASE: She’s not herself

Mrs. M, age 74, is brought to the ER by her husband after he finds her lying on their bedroom floor, incoherent and extremely drowsy. He reports that his wife, who suffers chronic arthritic back and joint pain, might have overdosed on pain medications.

According to her husband, Mrs. M has been taking combination oxycodone/acetaminophen and transdermal fentanyl at unknown dosages, but he is unsure when she started using these medications or if she is taking others. Serum toxicology screening shows twice the normal values for opioids and benzodiazepines; other laboratory results are normal.

Mrs. M is medically stable but her mental status is altered. She is oblivious to time, place and person, speaks to no one, and seems lost in her own world. The hospital’s medical service admits Mrs. M for stabilization and to determine whether the overdose was intentional.

Two days later, we evaluate Mrs. M’s mental status at the attending physician’s request. She appears confused and cannot answer our questions. Her husband tells us she was “doing fine” until approximately 4 months ago, when she started becoming increasingly forgetful and lethargic. He says she has been forgetting routine chores such as paying bills and grocery shopping. Recently, she has been getting lost during her evening walk; neighbors often help her find her way home.

The authors’ observations

Two opioid medications—oxycodone/acetaminophen combination and transdermal fentanyl—are commonly used to manage moderate or severe pain from any type of chronic arthritis.

• Oxycodone, a semisynthetic opioid analgesic indicated for moderate to moderately severe pain, is used when nondrug measures and nonnarcotic medications do not control the pain.
  
• Transdermal fentanyl, a potent analgesic indicated for persistent moderate to severe chronic pain, typically is prescribed to patients who tolerate oral oxycodone, 30 mg/d; morphine, 60 mg; hydromorphone, 8 mg; or an equianalgesic dosage of another opioid for ≥1 week.
  

Mrs. M also was taking a benzodiazepine, but which medication—and why she was taking it—were unclear. She had no psychiatric diagnosis, and her husband could not recall her medication history.

We also cannot explain Mrs. M’s negative cognitive and behavioral changes. Opioid overuse and onset of dementia-related cognitive decline are possibilities.

TRANSFER Why is she confused?

Based on information from the pharmacy department, doctors at the medical unit restart oxycodone/acetaminophen, 7.5/325 mg tid, and transdermal fentanyl, 25 mcg/hr every 3 days. After discussing how to treat Mrs. M, the psychiatric and medical services transfer her to the geriatric psychiatric inpatient unit 3 days after admission.

We visit Mrs. M hours after her transfer. She seems lethargic but not confused, although Mini-Mental State Examination (MMSE) score of 15 suggests moderate cognitive impairment. Vitamin B₁₂ and thyroid levels, erythrocyte sedimentation rate, and syphilis test results are normal, allowing us to rule out organic causes for her dementia. Brain MRI shows no neurologic damage. On a scale of 1 to 5 with 5 being most severe, Mrs. M scores her pain as 2 (mild) and her sedation as 3 (moderate).

With Mrs. M’s permission, we call her primary care physician for collateral information. The physician tells us Mrs. M has suffered severe joint pain for 2 years. Nonnarcotic medications and treatments—including counseling, support groups, massage, yoga, exercise, biofeedback, relaxation therapy, and physical therapy—were ineffective.

Approximately 10 months ago, the physician started oxycodone/acetaminophen at 2.5/325 mg bid and titrated it over 6 weeks to 7.5/325 mg tid for Mrs. M’s persistent joint pain. Four months ago, with her pain still severe, the physician added transdermal fentanyl, 25 mcg/hr every 3 days, after which the patient reported mild improvement.

One month after starting the fentanyl patch, Mrs. M complained of sudden forgetfulness, low energy, poor concentration, and increased sleep. The physician suspected depression with possible comorbid anxiety and prescribed sertraline, 50 mg/d, and alprazolam, 0.5 mg bid. Mrs. M stopped sertraline after 3 days because it was causing diarrhea but kept taking alprazolam.

Mrs. M saw her primary care physician once after starting alprazolam and sertraline but missed her most recent appointment last month. The physician says he inadvertently approved at least 1 premature request for an alprazolam refill.

Six days after admission, Mrs. M’s sedation, cognitive impairment, and lethargy persist. She reports no mood and anxiety problems, and we have not restarted alprazolam.

The authors’ observations

The fentanyl patch most likely began to diminish Mrs. M’s alertness soon after she started using it. The doctor, however, mistook cognitive slowing for new-onset depression or anxiety. Depressive symptoms can imitate dementia, but Mrs. M’s severe sedation and denial of depressive symptoms suggest a medication side effect.

The primary care physician’s reconstruction of Mrs. M’s history explained her positive benzodiazepine reading, and her use of the short-acting benzodiazepine alprazolam could account for her sudden-onset paranoia and cognitive decline (Box). Benzodiazepines can cause behavioral side effects such as disinhibition, agitation, or paranoia, and patients age ≥65 are at increased risk for these side effects.⁴ In particular, benzodiazepines with half-lives ≥6 to 8 hours such as clonazepam and oxazepam can cause short-term memory impairment, confusion, and delirium.^5-7

Box

When prescribing benzodiazepines (especially in older patients) watch for signs of overuse or abuse, such as early requests for refills, unkempt appearance, excessive sleepiness, or agitation (Table 1).

Warning signs of opioid, benzodiazepine overuse

The authors’ observations

Persistent chronic pain in the elderly can diminish health and quality of life, resulting in depression, social isolation, immobility, and sleep disturbance.

Managing an older patient’s pain can be challenging (Table 2). Opioids are effective painkillers, but even at relatively low dosages they can diminish function and cognition and increase risk of delirium. Also, patients’ ability to tolerate different opioids at different dosages varies widely.

Mrs. M’s opioid regimen was intolerable and numerous other treatments did not alleviate her pain. At this point, replacing fentanyl with another opioid was our best option.⁸

We decided to try methadone, which is indicated for moderate to severe pain that does not respond to nonnarcotic treatments. Methadone often is used for chronic pain associated with arthritis or malignancy.

Methadone is less sedating, more tolerable, and carries a lower risk of cognitive side effects than other opioids. Methadone also is fast- and long-acting—its analgesic effects begin within 30 minutes to 1 hour of oral administration⁹ and last approximately 12 hours, thus reducing the risk of breakthrough pain. Methadone also:

• has no active metabolites, which decreases the risk of hepatic side effects
  
• offers a high volume of distribution, thus allowing clinical effect with minimal dosing.
  

• watch for symptoms such as oversedation, memory and concentration problems, and sudden changes in personality
  
• call you to clarify if these symptoms are methadone side effects.
  

Watch for other potential side effects of methadone, such as constipation, sedation, breakthrough pain, sexual dysfunction, decreased immunity, respiratory depression, or prolonged corrected QT intervals.

Patients usually tolerate an immediate switch from transdermal fentanyl to methadone, but a sudden switch from high-dose fentanyl can reduce methadone’s effectiveness. Starting methadone at a high dosage to compensate for loss of effectiveness could increase side effect risk. If the fentanyl dosage exceeds 100 mcg/hr, taper by 25 mcg weekly. Simultaneously start methadone at a low dosage and titrate by 5 to 10 mg weekly as needed.

Table 2

Chronic pain management in the elderly: Dos and don’ts

TREATMENT Medication change

We stop transdermal fentanyl and start oral methadone, 5 mg bid, while continuing oxycodone/acetaminophen at the previous dosage.

Two days later, Mrs. M is much more alert. Since admission 1 week ago, her sedation rating has improved from 3 (mildly sedated) to 4 (almost fully alert). She rates her pain as mild and reports no breakthrough pain or other side effects from methadone. Her MMSE score has improved to 24—suggesting close to normal cognition—and she is much more interactive with staff and family.

Eight days after we start methadone, we stop oxycodone/acetaminophen and increase methadone to 10 mg bid to further improve cognition and alertness and to see if 1 pain medication is suffcient. Two days later, we discharge Mrs. M. She is fully alert, feels little or no joint pain, and is tolerating the methadone increase.

At outpatient follow-up 4 weeks later, Mrs. M remains pain-free and her MMSE score is 29, suggesting normal cognition. Over 8 months, we continue to see her monthly and then bi-monthly, after which we refer her to her primary care physician.

Related resources

• American Association for Geriatric Psychiatry. www.aagponline.org.
  
• American Pain Society. www.ampainsoc.org.
  

• Alprazolam • Xanax
  
• Clonazepam • Klonopin
  
• Fentanyl (transdermal) • Duragesic
  
• Hydromorphone • Dilaudid
  
• Meperidine • Demerol
  
• Methadone • Dolophine
  
• Oxazepam • Serax
  
• Oxycodone • OxyContin, Roxicodone
  
• Oxycodone/acetaminophen • Percocet
  
• Propoxyphene • Darvon
  
• Sertraline • Zoloft
  

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

CASE: She’s not herself

Mrs. M, age 74, is brought to the ER by her husband after he finds her lying on their bedroom floor, incoherent and extremely drowsy. He reports that his wife, who suffers chronic arthritic back and joint pain, might have overdosed on pain medications.

According to her husband, Mrs. M has been taking combination oxycodone/acetaminophen and transdermal fentanyl at unknown dosages, but he is unsure when she started using these medications or if she is taking others. Serum toxicology screening shows twice the normal values for opioids and benzodiazepines; other laboratory results are normal.

Mrs. M is medically stable but her mental status is altered. She is oblivious to time, place and person, speaks to no one, and seems lost in her own world. The hospital’s medical service admits Mrs. M for stabilization and to determine whether the overdose was intentional.

Two days later, we evaluate Mrs. M’s mental status at the attending physician’s request. She appears confused and cannot answer our questions. Her husband tells us she was “doing fine” until approximately 4 months ago, when she started becoming increasingly forgetful and lethargic. He says she has been forgetting routine chores such as paying bills and grocery shopping. Recently, she has been getting lost during her evening walk; neighbors often help her find her way home.

The authors’ observations

Two opioid medications—oxycodone/acetaminophen combination and transdermal fentanyl—are commonly used to manage moderate or severe pain from any type of chronic arthritis.

• Oxycodone, a semisynthetic opioid analgesic indicated for moderate to moderately severe pain, is used when nondrug measures and nonnarcotic medications do not control the pain.
  
• Transdermal fentanyl, a potent analgesic indicated for persistent moderate to severe chronic pain, typically is prescribed to patients who tolerate oral oxycodone, 30 mg/d; morphine, 60 mg; hydromorphone, 8 mg; or an equianalgesic dosage of another opioid for ≥1 week.
  

Mrs. M also was taking a benzodiazepine, but which medication—and why she was taking it—were unclear. She had no psychiatric diagnosis, and her husband could not recall her medication history.

We also cannot explain Mrs. M’s negative cognitive and behavioral changes. Opioid overuse and onset of dementia-related cognitive decline are possibilities.

TRANSFER Why is she confused?

Based on information from the pharmacy department, doctors at the medical unit restart oxycodone/acetaminophen, 7.5/325 mg tid, and transdermal fentanyl, 25 mcg/hr every 3 days. After discussing how to treat Mrs. M, the psychiatric and medical services transfer her to the geriatric psychiatric inpatient unit 3 days after admission.

We visit Mrs. M hours after her transfer. She seems lethargic but not confused, although Mini-Mental State Examination (MMSE) score of 15 suggests moderate cognitive impairment. Vitamin B₁₂ and thyroid levels, erythrocyte sedimentation rate, and syphilis test results are normal, allowing us to rule out organic causes for her dementia. Brain MRI shows no neurologic damage. On a scale of 1 to 5 with 5 being most severe, Mrs. M scores her pain as 2 (mild) and her sedation as 3 (moderate).

With Mrs. M’s permission, we call her primary care physician for collateral information. The physician tells us Mrs. M has suffered severe joint pain for 2 years. Nonnarcotic medications and treatments—including counseling, support groups, massage, yoga, exercise, biofeedback, relaxation therapy, and physical therapy—were ineffective.

Approximately 10 months ago, the physician started oxycodone/acetaminophen at 2.5/325 mg bid and titrated it over 6 weeks to 7.5/325 mg tid for Mrs. M’s persistent joint pain. Four months ago, with her pain still severe, the physician added transdermal fentanyl, 25 mcg/hr every 3 days, after which the patient reported mild improvement.

One month after starting the fentanyl patch, Mrs. M complained of sudden forgetfulness, low energy, poor concentration, and increased sleep. The physician suspected depression with possible comorbid anxiety and prescribed sertraline, 50 mg/d, and alprazolam, 0.5 mg bid. Mrs. M stopped sertraline after 3 days because it was causing diarrhea but kept taking alprazolam.

Mrs. M saw her primary care physician once after starting alprazolam and sertraline but missed her most recent appointment last month. The physician says he inadvertently approved at least 1 premature request for an alprazolam refill.

Six days after admission, Mrs. M’s sedation, cognitive impairment, and lethargy persist. She reports no mood and anxiety problems, and we have not restarted alprazolam.

The authors’ observations

The fentanyl patch most likely began to diminish Mrs. M’s alertness soon after she started using it. The doctor, however, mistook cognitive slowing for new-onset depression or anxiety. Depressive symptoms can imitate dementia, but Mrs. M’s severe sedation and denial of depressive symptoms suggest a medication side effect.

The primary care physician’s reconstruction of Mrs. M’s history explained her positive benzodiazepine reading, and her use of the short-acting benzodiazepine alprazolam could account for her sudden-onset paranoia and cognitive decline (Box). Benzodiazepines can cause behavioral side effects such as disinhibition, agitation, or paranoia, and patients age ≥65 are at increased risk for these side effects.⁴ In particular, benzodiazepines with half-lives ≥6 to 8 hours such as clonazepam and oxazepam can cause short-term memory impairment, confusion, and delirium.^5-7

Box

When prescribing benzodiazepines (especially in older patients) watch for signs of overuse or abuse, such as early requests for refills, unkempt appearance, excessive sleepiness, or agitation (Table 1).

Warning signs of opioid, benzodiazepine overuse

The authors’ observations

Persistent chronic pain in the elderly can diminish health and quality of life, resulting in depression, social isolation, immobility, and sleep disturbance.

Managing an older patient’s pain can be challenging (Table 2). Opioids are effective painkillers, but even at relatively low dosages they can diminish function and cognition and increase risk of delirium. Also, patients’ ability to tolerate different opioids at different dosages varies widely.

Mrs. M’s opioid regimen was intolerable and numerous other treatments did not alleviate her pain. At this point, replacing fentanyl with another opioid was our best option.⁸

We decided to try methadone, which is indicated for moderate to severe pain that does not respond to nonnarcotic treatments. Methadone often is used for chronic pain associated with arthritis or malignancy.

Methadone is less sedating, more tolerable, and carries a lower risk of cognitive side effects than other opioids. Methadone also is fast- and long-acting—its analgesic effects begin within 30 minutes to 1 hour of oral administration⁹ and last approximately 12 hours, thus reducing the risk of breakthrough pain. Methadone also:

• has no active metabolites, which decreases the risk of hepatic side effects
  
• offers a high volume of distribution, thus allowing clinical effect with minimal dosing.
  

• watch for symptoms such as oversedation, memory and concentration problems, and sudden changes in personality
  
• call you to clarify if these symptoms are methadone side effects.
  

Watch for other potential side effects of methadone, such as constipation, sedation, breakthrough pain, sexual dysfunction, decreased immunity, respiratory depression, or prolonged corrected QT intervals.

Patients usually tolerate an immediate switch from transdermal fentanyl to methadone, but a sudden switch from high-dose fentanyl can reduce methadone’s effectiveness. Starting methadone at a high dosage to compensate for loss of effectiveness could increase side effect risk. If the fentanyl dosage exceeds 100 mcg/hr, taper by 25 mcg weekly. Simultaneously start methadone at a low dosage and titrate by 5 to 10 mg weekly as needed.

Table 2

Chronic pain management in the elderly: Dos and don’ts

TREATMENT Medication change

We stop transdermal fentanyl and start oral methadone, 5 mg bid, while continuing oxycodone/acetaminophen at the previous dosage.

Two days later, Mrs. M is much more alert. Since admission 1 week ago, her sedation rating has improved from 3 (mildly sedated) to 4 (almost fully alert). She rates her pain as mild and reports no breakthrough pain or other side effects from methadone. Her MMSE score has improved to 24—suggesting close to normal cognition—and she is much more interactive with staff and family.

Eight days after we start methadone, we stop oxycodone/acetaminophen and increase methadone to 10 mg bid to further improve cognition and alertness and to see if 1 pain medication is suffcient. Two days later, we discharge Mrs. M. She is fully alert, feels little or no joint pain, and is tolerating the methadone increase.

At outpatient follow-up 4 weeks later, Mrs. M remains pain-free and her MMSE score is 29, suggesting normal cognition. Over 8 months, we continue to see her monthly and then bi-monthly, after which we refer her to her primary care physician.

Related resources

• American Association for Geriatric Psychiatry. www.aagponline.org.
  
• American Pain Society. www.ampainsoc.org.
  

• Alprazolam • Xanax
  
• Clonazepam • Klonopin
  
• Fentanyl (transdermal) • Duragesic
  
• Hydromorphone • Dilaudid
  
• Meperidine • Demerol
  
• Methadone • Dolophine
  
• Oxazepam • Serax
  
• Oxycodone • OxyContin, Roxicodone
  
• Oxycodone/acetaminophen • Percocet
  
• Propoxyphene • Darvon
  
• Sertraline • Zoloft
  

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

The best intervention for a potentially life-threatening drug rash can happen before you choose a psychotropic. Carefully evaluating your patient’s risk for an adverse cutaneous drug reaction (ACDR) will guide safer prescribing. If your patient develops a rash, differentiating serious from benign reactions can help prevent morbidity, which can range from work loss or hospitalization to disfigurement or death.

In the first installment of this 2-part article on drug eruptions, we discussed how to recognize and manage benign rashes.¹ Here we explain how to reduce ACDR risk and identify 6 serious rashes.

Risk reduction strategies

Although it is impossible to eliminate drug rashes, you may be able to reduce ACDR risk by using sound prescribing methods. Ultimately your choice of a psychotropic comes down to whether the drug’s benefits outweigh the risks to your patient. Factors affecting ACDR risk fall into 3 categories:

• historical
  
• pharmacokinetic
  
• environmental/other.
  

A patient who has had an ACDR also may be hypersensitive to other drugs in the same class. One example is anticonvulsant hypersensitivity syndrome. Phenytoin, carbamazepine, and phenobarbital may be cross-reactive.³ A patient who is hypersensitive to carbamazepine may have a ≥30% risk of reacting to oxcarbazepine.⁴ A major predictor of rash associated with lamotrigine is history of a rash from another antiepileptic.⁵ Cross-reactivity also may occur among antidepressants, particularly selective serotonin reuptake inhibitors.⁶

Knowles et al³ suggests warning close relatives of a patient with anticonvulsant hypersensitivity syndrome about the risk of using potentially cross-reactive anticonvulsants.

If your patient reports that a relative had an ACDR—particularly a severe reaction—to a drug you are considering prescribing, reduce this patient’s risk by choosing an alternate drug or proceeding cautiously by slowly titrating the dosage and monitoring carefully.

Pharmacokinetic factors. In general, ACDRs and dosage are not correlated,² but anticonvulsants may be an exception. For example:

• lowering the starting dosage of lamotrigine reduces ACDR risk⁹
  
• rapid increase in dosages and high serum concentrations of phenytoin and carbamazepine appear to increase the risk of rash.¹⁰
  

Be vigilant for potential interactions between drugs. For instance, valproic acid inhibits lamotrigine metabolism, so when prescribing these 2 medications together, take steps to avoid a serious, life-threatening rash such as Stevens-Johnson syndrome (SJS). For bipolar patients age >12 taking valproic acid, titrate lamotrigine in a special regimen (initially 25 mg every other day, then gradually increased to ≤100 mg/d).¹¹ Remain in close contact with the patient’s other prescribers to ensure that all are aware of potential adverse reactions if the patient’s medications are changed.

Environmental /other factors. Psychotropic medications—particularly antipsychotics—are associated with ACDRs related to sun exposure.^12-14 Advise patients to use sunscreen and wear protective clothing, and consider recommending antioxidant supplements to help prevent photosensitive reactions.¹⁵

Populations at increased risk of developing a drug rash include African-Americans and persons age >70.⁷ Women have higher incidence of rash from lamotrigine use compared with men.⁹ Underlying diseases, such as human immunodeficiency virus, may increase ACDR risk.⁷ Strategies for reducing ACDR risk are summarized in Table 1.

Table 1

Steps to reduce ACDR risk

Serious drug eruptions

Most drug rashes are benign, but some can be life-threatening and require immediate drug discontinuation. Six serious ACDRs associated with psychotropics are listed in Table 2.

Table 2

Serious rashes associated with psychotropics*

• constitutional symptoms (fever, sore throat, malaise, arthralgia, lymphadenopathy, cough)
  
• facial involvement
  
• mucous membrane involvement
  
• skin tenderness or blistering, particularly if there is full-thickness epidermal detachment
  
• purpura.^16,17
  

Table 3

Managing a serious rash

© 2001-2008, DermAtlas
Erythema multiforme: Erythematous target or iris-like papules and vesicobullous eruptions that present on extremities and palmoplantar surfaces. The patient might present with detachment of the epidermis from the dermis. If this consider SJS spectrum disease (see below).^2,13,18,19

Because EM may be a harbinger of a more severe skin reaction, consult a dermatologist and—if the rash involves the eyes—an ophthalmologist.¹² Antihistamines and topical corticosteroids may be used to treat EM.¹⁸ Depending on the severity of the reaction, hospitalization might be indicated.

Stevens-Johnson syndrome/toxic epidermal necrolysis (TEN) are considered a spectrum of reactive skin disorders; TEN is the more severe variant. Patients may present with a prodrome of fever, cough, and malaise. Oral lesions—such as mucosal blistering (Photo 2)—may precede skin lesions. Look for widespread distribution of flat, atypical target lesions characterized by blisters on purpuric macules.² Compared with EM, SJS/TEN lesions are more far-reaching, and the more extensive mucous membrane involvement can affect the mouth, esophagus, and genitalia. Ocular involvement might lead to blindness.^20-23

© 2001-2008, DermAtlas
Stevens-Johnson syndrome/toxic epidermal necrosis: Mucosal blistering, widespread flat skin lesions, and epidermal detachment. Epidermal detachment also may be widespread. SJS and TEN are differentiated by the extent of skin detachment:

• 10% to 30% detachment is SJS/TEN
  
• >30% is TEN.²
  

Hypersensitivity syndrome—known as drug rash with eosinophilia and systemic symptoms (DRESS)—is a potentially life-threatening syndrome that presents as a triad of fever, rash, and internal organ involvement.²⁶ These symptoms typically present 2 to 6 weeks after the patient starts the offending drug.

Early symptoms may include fever, malaise, pharyngitis, and lymphadenopathy.² Cutaneous manifestations range from relatively benign exanthematous eruptions to more serious eruptions such as erythroderma or TEN.

Laboratory findings might show abnormalities of the liver, kidneys, lungs, or thyroid. Atypical lymphocytes and eosinophilia may be present.

Because hypersensitivity syndrome may present like a benign condition, consider the diagnosis when assessing any drug rash, particularly if the patient is receiving an anticonvulsant.^20,22,27 Appropriate, timely care may be best delivered in an inpatient setting, so hospitalization might be indicated. Laboratory tests to assess organ function may include complete blood count (CBC), urine analysis (UA), creatinine, liver function tests, and thyroid stimulating hormone (TSH).

Vasculitis may present with palpable purpura, fever, and rash generally in dependent areas (Photo 3). Patients often develop morbilliform or urticarial eruptions, and the condition might affect internal organs. Differential diagnosis includes:

• Henoch-Schönlein (allergic) purpura
  
• Wegener’s granulomatosis
  
• infections
  
• collagen vascular diseases.²
  

© 2001-2008, DermAtlas
Vasculitis: Palpable purpura, fever, and rash generally in dependent areas.

Erythroderma, also known as exfoliative dermatitis, can present as sudden, pruritic erythema that can generalize (Photo 4). Scaling will appear, followed by desquamation. Patients typically complain of irritation, feeling cold, and a sensation of tightness. Dilated dermal vessels can result in high-output cardiac failure. This potentially life-threatening condition can develop within 1 week of starting a drug.^2,29

© 2001-2008, DermAtlas
Erythroderma: Sudden, pruritic erythema that can generalize. Scaling precedes desquamation.

Erythema nodosum may present as painful erythematous nodules—usually in the lower extremities (Photo 5)—that are the result of fat necrosis.^13,30 Treatment typically involves best rest, nonsteroidal anti-inflammatory drugs, and potassium iodide.³⁰ Systemic corticosteroids also may be used.³¹

© 2001-2008, DermAtlas
Erythema nodosum: Painful erythematous nodules, usually in the lower extremities.

Resuming psychiatric treatment

Although medically necessary for patients with a serious rash, abruptly discontinuing a psychotropic might place them at risk for rapid psychiatric decompensation. Whenever possible, wait 2 weeks before restarting psychopharmacotherapy in a patient who has been treated for an ACDR. If that is not feasible because (for example) the patient is psychotic and agitated, you can cross-taper with a different medication from another class.

If your patient has experienced a serious ACDR, follow the 3 “A’s” to protect against recurrence (Table 4).

Box

3 ‘As’ to protect patients after a life-threatening ACDR

• Knowles SR, Shear NH. Recognition and management of severe cutaneous drug reactions. Dermatol Clin 2007;25(2):245-53.
  
• Dermatology Image Atlas. www.dermatlas.org.
  
• American Academy of Dermatology. www.aad.org.
  

• Alprazolam • Xanax
  
• Amitriptyline • Elavil
  
• Aripiprazole • Abilify
  
• Bupropion • Wellbutrin
  
• Carbamazepine • Tegretol
  
• Chlorpromazine • Thorazine
  
• Clomipramine • Anafranil
  
• Clozapine • Clozaril
  
• Cyclosporine • Neoral, Sandimmune
  
• Desipramine • Norpramin
  
• Diazepam • Valium
  
• Duloxetine • Cymbalta
  
• Eszopiclone • Lunesta
  
• Fluoxetine • Prozac
  
• Fluvoxamine • Luvox
  
• Haloperidol • Haldol
  
• Lamotrigine • Lamictal
  
• Methylphenidate • Ritalin
  
• Mirtazapine • Remeron
  
• Maprotiline • Ludiomil
  
• Olanzapine • Zyprexa
  
• Oxcarbazepine • Trileptal
  
• Paroxetine • Paxil
  
• Phenytoin • Dilantin
  
• Phenobarbital • Luminal
  
• Quetiapine • Seroquel
  
• Risperidone • Risperdal
  
• Sertraline • Zoloft
  
• Thioridazine • Mellaril
  
• Topiramate • Topamax
  
• Lithium • Lithobid, Eskalith
  
• Trazodone • Desyrel
  
• Valproic acid • Depakote
  
• Venlafaxine • Effexor
  
• Ziprasidone • Geodon
  

Dr. Skonicki reports no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Dr. Warnock receives research/grant support from Boehringer Ingelheim, Forest Pharmaceuticals, and Wyeth Pharmaceuticals.

The best intervention for a potentially life-threatening drug rash can happen before you choose a psychotropic. Carefully evaluating your patient’s risk for an adverse cutaneous drug reaction (ACDR) will guide safer prescribing. If your patient develops a rash, differentiating serious from benign reactions can help prevent morbidity, which can range from work loss or hospitalization to disfigurement or death.

In the first installment of this 2-part article on drug eruptions, we discussed how to recognize and manage benign rashes.¹ Here we explain how to reduce ACDR risk and identify 6 serious rashes.

Risk reduction strategies

Although it is impossible to eliminate drug rashes, you may be able to reduce ACDR risk by using sound prescribing methods. Ultimately your choice of a psychotropic comes down to whether the drug’s benefits outweigh the risks to your patient. Factors affecting ACDR risk fall into 3 categories:

• historical
  
• pharmacokinetic
  
• environmental/other.
  

A patient who has had an ACDR also may be hypersensitive to other drugs in the same class. One example is anticonvulsant hypersensitivity syndrome. Phenytoin, carbamazepine, and phenobarbital may be cross-reactive.³ A patient who is hypersensitive to carbamazepine may have a ≥30% risk of reacting to oxcarbazepine.⁴ A major predictor of rash associated with lamotrigine is history of a rash from another antiepileptic.⁵ Cross-reactivity also may occur among antidepressants, particularly selective serotonin reuptake inhibitors.⁶

Knowles et al³ suggests warning close relatives of a patient with anticonvulsant hypersensitivity syndrome about the risk of using potentially cross-reactive anticonvulsants.

If your patient reports that a relative had an ACDR—particularly a severe reaction—to a drug you are considering prescribing, reduce this patient’s risk by choosing an alternate drug or proceeding cautiously by slowly titrating the dosage and monitoring carefully.

Pharmacokinetic factors. In general, ACDRs and dosage are not correlated,² but anticonvulsants may be an exception. For example:

• lowering the starting dosage of lamotrigine reduces ACDR risk⁹
  
• rapid increase in dosages and high serum concentrations of phenytoin and carbamazepine appear to increase the risk of rash.¹⁰
  

Be vigilant for potential interactions between drugs. For instance, valproic acid inhibits lamotrigine metabolism, so when prescribing these 2 medications together, take steps to avoid a serious, life-threatening rash such as Stevens-Johnson syndrome (SJS). For bipolar patients age >12 taking valproic acid, titrate lamotrigine in a special regimen (initially 25 mg every other day, then gradually increased to ≤100 mg/d).¹¹ Remain in close contact with the patient’s other prescribers to ensure that all are aware of potential adverse reactions if the patient’s medications are changed.

Environmental /other factors. Psychotropic medications—particularly antipsychotics—are associated with ACDRs related to sun exposure.^12-14 Advise patients to use sunscreen and wear protective clothing, and consider recommending antioxidant supplements to help prevent photosensitive reactions.¹⁵

Populations at increased risk of developing a drug rash include African-Americans and persons age >70.⁷ Women have higher incidence of rash from lamotrigine use compared with men.⁹ Underlying diseases, such as human immunodeficiency virus, may increase ACDR risk.⁷ Strategies for reducing ACDR risk are summarized in Table 1.

Table 1

Steps to reduce ACDR risk

Serious drug eruptions

Most drug rashes are benign, but some can be life-threatening and require immediate drug discontinuation. Six serious ACDRs associated with psychotropics are listed in Table 2.

Table 2

Serious rashes associated with psychotropics*

• constitutional symptoms (fever, sore throat, malaise, arthralgia, lymphadenopathy, cough)
  
• facial involvement
  
• mucous membrane involvement
  
• skin tenderness or blistering, particularly if there is full-thickness epidermal detachment
  
• purpura.^16,17
  

Table 3

Managing a serious rash

© 2001-2008, DermAtlas
Erythema multiforme: Erythematous target or iris-like papules and vesicobullous eruptions that present on extremities and palmoplantar surfaces. The patient might present with detachment of the epidermis from the dermis. If this consider SJS spectrum disease (see below).^2,13,18,19

Because EM may be a harbinger of a more severe skin reaction, consult a dermatologist and—if the rash involves the eyes—an ophthalmologist.¹² Antihistamines and topical corticosteroids may be used to treat EM.¹⁸ Depending on the severity of the reaction, hospitalization might be indicated.

Stevens-Johnson syndrome/toxic epidermal necrolysis (TEN) are considered a spectrum of reactive skin disorders; TEN is the more severe variant. Patients may present with a prodrome of fever, cough, and malaise. Oral lesions—such as mucosal blistering (Photo 2)—may precede skin lesions. Look for widespread distribution of flat, atypical target lesions characterized by blisters on purpuric macules.² Compared with EM, SJS/TEN lesions are more far-reaching, and the more extensive mucous membrane involvement can affect the mouth, esophagus, and genitalia. Ocular involvement might lead to blindness.^20-23

© 2001-2008, DermAtlas
Stevens-Johnson syndrome/toxic epidermal necrosis: Mucosal blistering, widespread flat skin lesions, and epidermal detachment. Epidermal detachment also may be widespread. SJS and TEN are differentiated by the extent of skin detachment:

• 10% to 30% detachment is SJS/TEN
  
• >30% is TEN.²
  

Hypersensitivity syndrome—known as drug rash with eosinophilia and systemic symptoms (DRESS)—is a potentially life-threatening syndrome that presents as a triad of fever, rash, and internal organ involvement.²⁶ These symptoms typically present 2 to 6 weeks after the patient starts the offending drug.

Early symptoms may include fever, malaise, pharyngitis, and lymphadenopathy.² Cutaneous manifestations range from relatively benign exanthematous eruptions to more serious eruptions such as erythroderma or TEN.

Laboratory findings might show abnormalities of the liver, kidneys, lungs, or thyroid. Atypical lymphocytes and eosinophilia may be present.

Because hypersensitivity syndrome may present like a benign condition, consider the diagnosis when assessing any drug rash, particularly if the patient is receiving an anticonvulsant.^20,22,27 Appropriate, timely care may be best delivered in an inpatient setting, so hospitalization might be indicated. Laboratory tests to assess organ function may include complete blood count (CBC), urine analysis (UA), creatinine, liver function tests, and thyroid stimulating hormone (TSH).

Vasculitis may present with palpable purpura, fever, and rash generally in dependent areas (Photo 3). Patients often develop morbilliform or urticarial eruptions, and the condition might affect internal organs. Differential diagnosis includes:

• Henoch-Schönlein (allergic) purpura
  
• Wegener’s granulomatosis
  
• infections
  
• collagen vascular diseases.²
  

© 2001-2008, DermAtlas
Vasculitis: Palpable purpura, fever, and rash generally in dependent areas.

Erythroderma, also known as exfoliative dermatitis, can present as sudden, pruritic erythema that can generalize (Photo 4). Scaling will appear, followed by desquamation. Patients typically complain of irritation, feeling cold, and a sensation of tightness. Dilated dermal vessels can result in high-output cardiac failure. This potentially life-threatening condition can develop within 1 week of starting a drug.^2,29

© 2001-2008, DermAtlas
Erythroderma: Sudden, pruritic erythema that can generalize. Scaling precedes desquamation.

Erythema nodosum may present as painful erythematous nodules—usually in the lower extremities (Photo 5)—that are the result of fat necrosis.^13,30 Treatment typically involves best rest, nonsteroidal anti-inflammatory drugs, and potassium iodide.³⁰ Systemic corticosteroids also may be used.³¹

© 2001-2008, DermAtlas
Erythema nodosum: Painful erythematous nodules, usually in the lower extremities.

Resuming psychiatric treatment

Although medically necessary for patients with a serious rash, abruptly discontinuing a psychotropic might place them at risk for rapid psychiatric decompensation. Whenever possible, wait 2 weeks before restarting psychopharmacotherapy in a patient who has been treated for an ACDR. If that is not feasible because (for example) the patient is psychotic and agitated, you can cross-taper with a different medication from another class.

If your patient has experienced a serious ACDR, follow the 3 “A’s” to protect against recurrence (Table 4).

Box

3 ‘As’ to protect patients after a life-threatening ACDR

• Knowles SR, Shear NH. Recognition and management of severe cutaneous drug reactions. Dermatol Clin 2007;25(2):245-53.
  
• Dermatology Image Atlas. www.dermatlas.org.
  
• American Academy of Dermatology. www.aad.org.
  

• Alprazolam • Xanax
  
• Amitriptyline • Elavil
  
• Aripiprazole • Abilify
  
• Bupropion • Wellbutrin
  
• Carbamazepine • Tegretol
  
• Chlorpromazine • Thorazine
  
• Clomipramine • Anafranil
  
• Clozapine • Clozaril
  
• Cyclosporine • Neoral, Sandimmune
  
• Desipramine • Norpramin
  
• Diazepam • Valium
  
• Duloxetine • Cymbalta
  
• Eszopiclone • Lunesta
  
• Fluoxetine • Prozac
  
• Fluvoxamine • Luvox
  
• Haloperidol • Haldol
  
• Lamotrigine • Lamictal
  
• Methylphenidate • Ritalin
  
• Mirtazapine • Remeron
  
• Maprotiline • Ludiomil
  
• Olanzapine • Zyprexa
  
• Oxcarbazepine • Trileptal
  
• Paroxetine • Paxil
  
• Phenytoin • Dilantin
  
• Phenobarbital • Luminal
  
• Quetiapine • Seroquel
  
• Risperidone • Risperdal
  
• Sertraline • Zoloft
  
• Thioridazine • Mellaril
  
• Topiramate • Topamax
  
• Lithium • Lithobid, Eskalith
  
• Trazodone • Desyrel
  
• Valproic acid • Depakote
  
• Venlafaxine • Effexor
  
• Ziprasidone • Geodon
  

Dr. Skonicki reports no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Dr. Warnock receives research/grant support from Boehringer Ingelheim, Forest Pharmaceuticals, and Wyeth Pharmaceuticals.

The best intervention for a potentially life-threatening drug rash can happen before you choose a psychotropic. Carefully evaluating your patient’s risk for an adverse cutaneous drug reaction (ACDR) will guide safer prescribing. If your patient develops a rash, differentiating serious from benign reactions can help prevent morbidity, which can range from work loss or hospitalization to disfigurement or death.

In the first installment of this 2-part article on drug eruptions, we discussed how to recognize and manage benign rashes.¹ Here we explain how to reduce ACDR risk and identify 6 serious rashes.

Risk reduction strategies

Although it is impossible to eliminate drug rashes, you may be able to reduce ACDR risk by using sound prescribing methods. Ultimately your choice of a psychotropic comes down to whether the drug’s benefits outweigh the risks to your patient. Factors affecting ACDR risk fall into 3 categories:

• historical
  
• pharmacokinetic
  
• environmental/other.
  

A patient who has had an ACDR also may be hypersensitive to other drugs in the same class. One example is anticonvulsant hypersensitivity syndrome. Phenytoin, carbamazepine, and phenobarbital may be cross-reactive.³ A patient who is hypersensitive to carbamazepine may have a ≥30% risk of reacting to oxcarbazepine.⁴ A major predictor of rash associated with lamotrigine is history of a rash from another antiepileptic.⁵ Cross-reactivity also may occur among antidepressants, particularly selective serotonin reuptake inhibitors.⁶

Knowles et al³ suggests warning close relatives of a patient with anticonvulsant hypersensitivity syndrome about the risk of using potentially cross-reactive anticonvulsants.

If your patient reports that a relative had an ACDR—particularly a severe reaction—to a drug you are considering prescribing, reduce this patient’s risk by choosing an alternate drug or proceeding cautiously by slowly titrating the dosage and monitoring carefully.

Pharmacokinetic factors. In general, ACDRs and dosage are not correlated,² but anticonvulsants may be an exception. For example:

• lowering the starting dosage of lamotrigine reduces ACDR risk⁹
  
• rapid increase in dosages and high serum concentrations of phenytoin and carbamazepine appear to increase the risk of rash.¹⁰
  

Be vigilant for potential interactions between drugs. For instance, valproic acid inhibits lamotrigine metabolism, so when prescribing these 2 medications together, take steps to avoid a serious, life-threatening rash such as Stevens-Johnson syndrome (SJS). For bipolar patients age >12 taking valproic acid, titrate lamotrigine in a special regimen (initially 25 mg every other day, then gradually increased to ≤100 mg/d).¹¹ Remain in close contact with the patient’s other prescribers to ensure that all are aware of potential adverse reactions if the patient’s medications are changed.

Environmental /other factors. Psychotropic medications—particularly antipsychotics—are associated with ACDRs related to sun exposure.^12-14 Advise patients to use sunscreen and wear protective clothing, and consider recommending antioxidant supplements to help prevent photosensitive reactions.¹⁵

Populations at increased risk of developing a drug rash include African-Americans and persons age >70.⁷ Women have higher incidence of rash from lamotrigine use compared with men.⁹ Underlying diseases, such as human immunodeficiency virus, may increase ACDR risk.⁷ Strategies for reducing ACDR risk are summarized in Table 1.

Table 1

Steps to reduce ACDR risk

Serious drug eruptions

Most drug rashes are benign, but some can be life-threatening and require immediate drug discontinuation. Six serious ACDRs associated with psychotropics are listed in Table 2.

Table 2

Serious rashes associated with psychotropics*

• constitutional symptoms (fever, sore throat, malaise, arthralgia, lymphadenopathy, cough)
  
• facial involvement
  
• mucous membrane involvement
  
• skin tenderness or blistering, particularly if there is full-thickness epidermal detachment
  
• purpura.^16,17
  

Table 3

Managing a serious rash

© 2001-2008, DermAtlas
Erythema multiforme: Erythematous target or iris-like papules and vesicobullous eruptions that present on extremities and palmoplantar surfaces. The patient might present with detachment of the epidermis from the dermis. If this consider SJS spectrum disease (see below).^2,13,18,19

Because EM may be a harbinger of a more severe skin reaction, consult a dermatologist and—if the rash involves the eyes—an ophthalmologist.¹² Antihistamines and topical corticosteroids may be used to treat EM.¹⁸ Depending on the severity of the reaction, hospitalization might be indicated.

Stevens-Johnson syndrome/toxic epidermal necrolysis (TEN) are considered a spectrum of reactive skin disorders; TEN is the more severe variant. Patients may present with a prodrome of fever, cough, and malaise. Oral lesions—such as mucosal blistering (Photo 2)—may precede skin lesions. Look for widespread distribution of flat, atypical target lesions characterized by blisters on purpuric macules.² Compared with EM, SJS/TEN lesions are more far-reaching, and the more extensive mucous membrane involvement can affect the mouth, esophagus, and genitalia. Ocular involvement might lead to blindness.^20-23

© 2001-2008, DermAtlas
Stevens-Johnson syndrome/toxic epidermal necrosis: Mucosal blistering, widespread flat skin lesions, and epidermal detachment. Epidermal detachment also may be widespread. SJS and TEN are differentiated by the extent of skin detachment:

• 10% to 30% detachment is SJS/TEN
  
• >30% is TEN.²
  

Hypersensitivity syndrome—known as drug rash with eosinophilia and systemic symptoms (DRESS)—is a potentially life-threatening syndrome that presents as a triad of fever, rash, and internal organ involvement.²⁶ These symptoms typically present 2 to 6 weeks after the patient starts the offending drug.

Early symptoms may include fever, malaise, pharyngitis, and lymphadenopathy.² Cutaneous manifestations range from relatively benign exanthematous eruptions to more serious eruptions such as erythroderma or TEN.

Laboratory findings might show abnormalities of the liver, kidneys, lungs, or thyroid. Atypical lymphocytes and eosinophilia may be present.

Because hypersensitivity syndrome may present like a benign condition, consider the diagnosis when assessing any drug rash, particularly if the patient is receiving an anticonvulsant.^20,22,27 Appropriate, timely care may be best delivered in an inpatient setting, so hospitalization might be indicated. Laboratory tests to assess organ function may include complete blood count (CBC), urine analysis (UA), creatinine, liver function tests, and thyroid stimulating hormone (TSH).

Vasculitis may present with palpable purpura, fever, and rash generally in dependent areas (Photo 3). Patients often develop morbilliform or urticarial eruptions, and the condition might affect internal organs. Differential diagnosis includes:

• Henoch-Schönlein (allergic) purpura
  
• Wegener’s granulomatosis
  
• infections
  
• collagen vascular diseases.²
  

© 2001-2008, DermAtlas
Vasculitis: Palpable purpura, fever, and rash generally in dependent areas.

Erythroderma, also known as exfoliative dermatitis, can present as sudden, pruritic erythema that can generalize (Photo 4). Scaling will appear, followed by desquamation. Patients typically complain of irritation, feeling cold, and a sensation of tightness. Dilated dermal vessels can result in high-output cardiac failure. This potentially life-threatening condition can develop within 1 week of starting a drug.^2,29

© 2001-2008, DermAtlas
Erythroderma: Sudden, pruritic erythema that can generalize. Scaling precedes desquamation.

Erythema nodosum may present as painful erythematous nodules—usually in the lower extremities (Photo 5)—that are the result of fat necrosis.^13,30 Treatment typically involves best rest, nonsteroidal anti-inflammatory drugs, and potassium iodide.³⁰ Systemic corticosteroids also may be used.³¹

© 2001-2008, DermAtlas
Erythema nodosum: Painful erythematous nodules, usually in the lower extremities.

Resuming psychiatric treatment

Although medically necessary for patients with a serious rash, abruptly discontinuing a psychotropic might place them at risk for rapid psychiatric decompensation. Whenever possible, wait 2 weeks before restarting psychopharmacotherapy in a patient who has been treated for an ACDR. If that is not feasible because (for example) the patient is psychotic and agitated, you can cross-taper with a different medication from another class.

If your patient has experienced a serious ACDR, follow the 3 “A’s” to protect against recurrence (Table 4).

Box

3 ‘As’ to protect patients after a life-threatening ACDR

• Knowles SR, Shear NH. Recognition and management of severe cutaneous drug reactions. Dermatol Clin 2007;25(2):245-53.
  
• Dermatology Image Atlas. www.dermatlas.org.
  
• American Academy of Dermatology. www.aad.org.
  

• Alprazolam • Xanax
  
• Amitriptyline • Elavil
  
• Aripiprazole • Abilify
  
• Bupropion • Wellbutrin
  
• Carbamazepine • Tegretol
  
• Chlorpromazine • Thorazine
  
• Clomipramine • Anafranil
  
• Clozapine • Clozaril
  
• Cyclosporine • Neoral, Sandimmune
  
• Desipramine • Norpramin
  
• Diazepam • Valium
  
• Duloxetine • Cymbalta
  
• Eszopiclone • Lunesta
  
• Fluoxetine • Prozac
  
• Fluvoxamine • Luvox
  
• Haloperidol • Haldol
  
• Lamotrigine • Lamictal
  
• Methylphenidate • Ritalin
  
• Mirtazapine • Remeron
  
• Maprotiline • Ludiomil
  
• Olanzapine • Zyprexa
  
• Oxcarbazepine • Trileptal
  
• Paroxetine • Paxil
  
• Phenytoin • Dilantin
  
• Phenobarbital • Luminal
  
• Quetiapine • Seroquel
  
• Risperidone • Risperdal
  
• Sertraline • Zoloft
  
• Thioridazine • Mellaril
  
• Topiramate • Topamax
  
• Lithium • Lithobid, Eskalith
  
• Trazodone • Desyrel
  
• Valproic acid • Depakote
  
• Venlafaxine • Effexor
  
• Ziprasidone • Geodon
  

Dr. Skonicki reports no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Dr. Warnock receives research/grant support from Boehringer Ingelheim, Forest Pharmaceuticals, and Wyeth Pharmaceuticals.

Mr. J, age 52, has a history of opioid dependence. Four weeks after starting interferon therapy for hepatitis C, he presents to the outpatient mental health clinic with depressed mood, irritability, decreased energy, poor concentration, insomnia, anhedonia, and suicidal ideation.

Because Mr. J has no history of depression, the psychiatrist diagnoses him with depressive disorder secondary to interferon. Interferon is stopped. Mr. J’s mood improves, but he wants to restart interferon.

The psychiatrist starts Mr. J on sertraline, 50 mg/d, then gradually increases the dose to 150 mg/d as Mr. J’s mood symptoms return. Subsequently, the patient continues interferon with a combination of sertraline and supportive psychotherapy.

Recognizing a medication as the possible cause of your patient’s psychiatric symptoms can avoid inaccurate diagnosis and nonindicated psychiatric treatment. Diligently evaluating patients for drug-related psychiatric side effects is critical because complications usually are reversed when the offending drug is discontinued. Unfortunately, a thin line separates available evidence from anecdotal myths about psychiatric complications of nonpsychotropics.

Almost two-thirds (65%) of drugs included in the Physicians’ Desk Reference list potential psychiatric side effects, according to a random sample review.¹ In some patients, such as Mr. J, these effects can exacerbate mood symptoms and result in perceptual, cognitive, or behavioral disturbances.

A wide range of drugs can cause psychosis, agitation, anxiety, depression, delirium, or insomnia (Table). On the other hand, certain psychiatric side effects of nonpsychotropics can be beneficial (Box 1).

Improve your assessments by examining the evidence linking psychiatric side effects to commonly prescribed and over-the-counter (OTC) compounds, including:

• cardiovascular medications
  
• steroids (prescription and illegal)
  
• hormones
  
• interferons
  
• antimicrobials.
  

Table

New-onset psychiatric symptoms? Check patient’s drug list

Cardiovascular medications

Beta blockers have CNS effects—some of which cause psychiatric syndromes—that might depend on an ancillary property such as lipophilicity.² Unlike hydrophilic agents such as atenolol that are excreted unchanged by the kidneys, lipophilic drugs such as metoprolol and propranolol are metabolized by the liver and are believed to enter the brain. Metoprolol has a brain/plasma concentration ratio about 20 times higher than that of atenolol.³

Metoprolol and propranolol can induce delirium and psychosis.^4,5 Psychiatric side effects with metoprolol are frequent,⁴ and propranolol has been associated with:

• sedation (affecting >10% of patients)
  
• nightmares
  
• visual impairment
  
• hallucinations
  
• delirium
  
• depression.⁵
  

The relationship between depressive symptoms and beta blockers has been increasingly questioned, however. One study did not find a higher prevalence of depression in patients receiving beta blockers vs those receiving other medications, although this trial had major methodologic limitations.⁷ One large study found no significant association between beta-blocker use and major depression, regardless of patient age, gender, or race.⁸

These studies stress the importance of carefully assessing the individual patient before assigning neurotoxicity to beta blockers, as these drugs have considerable benefits for cardiovascular disease.⁹

Angiotensin-converting enzyme (ACE) inhibitors also affect the CNS. About 4% to 8% of patients taking an ACE inhibitor experience altered mental status—typically increased arousal and psychomotor activity—although

• anxiety
  
• mania
  
• insomnia
  
• fatigue
  
• paresthesias
  
• hallucinations.⁵
  

Clonidine is a centrally acting alpha-agonist. The alpha-adrenergic system regulates arousal and has an important role in major depression, anxiety states, and other arousal disorders.

More than one-third (35%) of patients taking clonidine experience sedation or lethargy; less commonly, the drug causes anxiety (3%), agitation (3%), depression (1%), and insomnia (1%).⁵ Acute confusion, delirium, hypomania, and psychosis related to clonidine use have long been recognized, occurring in 5

Box 1

Cholesterol-lowering statins might be linked to an increased risk of depression and suicide, but the evidence is inconclusive. Some studies have supported this link,^10,11 whereas others have strongly refuted it^12,13 or had mixed results.¹⁴ A recent review¹⁵ recommends being vigilant for psychiatric side effects in patients taking these drugs.

Steroids: prescription and illegal

Corticosteroids are prescribed for a variety of immune system-related diseases, including asthma, allergic rhinitis, rheumatoid arthritis, inflammatory bowel disease, and dermatologic disorders. Mood changes are the most common psychiatric symptoms caused by corticosteroid use; delirium is less common. Psychiatric side effects include:

• lethargy
  
• insomnia
  
• euphoria
  
• depression
  
• psychosis
  
• “personality changes”
  
• anxiety
  
• agitation.⁵
  

In a study of 50 ophthalmologic patients who did not have psychiatric illness receiving prednisolone (mean starting dose 119 mg/d) for 8 days, 26% developed mania and 10% depression.¹⁸ None reported psychotic symptoms.

The most common adverse effects of short-term corticosteroid therapy are euphoria and hypomania. Long-term therapy tends to induce depressive symptoms.¹⁹ A review of 79 cases of psychiatric syndromes induced by corticosteroids found that 41% reported depression, 28% mania, 6% mixed symptoms, and 14% psychosis.²⁰

A group of 16 healthy volunteers receiving 80 mg/d of prednisone over 5 days exhibited depressed or elevated mood, irritability, lability, increased energy, anxiety, and depersonalization.²¹ Numerous case studies have reported anxiety, agitation, mania, and psychotic symptoms in children and adults taking inhaled corticosteroids.

In general, psychiatric side effects of corticosteroids occur within 2 weeks of starting therapy and resolve with dosage reduction or discontinuation. In severe cases or situations in which the dosage cannot be reduced, the patient may require antipsychotics or mood stabilizers.¹⁹

Female gender and past psychiatric history might be risk factors for developing psychiatric symptoms with corticosteroids,²² although not all studies have confirmed these findings.¹⁸

Anabolic androgenic steroids (AAS) have limited therapeutic benefits but are used illegally by some bodybuilders, wrestlers, and other amateur and professional athletes to increase muscle mass, enhance performance, and gain a competitive edge. AAS can cause acute paranoia, delirium, mania or hypomania, homicidal rage, aggression, and extreme mood swings, as well as a marked increase in libido, irritability, agitation, and anger.

In a large observational cohort study of 320 bodybuilding amateur and recreational athletes,²³ AAS use induced many of these psychiatric side effects. The extent intensified as the abuse escalated. A study that used the Structured Clinical Interview for DSM-III-R to compare 88 athletes using steroids with 68 nonusers found that 23% of the AAS users reported major mood syndromes, including mania, hypomania, and major depression.²⁴

In a 2-week, double-blind, fixed-order, placebo-controlled, crossover study of healthy male inpatient volunteers, AAS had both:

• mood-elevating effects—euphoria (“steroid rush”), increased energy, and increased sexual arousal and drive
  
• mood-dysphoric effects, such as irritability, mood swings, increasingly violent feelings, increased hostility, and cognitive impairments.²⁵
  

Hormones

Gonadotropin-releasing hormone (GnRH) agonists such as leuprolide and nafarelin are approved for treating endometriosis, advanced prostate cancer, precocious puberty, and uterine leiomyomata. Some studies and case reports suggest that these agents cause depressive symptoms.²⁶

Interferon

Various forms of interferon are used to treat hepatitis C, melanoma, multiple sclerosis, chronic myelogenous leukemia, and other illnesses. Psychiatric complications—particularly depression—are the most frequent side effect of interferon therapy and mainly occur within the first 12 weeks of therapy.²⁸

In a prospective observational study of veterans undergoing interferon-alfa/ribavirin treatment for chronic hepatitis C:

• 48% of patients not receiving psychiatric care at baseline required treatment for neuropsychiatric side effects
  
• 23% developed symptoms of major depression.²⁹
  

Because patients who receive interferon are far more likely to require psychiatric intervention if they have a family history of mood disorders, closely monitor them for depressive symptoms and treat such symptoms aggressively. Also closely monitor patients with multiple psychiatric diagnoses receiving interferon-alfa therapy.³⁰

Jeungling et al³¹ speculated that hypometabolism in the prefrontal cortex may predispose patients to interferon-associated neuropsychological syndromes. Neuropsychiatric symptoms may be a characteristic of hepatitis C, interferon treatment, or both.³²

Antimicrobial agents

Antibiotic and antiviral drugs can cause psychiatric side effects:

• directly by affecting neuronal functions
  
• indirectly by entering the brain rapidly, taking advantage of the compromised blood-brain barrier during sepsis or infection.
  

Antibiotics. Penicillin and its analogues are associated with sedation, anxiety, and hallucinations. Delirium has been reported as a side effect of most cephalosporins, especially in patients with compromised renal function. Quinolones such as ciprofloxacin and ofloxacin rarely cause restlessness, irritability, lethargy, tremors, insomnia, mania, depression, psychosis, delirium, seizures, or catatonia (incidence ≤1%).⁵ Though not commonly used, chloramphenicol may cause depression, confusion, and delirium. Many case reports have strongly associated clarithromycin with delirium.³³

Isoniazid is one of the most commonly used antibiotics that can cause psychiatric side effects; it has been linked to delirium, mania, depression, and psychosis. Ethionamide is associated with sedation, irritability, depression, restlessness, and psychosis. Tetracyclines have been known to cause depression, insomnia, and irritability at high dosages.

Sulfonamides can cause delirium. Psychosis and confusion also have been reported, especially when sulfa drugs are combined with trimethoprim.⁵

Antivirals. When used intravenously and at high doses, acyclovir and ganciclovir can cause lethargy, anxiety, hallucinations, and frank delirium.⁵ Foscarnet—an antiviral used to treat herpes viruses—can cause depression, anxiety, hallucinations, and aggressive irritability.

Didanosine—an antiretroviral agent to treat HIV infections—can cause lethargy (5% to 7% of patients), depression (2%), anxiety (2%), emotional lability (25%), delirium (2%), insomnia (1%), and psychotic delusions (1%).⁵ Efavirenz treatment may be associated with major depression and severe suicidal ideation.³⁴ Tenofovir, a nucleotide reverse transcriptase inhibitor, has not been associated with psychiatric side effects.²⁷

Antifungals. Psychiatric side effects are rare.

OTC and other agents

Many common nonprescription agents can cause psychiatric symptoms. The most frequently used classes include cold and allergy preparations, reflux medications, and analgesics (Box 2).^5,35

Cold preparations. Combined antihistamines and decongestants—such as phenylpropanolamine, azatadine, loratadine, ephedrine, phenylephrine, pseudoephedrine, and naphazoline—can cause an atropine-like psychosis that typically manifests as confusion, disorientation, agitation, hallucinations, and memory problems. Decongestants can cause dangerously high levels of norepinephrine when combined with monoamine oxidase inhibitors (MAOIs) and are contraindicated in patients taking MAOIs. Ephedrine can induce restlessness, dysphoria, irritability, anxiety, and insomnia.

Time to onset of psychiatric side effects from H₂ antagonists varies. Ranitidine can cause depression 4 to 8 weeks after treatment begins. Cimetidine has been reported to cause adverse events within 2 to 3 weeks and delirium within 24 to 48 hours.³⁸ These effects usually resolve within 3 days of discontinuing the drug. Cimetidine is also associated with sexual dysfunction.

Other medications. Ondansetron is a 5-hydroxytryptamine subclass 3 (5-HT₃) antagonist used for antiemetic therapy. In case reports, it has been strongly associated with anxiety.⁴⁰ This association is complex, however, and studies are evaluating 5-HT₃ receptor antagonists for the treatment of anxiety, depression, phobia, and schizophrenia.

Isotretinoin—a retinoid used for severe acne—can cause severe depression and suicidal behavior.⁴¹

Aminophylline and salbutamol are associated with agitation, insomnia, euphoria, and delirium. Methotrexate is known to cause personality changes, irritability, and delirium.²⁷

Box 2

Treating drug-related mood effects

If you suspect a nonpsychotropic medication is causing your patient’s psychiatric symptoms, discuss this with the patient and the prescribing physician. Switching to another similar agent may be an option. If this is not possible:

• work closely with the patient’s primary physician
  
• treat mood symptoms with appropriate psychotropics.
  

• Turjanski N, Lloyd GG. Psychiatric side-effects of medications: recent developments. Advances in Psychiatric Treatment 2005;11:58-70.
  
• Brown TM, Stoudemire A. Psychiatric side effects of prescription and over-the-counter medications. Recognition and management. Washington, DC: American Psychiatric Publishing; 1998.
  
• Physicians’ Desk Reference. www.pdr.net.
  

• Acyclovir • Zovirax
  
• Aminophylline • Phyllocontin, Truphylline
  
• Atenolol • Tenormin
  
• Azatadine • Optimine
  
• Baclofen • Lioresal
  
• Chloramphenicol • Chloromycetin
  
• Cimetidine • Tagamet
  
• Ciprofloxacin • Cipro
  
• Clarithromycin • Biaxin
  
• Clonidine • Catapres
  
• Cyclosporine • Neoral, Sandimmune, others
  
• Dantrolene • Dantrium
  
• Didanosine • Videx
  
• Efavirenz • Sustiva
  
• Ethionamide • Trecator
  
• Famotidine • Pepcid
  
• Foscarnet • Foscavir
  
• Ganciclovir • Cytovene
  
• Indomethacin • Indocin
  
• Interferon alfa • Intron, Roferon
  
• Isoniazid • Nydrazid
  
• Isotretinoin • Accutane
  
• Lansoprazole • Prevacid
  
• Leuprolide • Lupron
  
• Lidocaine • Xylocaine, Xylocard
  
• Loratadine • Claritin
  
• Methotrexate • Rheumatrex, Trexall
  
• Methyldopa • Aldomet
  
• Metoprolol • Lopressor
  
• Mirtazapine • Remeron
  
• Nafarelin • Synarel
  
• Naloxone • Suboxone
  
• Naltrexone • Vivitrol
  
• Naphazoline • Naphcon-A, Clearine
  
• Naproxen • Aleve, others
  
• Nizatidine • Axid
  
• Ofloxacin • Floxin
  
• Omeprazole • Prilosec
  
• Ondansetron • Zofran
  
• Paroxetine • Paxil
  
• Peginterferon alfa • PEG-Intron, Pegasys
  
• Phenylephrine • Neo-Synephrine
  
• Prednisolone • Blephamide, Pred Forte, others
  
• Propranolol • Inderal
  
• Pseudoephedrine • Actifed, Sudafed
  
• Ranitidine • Zantac
  
• Reserpine • Serpasi
  
• Ribavirin • Copegus, Rebetol
  
• Salbutamol • Aerolin, Airomir, others
  
• Sertraline • Zoloft
  
• Sumatriptan • Imitrex
  
• Tenofovir • Viread
  
• Trimethoprim • Proloprim
  

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

Mr. J, age 52, has a history of opioid dependence. Four weeks after starting interferon therapy for hepatitis C, he presents to the outpatient mental health clinic with depressed mood, irritability, decreased energy, poor concentration, insomnia, anhedonia, and suicidal ideation.

Because Mr. J has no history of depression, the psychiatrist diagnoses him with depressive disorder secondary to interferon. Interferon is stopped. Mr. J’s mood improves, but he wants to restart interferon.

The psychiatrist starts Mr. J on sertraline, 50 mg/d, then gradually increases the dose to 150 mg/d as Mr. J’s mood symptoms return. Subsequently, the patient continues interferon with a combination of sertraline and supportive psychotherapy.

Recognizing a medication as the possible cause of your patient’s psychiatric symptoms can avoid inaccurate diagnosis and nonindicated psychiatric treatment. Diligently evaluating patients for drug-related psychiatric side effects is critical because complications usually are reversed when the offending drug is discontinued. Unfortunately, a thin line separates available evidence from anecdotal myths about psychiatric complications of nonpsychotropics.

Almost two-thirds (65%) of drugs included in the Physicians’ Desk Reference list potential psychiatric side effects, according to a random sample review.¹ In some patients, such as Mr. J, these effects can exacerbate mood symptoms and result in perceptual, cognitive, or behavioral disturbances.

A wide range of drugs can cause psychosis, agitation, anxiety, depression, delirium, or insomnia (Table). On the other hand, certain psychiatric side effects of nonpsychotropics can be beneficial (Box 1).

Improve your assessments by examining the evidence linking psychiatric side effects to commonly prescribed and over-the-counter (OTC) compounds, including:

• cardiovascular medications
  
• steroids (prescription and illegal)
  
• hormones
  
• interferons
  
• antimicrobials.
  

Table

New-onset psychiatric symptoms? Check patient’s drug list

Cardiovascular medications

Beta blockers have CNS effects—some of which cause psychiatric syndromes—that might depend on an ancillary property such as lipophilicity.² Unlike hydrophilic agents such as atenolol that are excreted unchanged by the kidneys, lipophilic drugs such as metoprolol and propranolol are metabolized by the liver and are believed to enter the brain. Metoprolol has a brain/plasma concentration ratio about 20 times higher than that of atenolol.³

Metoprolol and propranolol can induce delirium and psychosis.^4,5 Psychiatric side effects with metoprolol are frequent,⁴ and propranolol has been associated with:

• sedation (affecting >10% of patients)
  
• nightmares
  
• visual impairment
  
• hallucinations
  
• delirium
  
• depression.⁵
  

The relationship between depressive symptoms and beta blockers has been increasingly questioned, however. One study did not find a higher prevalence of depression in patients receiving beta blockers vs those receiving other medications, although this trial had major methodologic limitations.⁷ One large study found no significant association between beta-blocker use and major depression, regardless of patient age, gender, or race.⁸

These studies stress the importance of carefully assessing the individual patient before assigning neurotoxicity to beta blockers, as these drugs have considerable benefits for cardiovascular disease.⁹

Angiotensin-converting enzyme (ACE) inhibitors also affect the CNS. About 4% to 8% of patients taking an ACE inhibitor experience altered mental status—typically increased arousal and psychomotor activity—although

• anxiety
  
• mania
  
• insomnia
  
• fatigue
  
• paresthesias
  
• hallucinations.⁵
  

Clonidine is a centrally acting alpha-agonist. The alpha-adrenergic system regulates arousal and has an important role in major depression, anxiety states, and other arousal disorders.

More than one-third (35%) of patients taking clonidine experience sedation or lethargy; less commonly, the drug causes anxiety (3%), agitation (3%), depression (1%), and insomnia (1%).⁵ Acute confusion, delirium, hypomania, and psychosis related to clonidine use have long been recognized, occurring in 5

Box 1

Cholesterol-lowering statins might be linked to an increased risk of depression and suicide, but the evidence is inconclusive. Some studies have supported this link,^10,11 whereas others have strongly refuted it^12,13 or had mixed results.¹⁴ A recent review¹⁵ recommends being vigilant for psychiatric side effects in patients taking these drugs.

Steroids: prescription and illegal

Corticosteroids are prescribed for a variety of immune system-related diseases, including asthma, allergic rhinitis, rheumatoid arthritis, inflammatory bowel disease, and dermatologic disorders. Mood changes are the most common psychiatric symptoms caused by corticosteroid use; delirium is less common. Psychiatric side effects include:

• lethargy
  
• insomnia
  
• euphoria
  
• depression
  
• psychosis
  
• “personality changes”
  
• anxiety
  
• agitation.⁵
  

In a study of 50 ophthalmologic patients who did not have psychiatric illness receiving prednisolone (mean starting dose 119 mg/d) for 8 days, 26% developed mania and 10% depression.¹⁸ None reported psychotic symptoms.

The most common adverse effects of short-term corticosteroid therapy are euphoria and hypomania. Long-term therapy tends to induce depressive symptoms.¹⁹ A review of 79 cases of psychiatric syndromes induced by corticosteroids found that 41% reported depression, 28% mania, 6% mixed symptoms, and 14% psychosis.²⁰

A group of 16 healthy volunteers receiving 80 mg/d of prednisone over 5 days exhibited depressed or elevated mood, irritability, lability, increased energy, anxiety, and depersonalization.²¹ Numerous case studies have reported anxiety, agitation, mania, and psychotic symptoms in children and adults taking inhaled corticosteroids.

In general, psychiatric side effects of corticosteroids occur within 2 weeks of starting therapy and resolve with dosage reduction or discontinuation. In severe cases or situations in which the dosage cannot be reduced, the patient may require antipsychotics or mood stabilizers.¹⁹

Female gender and past psychiatric history might be risk factors for developing psychiatric symptoms with corticosteroids,²² although not all studies have confirmed these findings.¹⁸

Anabolic androgenic steroids (AAS) have limited therapeutic benefits but are used illegally by some bodybuilders, wrestlers, and other amateur and professional athletes to increase muscle mass, enhance performance, and gain a competitive edge. AAS can cause acute paranoia, delirium, mania or hypomania, homicidal rage, aggression, and extreme mood swings, as well as a marked increase in libido, irritability, agitation, and anger.

In a large observational cohort study of 320 bodybuilding amateur and recreational athletes,²³ AAS use induced many of these psychiatric side effects. The extent intensified as the abuse escalated. A study that used the Structured Clinical Interview for DSM-III-R to compare 88 athletes using steroids with 68 nonusers found that 23% of the AAS users reported major mood syndromes, including mania, hypomania, and major depression.²⁴

In a 2-week, double-blind, fixed-order, placebo-controlled, crossover study of healthy male inpatient volunteers, AAS had both:

• mood-elevating effects—euphoria (“steroid rush”), increased energy, and increased sexual arousal and drive
  
• mood-dysphoric effects, such as irritability, mood swings, increasingly violent feelings, increased hostility, and cognitive impairments.²⁵
  

Hormones

Gonadotropin-releasing hormone (GnRH) agonists such as leuprolide and nafarelin are approved for treating endometriosis, advanced prostate cancer, precocious puberty, and uterine leiomyomata. Some studies and case reports suggest that these agents cause depressive symptoms.²⁶

Interferon

Various forms of interferon are used to treat hepatitis C, melanoma, multiple sclerosis, chronic myelogenous leukemia, and other illnesses. Psychiatric complications—particularly depression—are the most frequent side effect of interferon therapy and mainly occur within the first 12 weeks of therapy.²⁸

In a prospective observational study of veterans undergoing interferon-alfa/ribavirin treatment for chronic hepatitis C:

• 48% of patients not receiving psychiatric care at baseline required treatment for neuropsychiatric side effects
  
• 23% developed symptoms of major depression.²⁹
  

Because patients who receive interferon are far more likely to require psychiatric intervention if they have a family history of mood disorders, closely monitor them for depressive symptoms and treat such symptoms aggressively. Also closely monitor patients with multiple psychiatric diagnoses receiving interferon-alfa therapy.³⁰

Jeungling et al³¹ speculated that hypometabolism in the prefrontal cortex may predispose patients to interferon-associated neuropsychological syndromes. Neuropsychiatric symptoms may be a characteristic of hepatitis C, interferon treatment, or both.³²

Antimicrobial agents

Antibiotic and antiviral drugs can cause psychiatric side effects:

• directly by affecting neuronal functions
  
• indirectly by entering the brain rapidly, taking advantage of the compromised blood-brain barrier during sepsis or infection.
  

Antibiotics. Penicillin and its analogues are associated with sedation, anxiety, and hallucinations. Delirium has been reported as a side effect of most cephalosporins, especially in patients with compromised renal function. Quinolones such as ciprofloxacin and ofloxacin rarely cause restlessness, irritability, lethargy, tremors, insomnia, mania, depression, psychosis, delirium, seizures, or catatonia (incidence ≤1%).⁵ Though not commonly used, chloramphenicol may cause depression, confusion, and delirium. Many case reports have strongly associated clarithromycin with delirium.³³

Isoniazid is one of the most commonly used antibiotics that can cause psychiatric side effects; it has been linked to delirium, mania, depression, and psychosis. Ethionamide is associated with sedation, irritability, depression, restlessness, and psychosis. Tetracyclines have been known to cause depression, insomnia, and irritability at high dosages.

Sulfonamides can cause delirium. Psychosis and confusion also have been reported, especially when sulfa drugs are combined with trimethoprim.⁵

Antivirals. When used intravenously and at high doses, acyclovir and ganciclovir can cause lethargy, anxiety, hallucinations, and frank delirium.⁵ Foscarnet—an antiviral used to treat herpes viruses—can cause depression, anxiety, hallucinations, and aggressive irritability.

Didanosine—an antiretroviral agent to treat HIV infections—can cause lethargy (5% to 7% of patients), depression (2%), anxiety (2%), emotional lability (25%), delirium (2%), insomnia (1%), and psychotic delusions (1%).⁵ Efavirenz treatment may be associated with major depression and severe suicidal ideation.³⁴ Tenofovir, a nucleotide reverse transcriptase inhibitor, has not been associated with psychiatric side effects.²⁷

Antifungals. Psychiatric side effects are rare.

OTC and other agents

Many common nonprescription agents can cause psychiatric symptoms. The most frequently used classes include cold and allergy preparations, reflux medications, and analgesics (Box 2).^5,35

Cold preparations. Combined antihistamines and decongestants—such as phenylpropanolamine, azatadine, loratadine, ephedrine, phenylephrine, pseudoephedrine, and naphazoline—can cause an atropine-like psychosis that typically manifests as confusion, disorientation, agitation, hallucinations, and memory problems. Decongestants can cause dangerously high levels of norepinephrine when combined with monoamine oxidase inhibitors (MAOIs) and are contraindicated in patients taking MAOIs. Ephedrine can induce restlessness, dysphoria, irritability, anxiety, and insomnia.

Time to onset of psychiatric side effects from H₂ antagonists varies. Ranitidine can cause depression 4 to 8 weeks after treatment begins. Cimetidine has been reported to cause adverse events within 2 to 3 weeks and delirium within 24 to 48 hours.³⁸ These effects usually resolve within 3 days of discontinuing the drug. Cimetidine is also associated with sexual dysfunction.

Other medications. Ondansetron is a 5-hydroxytryptamine subclass 3 (5-HT₃) antagonist used for antiemetic therapy. In case reports, it has been strongly associated with anxiety.⁴⁰ This association is complex, however, and studies are evaluating 5-HT₃ receptor antagonists for the treatment of anxiety, depression, phobia, and schizophrenia.

Isotretinoin—a retinoid used for severe acne—can cause severe depression and suicidal behavior.⁴¹

Aminophylline and salbutamol are associated with agitation, insomnia, euphoria, and delirium. Methotrexate is known to cause personality changes, irritability, and delirium.²⁷

Box 2

Treating drug-related mood effects

If you suspect a nonpsychotropic medication is causing your patient’s psychiatric symptoms, discuss this with the patient and the prescribing physician. Switching to another similar agent may be an option. If this is not possible:

• work closely with the patient’s primary physician
  
• treat mood symptoms with appropriate psychotropics.
  

• Turjanski N, Lloyd GG. Psychiatric side-effects of medications: recent developments. Advances in Psychiatric Treatment 2005;11:58-70.
  
• Brown TM, Stoudemire A. Psychiatric side effects of prescription and over-the-counter medications. Recognition and management. Washington, DC: American Psychiatric Publishing; 1998.
  
• Physicians’ Desk Reference. www.pdr.net.
  

• Acyclovir • Zovirax
  
• Aminophylline • Phyllocontin, Truphylline
  
• Atenolol • Tenormin
  
• Azatadine • Optimine
  
• Baclofen • Lioresal
  
• Chloramphenicol • Chloromycetin
  
• Cimetidine • Tagamet
  
• Ciprofloxacin • Cipro
  
• Clarithromycin • Biaxin
  
• Clonidine • Catapres
  
• Cyclosporine • Neoral, Sandimmune, others
  
• Dantrolene • Dantrium
  
• Didanosine • Videx
  
• Efavirenz • Sustiva
  
• Ethionamide • Trecator
  
• Famotidine • Pepcid
  
• Foscarnet • Foscavir
  
• Ganciclovir • Cytovene
  
• Indomethacin • Indocin
  
• Interferon alfa • Intron, Roferon
  
• Isoniazid • Nydrazid
  
• Isotretinoin • Accutane
  
• Lansoprazole • Prevacid
  
• Leuprolide • Lupron
  
• Lidocaine • Xylocaine, Xylocard
  
• Loratadine • Claritin
  
• Methotrexate • Rheumatrex, Trexall
  
• Methyldopa • Aldomet
  
• Metoprolol • Lopressor
  
• Mirtazapine • Remeron
  
• Nafarelin • Synarel
  
• Naloxone • Suboxone
  
• Naltrexone • Vivitrol
  
• Naphazoline • Naphcon-A, Clearine
  
• Naproxen • Aleve, others
  
• Nizatidine • Axid
  
• Ofloxacin • Floxin
  
• Omeprazole • Prilosec
  
• Ondansetron • Zofran
  
• Paroxetine • Paxil
  
• Peginterferon alfa • PEG-Intron, Pegasys
  
• Phenylephrine • Neo-Synephrine
  
• Prednisolone • Blephamide, Pred Forte, others
  
• Propranolol • Inderal
  
• Pseudoephedrine • Actifed, Sudafed
  
• Ranitidine • Zantac
  
• Reserpine • Serpasi
  
• Ribavirin • Copegus, Rebetol
  
• Salbutamol • Aerolin, Airomir, others
  
• Sertraline • Zoloft
  
• Sumatriptan • Imitrex
  
• Tenofovir • Viread
  
• Trimethoprim • Proloprim
  

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

Mr. J, age 52, has a history of opioid dependence. Four weeks after starting interferon therapy for hepatitis C, he presents to the outpatient mental health clinic with depressed mood, irritability, decreased energy, poor concentration, insomnia, anhedonia, and suicidal ideation.

Because Mr. J has no history of depression, the psychiatrist diagnoses him with depressive disorder secondary to interferon. Interferon is stopped. Mr. J’s mood improves, but he wants to restart interferon.

The psychiatrist starts Mr. J on sertraline, 50 mg/d, then gradually increases the dose to 150 mg/d as Mr. J’s mood symptoms return. Subsequently, the patient continues interferon with a combination of sertraline and supportive psychotherapy.

Recognizing a medication as the possible cause of your patient’s psychiatric symptoms can avoid inaccurate diagnosis and nonindicated psychiatric treatment. Diligently evaluating patients for drug-related psychiatric side effects is critical because complications usually are reversed when the offending drug is discontinued. Unfortunately, a thin line separates available evidence from anecdotal myths about psychiatric complications of nonpsychotropics.

Almost two-thirds (65%) of drugs included in the Physicians’ Desk Reference list potential psychiatric side effects, according to a random sample review.¹ In some patients, such as Mr. J, these effects can exacerbate mood symptoms and result in perceptual, cognitive, or behavioral disturbances.

A wide range of drugs can cause psychosis, agitation, anxiety, depression, delirium, or insomnia (Table). On the other hand, certain psychiatric side effects of nonpsychotropics can be beneficial (Box 1).

Improve your assessments by examining the evidence linking psychiatric side effects to commonly prescribed and over-the-counter (OTC) compounds, including:

• cardiovascular medications
  
• steroids (prescription and illegal)
  
• hormones
  
• interferons
  
• antimicrobials.
  

Table

New-onset psychiatric symptoms? Check patient’s drug list

Cardiovascular medications

Beta blockers have CNS effects—some of which cause psychiatric syndromes—that might depend on an ancillary property such as lipophilicity.² Unlike hydrophilic agents such as atenolol that are excreted unchanged by the kidneys, lipophilic drugs such as metoprolol and propranolol are metabolized by the liver and are believed to enter the brain. Metoprolol has a brain/plasma concentration ratio about 20 times higher than that of atenolol.³

Metoprolol and propranolol can induce delirium and psychosis.^4,5 Psychiatric side effects with metoprolol are frequent,⁴ and propranolol has been associated with:

• sedation (affecting >10% of patients)
  
• nightmares
  
• visual impairment
  
• hallucinations
  
• delirium
  
• depression.⁵
  

The relationship between depressive symptoms and beta blockers has been increasingly questioned, however. One study did not find a higher prevalence of depression in patients receiving beta blockers vs those receiving other medications, although this trial had major methodologic limitations.⁷ One large study found no significant association between beta-blocker use and major depression, regardless of patient age, gender, or race.⁸

These studies stress the importance of carefully assessing the individual patient before assigning neurotoxicity to beta blockers, as these drugs have considerable benefits for cardiovascular disease.⁹

Angiotensin-converting enzyme (ACE) inhibitors also affect the CNS. About 4% to 8% of patients taking an ACE inhibitor experience altered mental status—typically increased arousal and psychomotor activity—although

• anxiety
  
• mania
  
• insomnia
  
• fatigue
  
• paresthesias
  
• hallucinations.⁵
  

Clonidine is a centrally acting alpha-agonist. The alpha-adrenergic system regulates arousal and has an important role in major depression, anxiety states, and other arousal disorders.

More than one-third (35%) of patients taking clonidine experience sedation or lethargy; less commonly, the drug causes anxiety (3%), agitation (3%), depression (1%), and insomnia (1%).⁵ Acute confusion, delirium, hypomania, and psychosis related to clonidine use have long been recognized, occurring in 5

Box 1

Cholesterol-lowering statins might be linked to an increased risk of depression and suicide, but the evidence is inconclusive. Some studies have supported this link,^10,11 whereas others have strongly refuted it^12,13 or had mixed results.¹⁴ A recent review¹⁵ recommends being vigilant for psychiatric side effects in patients taking these drugs.

Steroids: prescription and illegal

Corticosteroids are prescribed for a variety of immune system-related diseases, including asthma, allergic rhinitis, rheumatoid arthritis, inflammatory bowel disease, and dermatologic disorders. Mood changes are the most common psychiatric symptoms caused by corticosteroid use; delirium is less common. Psychiatric side effects include:

• lethargy
  
• insomnia
  
• euphoria
  
• depression
  
• psychosis
  
• “personality changes”
  
• anxiety
  
• agitation.⁵
  

In a study of 50 ophthalmologic patients who did not have psychiatric illness receiving prednisolone (mean starting dose 119 mg/d) for 8 days, 26% developed mania and 10% depression.¹⁸ None reported psychotic symptoms.

The most common adverse effects of short-term corticosteroid therapy are euphoria and hypomania. Long-term therapy tends to induce depressive symptoms.¹⁹ A review of 79 cases of psychiatric syndromes induced by corticosteroids found that 41% reported depression, 28% mania, 6% mixed symptoms, and 14% psychosis.²⁰

A group of 16 healthy volunteers receiving 80 mg/d of prednisone over 5 days exhibited depressed or elevated mood, irritability, lability, increased energy, anxiety, and depersonalization.²¹ Numerous case studies have reported anxiety, agitation, mania, and psychotic symptoms in children and adults taking inhaled corticosteroids.

In general, psychiatric side effects of corticosteroids occur within 2 weeks of starting therapy and resolve with dosage reduction or discontinuation. In severe cases or situations in which the dosage cannot be reduced, the patient may require antipsychotics or mood stabilizers.¹⁹

Female gender and past psychiatric history might be risk factors for developing psychiatric symptoms with corticosteroids,²² although not all studies have confirmed these findings.¹⁸

Anabolic androgenic steroids (AAS) have limited therapeutic benefits but are used illegally by some bodybuilders, wrestlers, and other amateur and professional athletes to increase muscle mass, enhance performance, and gain a competitive edge. AAS can cause acute paranoia, delirium, mania or hypomania, homicidal rage, aggression, and extreme mood swings, as well as a marked increase in libido, irritability, agitation, and anger.

In a large observational cohort study of 320 bodybuilding amateur and recreational athletes,²³ AAS use induced many of these psychiatric side effects. The extent intensified as the abuse escalated. A study that used the Structured Clinical Interview for DSM-III-R to compare 88 athletes using steroids with 68 nonusers found that 23% of the AAS users reported major mood syndromes, including mania, hypomania, and major depression.²⁴

In a 2-week, double-blind, fixed-order, placebo-controlled, crossover study of healthy male inpatient volunteers, AAS had both:

• mood-elevating effects—euphoria (“steroid rush”), increased energy, and increased sexual arousal and drive
  
• mood-dysphoric effects, such as irritability, mood swings, increasingly violent feelings, increased hostility, and cognitive impairments.²⁵
  

Hormones

Gonadotropin-releasing hormone (GnRH) agonists such as leuprolide and nafarelin are approved for treating endometriosis, advanced prostate cancer, precocious puberty, and uterine leiomyomata. Some studies and case reports suggest that these agents cause depressive symptoms.²⁶

Interferon

Various forms of interferon are used to treat hepatitis C, melanoma, multiple sclerosis, chronic myelogenous leukemia, and other illnesses. Psychiatric complications—particularly depression—are the most frequent side effect of interferon therapy and mainly occur within the first 12 weeks of therapy.²⁸

In a prospective observational study of veterans undergoing interferon-alfa/ribavirin treatment for chronic hepatitis C:

• 48% of patients not receiving psychiatric care at baseline required treatment for neuropsychiatric side effects
  
• 23% developed symptoms of major depression.²⁹
  

Because patients who receive interferon are far more likely to require psychiatric intervention if they have a family history of mood disorders, closely monitor them for depressive symptoms and treat such symptoms aggressively. Also closely monitor patients with multiple psychiatric diagnoses receiving interferon-alfa therapy.³⁰

Jeungling et al³¹ speculated that hypometabolism in the prefrontal cortex may predispose patients to interferon-associated neuropsychological syndromes. Neuropsychiatric symptoms may be a characteristic of hepatitis C, interferon treatment, or both.³²

Antimicrobial agents

Antibiotic and antiviral drugs can cause psychiatric side effects:

• directly by affecting neuronal functions
  
• indirectly by entering the brain rapidly, taking advantage of the compromised blood-brain barrier during sepsis or infection.
  

Antibiotics. Penicillin and its analogues are associated with sedation, anxiety, and hallucinations. Delirium has been reported as a side effect of most cephalosporins, especially in patients with compromised renal function. Quinolones such as ciprofloxacin and ofloxacin rarely cause restlessness, irritability, lethargy, tremors, insomnia, mania, depression, psychosis, delirium, seizures, or catatonia (incidence ≤1%).⁵ Though not commonly used, chloramphenicol may cause depression, confusion, and delirium. Many case reports have strongly associated clarithromycin with delirium.³³

Isoniazid is one of the most commonly used antibiotics that can cause psychiatric side effects; it has been linked to delirium, mania, depression, and psychosis. Ethionamide is associated with sedation, irritability, depression, restlessness, and psychosis. Tetracyclines have been known to cause depression, insomnia, and irritability at high dosages.

Sulfonamides can cause delirium. Psychosis and confusion also have been reported, especially when sulfa drugs are combined with trimethoprim.⁵

Antivirals. When used intravenously and at high doses, acyclovir and ganciclovir can cause lethargy, anxiety, hallucinations, and frank delirium.⁵ Foscarnet—an antiviral used to treat herpes viruses—can cause depression, anxiety, hallucinations, and aggressive irritability.

Didanosine—an antiretroviral agent to treat HIV infections—can cause lethargy (5% to 7% of patients), depression (2%), anxiety (2%), emotional lability (25%), delirium (2%), insomnia (1%), and psychotic delusions (1%).⁵ Efavirenz treatment may be associated with major depression and severe suicidal ideation.³⁴ Tenofovir, a nucleotide reverse transcriptase inhibitor, has not been associated with psychiatric side effects.²⁷

Antifungals. Psychiatric side effects are rare.

OTC and other agents

Many common nonprescription agents can cause psychiatric symptoms. The most frequently used classes include cold and allergy preparations, reflux medications, and analgesics (Box 2).^5,35

Cold preparations. Combined antihistamines and decongestants—such as phenylpropanolamine, azatadine, loratadine, ephedrine, phenylephrine, pseudoephedrine, and naphazoline—can cause an atropine-like psychosis that typically manifests as confusion, disorientation, agitation, hallucinations, and memory problems. Decongestants can cause dangerously high levels of norepinephrine when combined with monoamine oxidase inhibitors (MAOIs) and are contraindicated in patients taking MAOIs. Ephedrine can induce restlessness, dysphoria, irritability, anxiety, and insomnia.

Time to onset of psychiatric side effects from H₂ antagonists varies. Ranitidine can cause depression 4 to 8 weeks after treatment begins. Cimetidine has been reported to cause adverse events within 2 to 3 weeks and delirium within 24 to 48 hours.³⁸ These effects usually resolve within 3 days of discontinuing the drug. Cimetidine is also associated with sexual dysfunction.

Other medications. Ondansetron is a 5-hydroxytryptamine subclass 3 (5-HT₃) antagonist used for antiemetic therapy. In case reports, it has been strongly associated with anxiety.⁴⁰ This association is complex, however, and studies are evaluating 5-HT₃ receptor antagonists for the treatment of anxiety, depression, phobia, and schizophrenia.

Isotretinoin—a retinoid used for severe acne—can cause severe depression and suicidal behavior.⁴¹

Aminophylline and salbutamol are associated with agitation, insomnia, euphoria, and delirium. Methotrexate is known to cause personality changes, irritability, and delirium.²⁷

Box 2

Treating drug-related mood effects

If you suspect a nonpsychotropic medication is causing your patient’s psychiatric symptoms, discuss this with the patient and the prescribing physician. Switching to another similar agent may be an option. If this is not possible:

• work closely with the patient’s primary physician
  
• treat mood symptoms with appropriate psychotropics.
  

• Turjanski N, Lloyd GG. Psychiatric side-effects of medications: recent developments. Advances in Psychiatric Treatment 2005;11:58-70.
  
• Brown TM, Stoudemire A. Psychiatric side effects of prescription and over-the-counter medications. Recognition and management. Washington, DC: American Psychiatric Publishing; 1998.
  
• Physicians’ Desk Reference. www.pdr.net.
  

• Acyclovir • Zovirax
  
• Aminophylline • Phyllocontin, Truphylline
  
• Atenolol • Tenormin
  
• Azatadine • Optimine
  
• Baclofen • Lioresal
  
• Chloramphenicol • Chloromycetin
  
• Cimetidine • Tagamet
  
• Ciprofloxacin • Cipro
  
• Clarithromycin • Biaxin
  
• Clonidine • Catapres
  
• Cyclosporine • Neoral, Sandimmune, others
  
• Dantrolene • Dantrium
  
• Didanosine • Videx
  
• Efavirenz • Sustiva
  
• Ethionamide • Trecator
  
• Famotidine • Pepcid
  
• Foscarnet • Foscavir
  
• Ganciclovir • Cytovene
  
• Indomethacin • Indocin
  
• Interferon alfa • Intron, Roferon
  
• Isoniazid • Nydrazid
  
• Isotretinoin • Accutane
  
• Lansoprazole • Prevacid
  
• Leuprolide • Lupron
  
• Lidocaine • Xylocaine, Xylocard
  
• Loratadine • Claritin
  
• Methotrexate • Rheumatrex, Trexall
  
• Methyldopa • Aldomet
  
• Metoprolol • Lopressor
  
• Mirtazapine • Remeron
  
• Nafarelin • Synarel
  
• Naloxone • Suboxone
  
• Naltrexone • Vivitrol
  
• Naphazoline • Naphcon-A, Clearine
  
• Naproxen • Aleve, others
  
• Nizatidine • Axid
  
• Ofloxacin • Floxin
  
• Omeprazole • Prilosec
  
• Ondansetron • Zofran
  
• Paroxetine • Paxil
  
• Peginterferon alfa • PEG-Intron, Pegasys
  
• Phenylephrine • Neo-Synephrine
  
• Prednisolone • Blephamide, Pred Forte, others
  
• Propranolol • Inderal
  
• Pseudoephedrine • Actifed, Sudafed
  
• Ranitidine • Zantac
  
• Reserpine • Serpasi
  
• Ribavirin • Copegus, Rebetol
  
• Salbutamol • Aerolin, Airomir, others
  
• Sertraline • Zoloft
  
• Sumatriptan • Imitrex
  
• Tenofovir • Viread
  
• Trimethoprim • Proloprim
  

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

Mr. R, age 67, presents with what he describes as uncharacteristic “memory loss” that is affecting his ability to run his accounting business. He and his wife relate that he was doing well until approximately 9 months ago, when he started showing difficulties remembering clients’ names and addresses.

His wife became extremely concerned when he made serious accounting errors in a 1-month period that resulted in the loss of a longtime customer. Mr. R has become easily distracted and absentminded as well, and his wife reports he is misplacing things around the house.

Screening for mild cognitive impairment (MCI) is not recommended for asymptomatic, cognitively healthy older persons, but memory complaints in individuals age >50—especially when corroborated by a reliable informant—warrant further assessment. Your challenge is to determine whether subtle cognitive changes in patients such as Mr. R are part of normal aging, caused by medical or mental illnesses, or a harbinger of Alzheimer’s disease (AD) or another dementia.

Although no treatments can yet prevent dementia, substantial new research is defining the MCI diagnosis for clinicians. This article describes:

• the evolving understanding of MCI and its subtypes
  
• risk factors for converting from MCI to AD
  
• an evidence-based work-up (including functional, cognitive, and neuropsychological testing)
  
• neuroprotective strategies for patients with an MCI diagnosis, including evidence on cholinesterase inhibitors, vitamin E, and anti-inflammatory agents.
  

MCI’s evolving definition

MCI is characterized by subjective and objective cognitive decline greater than expected for an individual’s age and education but less than the functional deficit required for a dementia diagnosis. MCI is proposed to identify persons with early but pathologic cognitive impairment that has a high risk to progress to AD and possibly other dementias.

MCI subtypes. Some experts view MCI as a single entity, whereas others suggest amnestic (aMCI) and nonamnestic (nMCI) subtypes.^1,3 Each subtype is further divided into single and multiple cognitive domains. So, for example, the diagnosis would be:

• aMCI-single cognitive domain for memory impairment only
  
• aMCI-multiple cognitive domains for memory impairment plus nonmemory deficits, such as in language, executive function, or visuospatial function
  
• nMCI-single or multiple cognitive domains for nonmemory deficits without memory impairment.
  

Determining a patient’s MCI subtype is still a research activity and calls for comprehensive neuropsychological testing. MCI patients perform at least 1.5 standard deviations below the average for age- and education-matched healthy individuals on objective measures of memory.¹

Conversion to dementia

In longitudinal population studies patients with MCI have shown an 11% to 33% risk of developing dementia within 2 years, whereas 44% reverted to normal 1 year later. Reasons for reversibility may include variable definitions of MCI among the longitudinal studies and the possibility that patients who recovered or improved may have had reversible causes of dementia.¹

Individuals with aMCI (Table 1)⁸ progress to AD at a rate of 10% to 15% per year, compared with 1% to 2% per year in normal elderly persons. The Mayo AD research center studies reported a conversion rate of up to 80% from aMCI to AD within 6 years.⁹

Research focuses on identifying preclinical AD states and potential targets for intervention using disease-modifying therapies. Some experts consider MCI to be the earliest clinical manifestation of AD, at least in a subgroup of patients.

Identifying markers to predict which patients are likely to convert from MCI to dementia also is a major research objective. In addition to ApoE status (Table 2),^7,9-15 predictors of conversion may include:

• hippocampal atrophy¹³
  
• reduced metabolism in the temporoparietal cortex and posterior cingulum¹⁴
  
• elevated CSF tau and the 42 amino acid form of ß-amyloid (Aß 42).¹⁵
  

Neuropsychiatric symptoms. Individuals with MCI and neuropsychiatric symptoms may be at particular risk for progressing to dementia. Agitation or depression are more prevalent in persons with MCI than in normal elderly but less prevalent than in those with dementia (Table 3).^10,16

• 16% of normal healthy elderly
  
• 43% of MCI patients
  
• 75% of dementia patients.¹⁶
  

Sleep disturbances and anxiety in persons with MCI may predict progression to AD.¹⁰ A baseline high frequency of apathy in aMCI patients has been associated with progression to AD within 1 year.¹¹

Table 1

Amnestic MCI: Proposed diagnostic criteria

Factors shown to predict conversion from MCI to dementia

Neuropsychiatric symptoms: Rising prevalence mirrors cognitive deterioration in elderly patients*

Depression and MCI

Depression and cognitive complaints overlap considerably in older adults. Depressed patients without dementia show persistent cognitive deficits even after depression remits. In some patients, new-onset geriatric depression is considered a prodrome of MCI and AD. Given that AD neuropathologic changes precede clinical symptoms by many years, depression and AD have been proposed as different clinical manifestations of AD pathology.¹⁷

Depressed MCI patients are at higher risk of developing dementia than those without depression, especially if cognitive measures do not improve after depression is treated.¹² Similarly, cognitively intact older persons who develop depression are at increased risk for MCI, particularly if they carry the ApoE e4 genotype.¹⁹

In the only study in which MCI patients’ neuropsychiatric symptoms have been treated, 39 elderly patients with depression and MCI received open-label sertraline, ≤200 mg/d, for 12 weeks. Among the 26 patients who completed the trial, 17 showed moderate improvement in depressive symptoms, attention, and executive function, and 9 showed no response.²⁰

Recommendation. In clinical practice, antidepressant treatment—usually a selective serotonin reuptake inhibitor (SSRI), with or without psychotherapy—is recommended for the MCI patient with comorbid major depression.

CASE CONTINUED: No signs of depression

Mr. R’s medical, neurologic, and substance use history is unremarkable. Family history includes AD in a paternal aunt diagnosed at age 82. Mr. R reports no history of mood, sleep, or appetite changes and no psychotic symptoms. He shows no deficits in activities of daily living (ADL), although his wife recently took over paying household bills after he forgot to make a payment.

Evidence-based workup

Functional assessment. In the differential diagnosis of MCI, give special attention to functional impairment, which points toward dementia. ADL generally are preserved in MCI, and minimal deterioration is seen in instrumental activities of daily living (IADL). A relatively easy way to assess function is to use the Alzheimer’s Disease Functional Assessment and Change Scale (ADFACS), which is based on 16 ADL and IADL items (Table 4).²¹

Cognitive assessment. Because most individuals with MCI score in the normal range on the Folstein Mini-Mental State Examination (MMSE), the modified MMSE (3MS)²² may be more sensitive for detecting MCI. The 3MS retains the MMSE’s brevity (≤10 minutes to administer) but incorporates 4 additional items, has more graded scoring responses, and broadens the score range to 0 to 100. The clock-drawing test also is sensitive for MCI, especially in detecting early visuoconstructional dysfunction.

The Montreal Cognitive Assessment (MoCA) is a 10-minute, 30-point scale designed to help clinicians detect MCI (see Related Resources). The MoCA usually is given with the modified MMSE for a comprehensive cognitive assessment.

Nasreddine et al²³ administered the MoCA and MMSE to 94 patients who met clinical criteria for MCI, 93 patients with mild AD, and 90 healthy cognitively normal elderly persons, using a cutoff score of 26. MoCA showed:

• 90% sensitivity for detecting MCI (compared with 18% for the MMSE)
  
• 87% specificity to exclude normal elderly persons.
  

Neuropsychological testing can be more sensitive than office-based screening tools in defining MCI subtypes. In the Alzheimer’s Disease Cooperative Study (ADCS), the neuropsychological measures that most accurately predicted progression of patients with aMCI to AD within 36 months were the:

• Symbol Digit Modalities Test
  
• New York University Paragraph Recall Test (Delayed)
  
• Delayed 10-Word List Recall
  
• Alzheimer’s Disease Assessment Scale-cognitive subscale (ADAS-Cog) total score.²⁴
  

Table 4

Alzheimer’s Disease Functional Assessment and Change Scale (ADFACS)

Lab studies to rule out reversible causes of MCI

CASE CONTINUED: Subtle cognitive deficits

Mr. R scores 27/30 on the MMSE (losing 3 points on recall) and 25/30 on the MoCA (losing points on visuospatial/executive function, fluency, and delayed recall). Thyroid stimulating hormone, vitamin B12, folate, and rapid plasma reagin tests are unremarkable; brain MRI shows no significant abnormalities.

You refer Mr. R for neuropsychological testing, and most cognitive domains are normal. Exceptions include moderate impairment in immediate and delayed verbal and visual memory and mild executive dysfunction.

Based on your clinical evaluation and neuropsychological testing, you diagnose amnestic MCI. Mr. R shows abnormalities in memory and executive functioning without significant decline in basic and instrumental ADLs, is not taking medications, and has no other medical or psychiatric condition that could explain his cognitive deficits.

You discuss the diagnosis with him and his wife, including evidence on his risk for progression to dementia, neuroprotective strategies, and medications.

After an MCI diagnosis

Neuroprotection. Eliminate medications with anticholinergic effects, including:

• tricyclic antidepressants
  
• conventional antipsychotics
  
• antihistamines
  
• drugs used to treat urinary incontinence, such as oxybutynin
  
• muscle relaxants, such as cyclobenzaprine
  
• certain antiparkinsonian drugs, such as benztropine.
  

Monitoring. The American Academy of Neurology recommends monitoring patients diagnosed with MCI every 6 to 12 months for cognitive and functional decline.

In these visits, include:

• repeat office-based cognitive assessment, especially the modified MMSE, clock-drawing test, and MoCA
  
• careful history-taking from the patient and reliable informant
  
• repeat neuropsychological testing annually or when dementia is suspected
  
• assessment of the caregiver for distress.
  

• moderate exercise (at least 30 minutes per session, 3 times a week)
  
• cognitively stimulating activities that involve language and psychomotor coordination, such as dancing, crossword puzzles, and volunteer work.
  

Medications—yes or no? Cholinesterase inhibitors, rofecoxib, and vitamin E have not been shown to prevent MCI from progressing to AD. Thus, insufficient evidence exists to recommend medications for patients with MCI.

Donepezil has shown possible short-term benefits, however, and patients may choose to try this medication. Some find comfort in seeking this “extra time” to make decisions about advanced directives, attend to estate and will issues, and optimize relationships while they have only mild cognitive deficits and possess decision-making capacity.

Donepezil. The Alzheimer’s Disease Cooperative Study—supported by the National Institute on Aging—was designed to determine whether daily doses of donepezil or vitamin E can delay or prevent progression of aMCI to AD.²⁵ In the double-blind, placebo-controlled, parallel group study, 769 patients with aMCI were randomly assigned to receive donepezil, 10 mg/d; vitamin E, 1,000 IU bid; or placebo for 3 years.

Overall progression to AD was 16% per year, and the 3-year risk of progression was the same in all 3 groups. Donepezil therapy was associated with a reduced rate of progression to AD compared with placebo during the first year of treatment. Donepezil’s benefit was evident among ApoE e4 carriers at 2-year follow-up, but none of the 3 groups showed statistically significant differences after 3 years. Vitamin E showed no effect on AD progression throughout the study.

Rivastigmine. A randomized, placebo-controlled trial in which 1,018 MCI patients received rivastigmine or placebo for 4 years found no statistically significant benefit of rivastigmine on AD progression.²⁶

Galantamine. Two international randomized, double-blind, placebo-controlled trials failed to show a statistically significant benefit of galantamine in slowing progression from aMCI to AD. MRI data from one of these studies suggested that galantamine may have reduced the rate of brain atrophy over a 2-year period.²⁷

Rofecoxib. Epidemiologic studies indicate that anti-inflammatory drugs may reduce the risk of developing AD, but the COX-2 inhibitor rofecoxib did not delay progression to AD among aMCI patients in a large, placebo-controlled trial.²⁸

Educate patients and family members about supportive nonpharmacologic treatments and cholinesterase inhibitors. The Alzheimer’s Association, National Institute on Aging, and local department of aging agencies offer useful resources (see Related Resources).

CASE CONTINUED: Dealing with uncertainty

Mr. R and his wife are unsettled by his MCI diagnosis. They prefer to take a “wait and watch” approach, decline initiation of a cholinesterase inhibitor, and agree to adhere to nonpharmacologic interventions you discussed. You schedule a follow-up visit in 6 months and encourage them to call you with questions.

Related resources

• Rosenberg PB, Johnston D, Lyketsos CG. A clinical approach to mild cognitive impairment. Am J Psychiatry 2006;163: 1884-90.
  
• Montreal Cognitive Assessment (MoCA). 10-minute screening test designed to help clinicians detect mild cognitive impairment. www.mocatest.org.
  
• Alzheimer’s Association. www.alz.org.
  
• National Institute on Aging. www.nia.nih.gov.
  

• Benztropine • Cogentin
  
• Cyclobenzaprine • Flexeril
  
• Donepezil • Aricept
  
• Galantamine • Razadyne
  
• Oxybutynin • Ditropan
  
• Rivastigmine • Exelon
  
• Rofecoxib • Vioxx
  
• Sertraline • Zoloft
  

Dr. Goveas and Dr. Dixon-Holbrook report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Dr. Kerwin is a consultant to Pfizer and a speaker for Pfizer and Novartis.

Dr. Antuono receives research support from Eisai, Pfizer, and Elan and is a speaker for Pfizer and Forest Pharmaceuticals.

Mr. R, age 67, presents with what he describes as uncharacteristic “memory loss” that is affecting his ability to run his accounting business. He and his wife relate that he was doing well until approximately 9 months ago, when he started showing difficulties remembering clients’ names and addresses.

His wife became extremely concerned when he made serious accounting errors in a 1-month period that resulted in the loss of a longtime customer. Mr. R has become easily distracted and absentminded as well, and his wife reports he is misplacing things around the house.

Screening for mild cognitive impairment (MCI) is not recommended for asymptomatic, cognitively healthy older persons, but memory complaints in individuals age >50—especially when corroborated by a reliable informant—warrant further assessment. Your challenge is to determine whether subtle cognitive changes in patients such as Mr. R are part of normal aging, caused by medical or mental illnesses, or a harbinger of Alzheimer’s disease (AD) or another dementia.

Although no treatments can yet prevent dementia, substantial new research is defining the MCI diagnosis for clinicians. This article describes:

• the evolving understanding of MCI and its subtypes
  
• risk factors for converting from MCI to AD
  
• an evidence-based work-up (including functional, cognitive, and neuropsychological testing)
  
• neuroprotective strategies for patients with an MCI diagnosis, including evidence on cholinesterase inhibitors, vitamin E, and anti-inflammatory agents.
  

MCI’s evolving definition

MCI is characterized by subjective and objective cognitive decline greater than expected for an individual’s age and education but less than the functional deficit required for a dementia diagnosis. MCI is proposed to identify persons with early but pathologic cognitive impairment that has a high risk to progress to AD and possibly other dementias.

MCI subtypes. Some experts view MCI as a single entity, whereas others suggest amnestic (aMCI) and nonamnestic (nMCI) subtypes.^1,3 Each subtype is further divided into single and multiple cognitive domains. So, for example, the diagnosis would be:

• aMCI-single cognitive domain for memory impairment only
  
• aMCI-multiple cognitive domains for memory impairment plus nonmemory deficits, such as in language, executive function, or visuospatial function
  
• nMCI-single or multiple cognitive domains for nonmemory deficits without memory impairment.
  

Determining a patient’s MCI subtype is still a research activity and calls for comprehensive neuropsychological testing. MCI patients perform at least 1.5 standard deviations below the average for age- and education-matched healthy individuals on objective measures of memory.¹

Conversion to dementia

In longitudinal population studies patients with MCI have shown an 11% to 33% risk of developing dementia within 2 years, whereas 44% reverted to normal 1 year later. Reasons for reversibility may include variable definitions of MCI among the longitudinal studies and the possibility that patients who recovered or improved may have had reversible causes of dementia.¹

Individuals with aMCI (Table 1)⁸ progress to AD at a rate of 10% to 15% per year, compared with 1% to 2% per year in normal elderly persons. The Mayo AD research center studies reported a conversion rate of up to 80% from aMCI to AD within 6 years.⁹

Research focuses on identifying preclinical AD states and potential targets for intervention using disease-modifying therapies. Some experts consider MCI to be the earliest clinical manifestation of AD, at least in a subgroup of patients.

Identifying markers to predict which patients are likely to convert from MCI to dementia also is a major research objective. In addition to ApoE status (Table 2),^7,9-15 predictors of conversion may include:

• hippocampal atrophy¹³
  
• reduced metabolism in the temporoparietal cortex and posterior cingulum¹⁴
  
• elevated CSF tau and the 42 amino acid form of ß-amyloid (Aß 42).¹⁵
  

Neuropsychiatric symptoms. Individuals with MCI and neuropsychiatric symptoms may be at particular risk for progressing to dementia. Agitation or depression are more prevalent in persons with MCI than in normal elderly but less prevalent than in those with dementia (Table 3).^10,16

• 16% of normal healthy elderly
  
• 43% of MCI patients
  
• 75% of dementia patients.¹⁶
  

Sleep disturbances and anxiety in persons with MCI may predict progression to AD.¹⁰ A baseline high frequency of apathy in aMCI patients has been associated with progression to AD within 1 year.¹¹

Table 1

Amnestic MCI: Proposed diagnostic criteria

Factors shown to predict conversion from MCI to dementia

Neuropsychiatric symptoms: Rising prevalence mirrors cognitive deterioration in elderly patients*

Depression and MCI

Depression and cognitive complaints overlap considerably in older adults. Depressed patients without dementia show persistent cognitive deficits even after depression remits. In some patients, new-onset geriatric depression is considered a prodrome of MCI and AD. Given that AD neuropathologic changes precede clinical symptoms by many years, depression and AD have been proposed as different clinical manifestations of AD pathology.¹⁷

Depressed MCI patients are at higher risk of developing dementia than those without depression, especially if cognitive measures do not improve after depression is treated.¹² Similarly, cognitively intact older persons who develop depression are at increased risk for MCI, particularly if they carry the ApoE e4 genotype.¹⁹

In the only study in which MCI patients’ neuropsychiatric symptoms have been treated, 39 elderly patients with depression and MCI received open-label sertraline, ≤200 mg/d, for 12 weeks. Among the 26 patients who completed the trial, 17 showed moderate improvement in depressive symptoms, attention, and executive function, and 9 showed no response.²⁰

Recommendation. In clinical practice, antidepressant treatment—usually a selective serotonin reuptake inhibitor (SSRI), with or without psychotherapy—is recommended for the MCI patient with comorbid major depression.

CASE CONTINUED: No signs of depression

Mr. R’s medical, neurologic, and substance use history is unremarkable. Family history includes AD in a paternal aunt diagnosed at age 82. Mr. R reports no history of mood, sleep, or appetite changes and no psychotic symptoms. He shows no deficits in activities of daily living (ADL), although his wife recently took over paying household bills after he forgot to make a payment.

Evidence-based workup

Functional assessment. In the differential diagnosis of MCI, give special attention to functional impairment, which points toward dementia. ADL generally are preserved in MCI, and minimal deterioration is seen in instrumental activities of daily living (IADL). A relatively easy way to assess function is to use the Alzheimer’s Disease Functional Assessment and Change Scale (ADFACS), which is based on 16 ADL and IADL items (Table 4).²¹

Cognitive assessment. Because most individuals with MCI score in the normal range on the Folstein Mini-Mental State Examination (MMSE), the modified MMSE (3MS)²² may be more sensitive for detecting MCI. The 3MS retains the MMSE’s brevity (≤10 minutes to administer) but incorporates 4 additional items, has more graded scoring responses, and broadens the score range to 0 to 100. The clock-drawing test also is sensitive for MCI, especially in detecting early visuoconstructional dysfunction.

The Montreal Cognitive Assessment (MoCA) is a 10-minute, 30-point scale designed to help clinicians detect MCI (see Related Resources). The MoCA usually is given with the modified MMSE for a comprehensive cognitive assessment.

Nasreddine et al²³ administered the MoCA and MMSE to 94 patients who met clinical criteria for MCI, 93 patients with mild AD, and 90 healthy cognitively normal elderly persons, using a cutoff score of 26. MoCA showed:

• 90% sensitivity for detecting MCI (compared with 18% for the MMSE)
  
• 87% specificity to exclude normal elderly persons.
  

Neuropsychological testing can be more sensitive than office-based screening tools in defining MCI subtypes. In the Alzheimer’s Disease Cooperative Study (ADCS), the neuropsychological measures that most accurately predicted progression of patients with aMCI to AD within 36 months were the:

• Symbol Digit Modalities Test
  
• New York University Paragraph Recall Test (Delayed)
  
• Delayed 10-Word List Recall
  
• Alzheimer’s Disease Assessment Scale-cognitive subscale (ADAS-Cog) total score.²⁴
  

Table 4

Alzheimer’s Disease Functional Assessment and Change Scale (ADFACS)