Current Psychiatry

History: ‘leaving town’

Mr. S, age 58, escaped repeatedly from his group home over 4 weeks. During one episode, he removed mail from neighbors’ mailboxes and tried to direct midday traffic. He would disappear for a few hours, sometimes overnight, before returning or being brought back by police.

The patient—who has had schizophrenia with catatonic features for 30 years—offered assorted explanations for escaping, most of them based on delusional beliefs, such as “I’m leaving town to get married” or “I’m late for engineering class.”

Since his last escape 3 weeks ago, Mr. S has remained in the group home without incident but has not been reporting for his usual outpatient psychiatric care. One day, he finally presents to us at the group home sponsor’s urging.

On evaluation, Mr. S shows stereotyped speech, staring, posturing, speech-prompt mutism, and odd mannerisms such as saluting. He has not been bathing or sleeping and smiles inappropriately. He speaks only when spoken to and answers with short phrases punctuated with ”By the grace of the good Lord.”

The authors’ observations

DSM-IV-TR requires at least two features to diagnose catatonic schizophrenia:

• peculiar voluntary movements
  
• extreme negativism
  
• excessive motor activity
  
• echolalia or echopraxia
  
• motoric immobility.¹
  

Catatonia is common among the chronic mentally ill,² yet it often goes undiagnosed.³ As a form of psychosis, catatonia might lead to greater functional impairment if not treated.

Treatment: time to try clozapine?

Over 10 years, numerous antipsychotic regimens plus adjunctive valproic acid, 500 mg tid, or lorazepam, up to 2 mg tid, have not lessened Mr. S’ psychosis and impulsivity. We start clozapine, 400 mg/d, and order twice-monthly blood tests to check for clozapine-induced agranulocytosis.

After nearly 6 months, some catatonic features improve gradually based on clinical interview. Serum clozapine is 363 ng/mL.

poll here

The authors’ observations

Second-generation antipsychotics (SGAs) are favored over first-generation antipsychotics to treat schizophrenia with catatonic features (Table),^4,5 but no drug in either class has worked for Mr. S.

ECT can alleviate catatonic schizophrenia,^4,6 but this option often is not available because the clinician fears a negative outcome would prompt legal action, or the guardian or next of kin do not consent to the procedure.³ We considered referring Mr. S to an ECT provider, but he has no legal guardian to provide consent. The group home sponsor also objected to ECT because Mr. S would have been sent out of town for treatment.

Catatonia patients who are immobile, physically compromised, and refuse food and drink typically are considered ECT candidates. Mr. S eats and drinks regularly and is physically able.

Lorazepam can produce rapid response, but it can be addictive.² Also, an adjunctive 2 mg/d dosage showed no effect.

Clozapine monotherapy has shown effectiveness in catatonic schizophrenia⁷ and might be an option after other antipsychotics have failed.

Table 1

Treatments for catatonia: risks and benefits

Complication: agranulocytosis, then nms

Six months after starting clozapine, Mr. S starts having diaphoresis and night sweats, suggesting neutropenia. Blood testing shows a white blood cell count (WBC) of 3.6/μL, down from 4.6/μL 2 weeks before (normal range, 4.6 to 11/μL).

One week later, Mr. S’ WBC is 1.6/μL with a 46% relative neutrophil value (normal range, 50% to 70%) and an absolute neutrophil count of 736 (normal range, 2,500 to 7,000).

We diagnose agranulocytosis and stop clozapine, but Mr. S’ WBC continues to fall over 2 weeks to 0.8/μL with a 16% relative and 128 absolute neutrophil count. After 1 more week, his WBC increases to 2.6/μL and returns to normal 1 week later—4 weeks after stopping clozapine

We then target Mr. S’ catatonia with intramuscular haloperidol, 100 mg/d for 4 weeks, and ziprasidone, 80 mg bid with food. He tolerates this combination but gradually develops tremor and rigidity. Six weeks later, we add levodopa/carbidopa, 25/250 mg bid for his movement problems.

Two weeks later, Mr. S is sweating profusely, disoriented, rigid, and febrile (104.6°F). We diagnose neuroleptic malignant syndrome (NMS), stop both antipsychotics, and admit him for treatment. We start lorazepam, 1 mg tid for catatonia; bromocriptine, 250 mg bid for rigidity; and continue levodopa/carbidopa at the same dosage. We also add dantrolene, 25 mg tid for 5 days for fever and rigidity, and provide a cooling blanket for hyperthermia.

Mr. S’ fever, autonomic changes, and diaphoresis diminish within 3 days. Rigidity and mental status improve gradually over 2 weeks. We discharge him after 10 days.

poll here

The author’s observations

Catatonia is a recognized risk factor for NMS. White and Robins⁸ described 17 patients with a catatonic syndrome that developed into NMS within 5 to 96 hours of starting a neuroleptic. Sachdev developed an NMS rating scale that includes catatonic symptoms.⁹

Northoff,¹⁰ however, associates NMS with D2 receptor blockage in the basal ganglia and relates catatonia to a frontocortical gamma-aminobutyric acid (GABA) dysfunction. Based on this theory, haloperidol—which offers a higher D2 blockade than do SGAs such as ziprasidone—might have contributed to Mr. S’ NMS.

Some evidence suggests that lorazepam—which works on gamma-aminobutyric acid ionotropic type A (GABAA) receptors—helps treat catatonia in NMS and improves rigidity, hyperthermia, and autonomic signs.¹¹

Treatment: which agents will work?

Three weeks after his discharge, we restart ziprasidone, 40 mg bid for Mr. S’ catatonic schizophrenia. He remains free of NMS symptoms but still has mannerisms (posturing, staring, immobility, stereotypic scratching on his face).

Over 1 year, Mr. S is hospitalized repeatedly because of persistent impulsivity and delusions. He has failed numerous antipsychotic regimens lasting 1 month or longer, including olanzapine, up to 30 mg/d; quetiapine, 300 mg tid; and risperidone, 2 mg tid. Adding a first-generation antipsychotic either does not help (as with perphenazine, 12 mg/d) or diminishes his memory (as with chlorpromazine, 250 mg/d). The anticholinergic benztropine, 2 mg bid, also is ineffective.

Combination quetiapine, 300 mg/d, and the antiviral amantadine, 100 mg tid, improve Mr. S’ stereotypy at first, but his delusions intensify within 1 week. His Bush-Francis Catatonia Rating Scale scores range from 9 (indicating moderate catatonia) to 16 (persistent catatonic features).¹²

poll here

The authors’ observations

Catatonic schizophrenia’s pathophysiology and response to medication might differ compared with other schizophrenia forms.¹³ Dopamine D2 hypoactivity, glutamate N-methyl-D-aspartate (NMDA) hyperactivity, or GABAA hypoactivity are believed to cause catatonia.^3,6,7 GABA agonists, anticonvulsants, dopamine agonists, SGAs, and NMDA antagonists target these pathophysiologies, but patients with a catatonia subtype often respond to only one type of medication.

Lorazepam exerts an anticatatonic effect by binding to GABAA receptors and increasing GABA activity. Lorazepam can help some patients with schizophrenia but has not shown benefit when added to an antipsychotic for chronic catatonia.^6,14

SGAs can provide marked improvement in patients with catatonic schizophrenia.⁵

Salokangas et al¹⁵ note that “atypicals” pass more dopamine to the D2 receptor when dopamine is low in the basal ganglia. This suggests that SGAs with low D2 binding—such as clozapine, olanzapine, and quetiapine—are more beneficial than other SGAs for catatonia. Serotonin binding or other mechanisms might add to these drugs’ anticatatonic effect.⁷

Anticonvulsants. Adjunctive anticonvulsant therapy might alleviate catatonia by increasing GABA activity or by causing a modest antiglutaminergic effect, as reported with carbamazepine or valproic acid.¹⁶ Anticholinergics also might help treat neuroleptic-induced catatonia.¹⁷

Amantadine—FDA-approved to treat Parkinson’s disease and extrapyramidal disease—can alleviate catatonia by blocking hyperglutamatergic excitotoxicity in neurons, thus blocking NMDA receptors.¹⁸ As with Mr. S, however, amantadine can worsen psychosis by increasing dopamine release.

Memantine—an NMDA receptor antagonist indicated for moderate to severe Alzheimer’s disease—also blocks hyperglutamatergic excitotoxicity in neurons. The medication has shown effectiveness for treating catatonic schizophrenia in case reports,^19-21 but 3 patients have reported memantine-induced psychosis and seizures.²¹

Some might argue that Mr. S’ delusions are predominant and more compelling than his catatonia, but these did not hamper his ability to live in a group home. His catatonia-related negativism, impulsivity, and inability to cooperate are what led to frequent hospitalization.

Follow-up: treatment change

We stop amantadine, add memantine, 10 mg bid, and titrate quetiapine over 2 weeks to 900 mg/d. Mr. S’ catatonia improves but some delusions persist. We add olanzapine, 7.5 mg bid, and within 2 weeks Mr. S is less delusional and more cooperative.

We discharge Mr. S on the above medications, plus:

• lorazepam, 1 mg each morning and 2 mg nightly, which he has been taking for catatonia for about 1 year
  
• trazodone, 150 mg bid, which we added 6 months ago to help him sleep and reduce psychomotor excitement
  
• ranitidine, 150 mg bid, for gastroesophageal reflux disorder
  
• and levothyroxine, 0.5 mg/d, for comobrid hypothyroidism. His thyroid-stimulating hormone level is normal.
  

We see Mr. S monthly. He is still impulsive at times, occasionally collecting his neighbors’ newspapers and mail despite instructions from group home staff not to do so. Yet his sponsors say Mr. S is “like a new person.” He talks spontaneously, interacts, and is cooperative. He has not been hospitalized for more than 1 year.

The authors’ observations

Mr. S responded favorably to clozapine but cannot tolerate it. With a combination of two other SGAs, a patient might gain the benefits of clozapine without the need for frequent blood draws or the risk of agranulocytosis, other side effects, or interactions between clozapine and other drugs. Adding memantine was necessary to improve the catatonic features that prevented his return to the group home.

Related resources

• World Federation of Societies of Biological Psychiatry. www.wfsbp.com.
  
• Neuroleptic Malignant Syndrome Information Service. www.nmsis.org.
  
• Mann SC, Caroff SN, Keck PE Jr, Lazarus A. Neuroleptic malignant syndrome and related conditions, 2nd ed. Arlington, VA: American Psychiatric Press; 2003:1-44.
  
• Ungvari GS (ed). Catatonia-an anthology of classical contributions. Hong Kong: Scientific Communications International; 2006.
  

• Amantadine • Symmetrel
  
• Benztropine • Cogentin
  
• Bromocriptine • Parlodel
  
• Carbamazepine • Equetro, others
  
• Chlorpromazine • Thorazine
  
• Clozapine • Clozaril
  
• Dantrolene • Dantrium
  
• Haloperidol • Haldol
  
• Levodopa/carbidopa • Sinemet
  
• Levothyroxine • Synthroid
  
• Lorazepam • Ativan
  
• Memantine • Namenda
  
• Olanzapine • Zyprexa
  
• Perphenazine • Trilafon
  
• Quetiapine • Seroquel
  
• Ranitidine • Zantac
  
• Risperidone • Risperdal
  
• Trazodone • Desyrel
  
• Valproic acid • Depakene
  
• Ziprasidone • Geodon
  

Dr. Carroll is a speaker for Abbott Laboratories, AstraZeneca Pharmaceuticals, Bristol Myers-Squibb Co., Forest Pharmaceuticals, Janssen Pharmaceutica, and Pfizer.

Dr. Thomas receives grant support from Pfizer and is a speaker for Abbott Laboratories, AstraZeneca Pharmaceuticals, and Pfizer.

Dr. Tugrul is a consultant to and speaker for Bristol Myers-Squibb Co. and Eli Lilly and Co.

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

Acknowledgment

The authors thank Francisco José Appiani, MD, chairman, psychiatry department, Military Hospital of Campo de Mayo, Buenos Aires, Argentina, and Vijay Jayanti, BS, medical student, The Ohio State University, Columbus, for their help with this article.

History: ‘leaving town’

Mr. S, age 58, escaped repeatedly from his group home over 4 weeks. During one episode, he removed mail from neighbors’ mailboxes and tried to direct midday traffic. He would disappear for a few hours, sometimes overnight, before returning or being brought back by police.

The patient—who has had schizophrenia with catatonic features for 30 years—offered assorted explanations for escaping, most of them based on delusional beliefs, such as “I’m leaving town to get married” or “I’m late for engineering class.”

Since his last escape 3 weeks ago, Mr. S has remained in the group home without incident but has not been reporting for his usual outpatient psychiatric care. One day, he finally presents to us at the group home sponsor’s urging.

On evaluation, Mr. S shows stereotyped speech, staring, posturing, speech-prompt mutism, and odd mannerisms such as saluting. He has not been bathing or sleeping and smiles inappropriately. He speaks only when spoken to and answers with short phrases punctuated with ”By the grace of the good Lord.”

The authors’ observations

DSM-IV-TR requires at least two features to diagnose catatonic schizophrenia:

• peculiar voluntary movements
  
• extreme negativism
  
• excessive motor activity
  
• echolalia or echopraxia
  
• motoric immobility.¹
  

Catatonia is common among the chronic mentally ill,² yet it often goes undiagnosed.³ As a form of psychosis, catatonia might lead to greater functional impairment if not treated.

Treatment: time to try clozapine?

Over 10 years, numerous antipsychotic regimens plus adjunctive valproic acid, 500 mg tid, or lorazepam, up to 2 mg tid, have not lessened Mr. S’ psychosis and impulsivity. We start clozapine, 400 mg/d, and order twice-monthly blood tests to check for clozapine-induced agranulocytosis.

After nearly 6 months, some catatonic features improve gradually based on clinical interview. Serum clozapine is 363 ng/mL.

poll here

The authors’ observations

Second-generation antipsychotics (SGAs) are favored over first-generation antipsychotics to treat schizophrenia with catatonic features (Table),^4,5 but no drug in either class has worked for Mr. S.

ECT can alleviate catatonic schizophrenia,^4,6 but this option often is not available because the clinician fears a negative outcome would prompt legal action, or the guardian or next of kin do not consent to the procedure.³ We considered referring Mr. S to an ECT provider, but he has no legal guardian to provide consent. The group home sponsor also objected to ECT because Mr. S would have been sent out of town for treatment.

Catatonia patients who are immobile, physically compromised, and refuse food and drink typically are considered ECT candidates. Mr. S eats and drinks regularly and is physically able.

Lorazepam can produce rapid response, but it can be addictive.² Also, an adjunctive 2 mg/d dosage showed no effect.

Clozapine monotherapy has shown effectiveness in catatonic schizophrenia⁷ and might be an option after other antipsychotics have failed.

Table 1

Treatments for catatonia: risks and benefits

Complication: agranulocytosis, then nms

Six months after starting clozapine, Mr. S starts having diaphoresis and night sweats, suggesting neutropenia. Blood testing shows a white blood cell count (WBC) of 3.6/μL, down from 4.6/μL 2 weeks before (normal range, 4.6 to 11/μL).

One week later, Mr. S’ WBC is 1.6/μL with a 46% relative neutrophil value (normal range, 50% to 70%) and an absolute neutrophil count of 736 (normal range, 2,500 to 7,000).

We diagnose agranulocytosis and stop clozapine, but Mr. S’ WBC continues to fall over 2 weeks to 0.8/μL with a 16% relative and 128 absolute neutrophil count. After 1 more week, his WBC increases to 2.6/μL and returns to normal 1 week later—4 weeks after stopping clozapine

We then target Mr. S’ catatonia with intramuscular haloperidol, 100 mg/d for 4 weeks, and ziprasidone, 80 mg bid with food. He tolerates this combination but gradually develops tremor and rigidity. Six weeks later, we add levodopa/carbidopa, 25/250 mg bid for his movement problems.

Two weeks later, Mr. S is sweating profusely, disoriented, rigid, and febrile (104.6°F). We diagnose neuroleptic malignant syndrome (NMS), stop both antipsychotics, and admit him for treatment. We start lorazepam, 1 mg tid for catatonia; bromocriptine, 250 mg bid for rigidity; and continue levodopa/carbidopa at the same dosage. We also add dantrolene, 25 mg tid for 5 days for fever and rigidity, and provide a cooling blanket for hyperthermia.

Mr. S’ fever, autonomic changes, and diaphoresis diminish within 3 days. Rigidity and mental status improve gradually over 2 weeks. We discharge him after 10 days.

poll here

The author’s observations

Catatonia is a recognized risk factor for NMS. White and Robins⁸ described 17 patients with a catatonic syndrome that developed into NMS within 5 to 96 hours of starting a neuroleptic. Sachdev developed an NMS rating scale that includes catatonic symptoms.⁹

Northoff,¹⁰ however, associates NMS with D2 receptor blockage in the basal ganglia and relates catatonia to a frontocortical gamma-aminobutyric acid (GABA) dysfunction. Based on this theory, haloperidol—which offers a higher D2 blockade than do SGAs such as ziprasidone—might have contributed to Mr. S’ NMS.

Some evidence suggests that lorazepam—which works on gamma-aminobutyric acid ionotropic type A (GABAA) receptors—helps treat catatonia in NMS and improves rigidity, hyperthermia, and autonomic signs.¹¹

Treatment: which agents will work?

Three weeks after his discharge, we restart ziprasidone, 40 mg bid for Mr. S’ catatonic schizophrenia. He remains free of NMS symptoms but still has mannerisms (posturing, staring, immobility, stereotypic scratching on his face).

Over 1 year, Mr. S is hospitalized repeatedly because of persistent impulsivity and delusions. He has failed numerous antipsychotic regimens lasting 1 month or longer, including olanzapine, up to 30 mg/d; quetiapine, 300 mg tid; and risperidone, 2 mg tid. Adding a first-generation antipsychotic either does not help (as with perphenazine, 12 mg/d) or diminishes his memory (as with chlorpromazine, 250 mg/d). The anticholinergic benztropine, 2 mg bid, also is ineffective.

Combination quetiapine, 300 mg/d, and the antiviral amantadine, 100 mg tid, improve Mr. S’ stereotypy at first, but his delusions intensify within 1 week. His Bush-Francis Catatonia Rating Scale scores range from 9 (indicating moderate catatonia) to 16 (persistent catatonic features).¹²

poll here

The authors’ observations

Catatonic schizophrenia’s pathophysiology and response to medication might differ compared with other schizophrenia forms.¹³ Dopamine D2 hypoactivity, glutamate N-methyl-D-aspartate (NMDA) hyperactivity, or GABAA hypoactivity are believed to cause catatonia.^3,6,7 GABA agonists, anticonvulsants, dopamine agonists, SGAs, and NMDA antagonists target these pathophysiologies, but patients with a catatonia subtype often respond to only one type of medication.

Lorazepam exerts an anticatatonic effect by binding to GABAA receptors and increasing GABA activity. Lorazepam can help some patients with schizophrenia but has not shown benefit when added to an antipsychotic for chronic catatonia.^6,14

SGAs can provide marked improvement in patients with catatonic schizophrenia.⁵

Salokangas et al¹⁵ note that “atypicals” pass more dopamine to the D2 receptor when dopamine is low in the basal ganglia. This suggests that SGAs with low D2 binding—such as clozapine, olanzapine, and quetiapine—are more beneficial than other SGAs for catatonia. Serotonin binding or other mechanisms might add to these drugs’ anticatatonic effect.⁷

Anticonvulsants. Adjunctive anticonvulsant therapy might alleviate catatonia by increasing GABA activity or by causing a modest antiglutaminergic effect, as reported with carbamazepine or valproic acid.¹⁶ Anticholinergics also might help treat neuroleptic-induced catatonia.¹⁷

Amantadine—FDA-approved to treat Parkinson’s disease and extrapyramidal disease—can alleviate catatonia by blocking hyperglutamatergic excitotoxicity in neurons, thus blocking NMDA receptors.¹⁸ As with Mr. S, however, amantadine can worsen psychosis by increasing dopamine release.

Memantine—an NMDA receptor antagonist indicated for moderate to severe Alzheimer’s disease—also blocks hyperglutamatergic excitotoxicity in neurons. The medication has shown effectiveness for treating catatonic schizophrenia in case reports,^19-21 but 3 patients have reported memantine-induced psychosis and seizures.²¹

Some might argue that Mr. S’ delusions are predominant and more compelling than his catatonia, but these did not hamper his ability to live in a group home. His catatonia-related negativism, impulsivity, and inability to cooperate are what led to frequent hospitalization.

Follow-up: treatment change

We stop amantadine, add memantine, 10 mg bid, and titrate quetiapine over 2 weeks to 900 mg/d. Mr. S’ catatonia improves but some delusions persist. We add olanzapine, 7.5 mg bid, and within 2 weeks Mr. S is less delusional and more cooperative.

We discharge Mr. S on the above medications, plus:

• lorazepam, 1 mg each morning and 2 mg nightly, which he has been taking for catatonia for about 1 year
  
• trazodone, 150 mg bid, which we added 6 months ago to help him sleep and reduce psychomotor excitement
  
• ranitidine, 150 mg bid, for gastroesophageal reflux disorder
  
• and levothyroxine, 0.5 mg/d, for comobrid hypothyroidism. His thyroid-stimulating hormone level is normal.
  

We see Mr. S monthly. He is still impulsive at times, occasionally collecting his neighbors’ newspapers and mail despite instructions from group home staff not to do so. Yet his sponsors say Mr. S is “like a new person.” He talks spontaneously, interacts, and is cooperative. He has not been hospitalized for more than 1 year.

The authors’ observations

Mr. S responded favorably to clozapine but cannot tolerate it. With a combination of two other SGAs, a patient might gain the benefits of clozapine without the need for frequent blood draws or the risk of agranulocytosis, other side effects, or interactions between clozapine and other drugs. Adding memantine was necessary to improve the catatonic features that prevented his return to the group home.

Related resources

• World Federation of Societies of Biological Psychiatry. www.wfsbp.com.
  
• Neuroleptic Malignant Syndrome Information Service. www.nmsis.org.
  
• Mann SC, Caroff SN, Keck PE Jr, Lazarus A. Neuroleptic malignant syndrome and related conditions, 2nd ed. Arlington, VA: American Psychiatric Press; 2003:1-44.
  
• Ungvari GS (ed). Catatonia-an anthology of classical contributions. Hong Kong: Scientific Communications International; 2006.
  

• Amantadine • Symmetrel
  
• Benztropine • Cogentin
  
• Bromocriptine • Parlodel
  
• Carbamazepine • Equetro, others
  
• Chlorpromazine • Thorazine
  
• Clozapine • Clozaril
  
• Dantrolene • Dantrium
  
• Haloperidol • Haldol
  
• Levodopa/carbidopa • Sinemet
  
• Levothyroxine • Synthroid
  
• Lorazepam • Ativan
  
• Memantine • Namenda
  
• Olanzapine • Zyprexa
  
• Perphenazine • Trilafon
  
• Quetiapine • Seroquel
  
• Ranitidine • Zantac
  
• Risperidone • Risperdal
  
• Trazodone • Desyrel
  
• Valproic acid • Depakene
  
• Ziprasidone • Geodon
  

Dr. Carroll is a speaker for Abbott Laboratories, AstraZeneca Pharmaceuticals, Bristol Myers-Squibb Co., Forest Pharmaceuticals, Janssen Pharmaceutica, and Pfizer.

Dr. Thomas receives grant support from Pfizer and is a speaker for Abbott Laboratories, AstraZeneca Pharmaceuticals, and Pfizer.

Dr. Tugrul is a consultant to and speaker for Bristol Myers-Squibb Co. and Eli Lilly and Co.

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

Acknowledgment

The authors thank Francisco José Appiani, MD, chairman, psychiatry department, Military Hospital of Campo de Mayo, Buenos Aires, Argentina, and Vijay Jayanti, BS, medical student, The Ohio State University, Columbus, for their help with this article.

History: ‘leaving town’

Mr. S, age 58, escaped repeatedly from his group home over 4 weeks. During one episode, he removed mail from neighbors’ mailboxes and tried to direct midday traffic. He would disappear for a few hours, sometimes overnight, before returning or being brought back by police.

The patient—who has had schizophrenia with catatonic features for 30 years—offered assorted explanations for escaping, most of them based on delusional beliefs, such as “I’m leaving town to get married” or “I’m late for engineering class.”

Since his last escape 3 weeks ago, Mr. S has remained in the group home without incident but has not been reporting for his usual outpatient psychiatric care. One day, he finally presents to us at the group home sponsor’s urging.

On evaluation, Mr. S shows stereotyped speech, staring, posturing, speech-prompt mutism, and odd mannerisms such as saluting. He has not been bathing or sleeping and smiles inappropriately. He speaks only when spoken to and answers with short phrases punctuated with ”By the grace of the good Lord.”

The authors’ observations

DSM-IV-TR requires at least two features to diagnose catatonic schizophrenia:

• peculiar voluntary movements
  
• extreme negativism
  
• excessive motor activity
  
• echolalia or echopraxia
  
• motoric immobility.¹
  

Catatonia is common among the chronic mentally ill,² yet it often goes undiagnosed.³ As a form of psychosis, catatonia might lead to greater functional impairment if not treated.

Treatment: time to try clozapine?

Over 10 years, numerous antipsychotic regimens plus adjunctive valproic acid, 500 mg tid, or lorazepam, up to 2 mg tid, have not lessened Mr. S’ psychosis and impulsivity. We start clozapine, 400 mg/d, and order twice-monthly blood tests to check for clozapine-induced agranulocytosis.

After nearly 6 months, some catatonic features improve gradually based on clinical interview. Serum clozapine is 363 ng/mL.

poll here

The authors’ observations

Second-generation antipsychotics (SGAs) are favored over first-generation antipsychotics to treat schizophrenia with catatonic features (Table),^4,5 but no drug in either class has worked for Mr. S.

ECT can alleviate catatonic schizophrenia,^4,6 but this option often is not available because the clinician fears a negative outcome would prompt legal action, or the guardian or next of kin do not consent to the procedure.³ We considered referring Mr. S to an ECT provider, but he has no legal guardian to provide consent. The group home sponsor also objected to ECT because Mr. S would have been sent out of town for treatment.

Catatonia patients who are immobile, physically compromised, and refuse food and drink typically are considered ECT candidates. Mr. S eats and drinks regularly and is physically able.

Lorazepam can produce rapid response, but it can be addictive.² Also, an adjunctive 2 mg/d dosage showed no effect.

Clozapine monotherapy has shown effectiveness in catatonic schizophrenia⁷ and might be an option after other antipsychotics have failed.

Table 1

Treatments for catatonia: risks and benefits

Complication: agranulocytosis, then nms

Six months after starting clozapine, Mr. S starts having diaphoresis and night sweats, suggesting neutropenia. Blood testing shows a white blood cell count (WBC) of 3.6/μL, down from 4.6/μL 2 weeks before (normal range, 4.6 to 11/μL).

One week later, Mr. S’ WBC is 1.6/μL with a 46% relative neutrophil value (normal range, 50% to 70%) and an absolute neutrophil count of 736 (normal range, 2,500 to 7,000).

We diagnose agranulocytosis and stop clozapine, but Mr. S’ WBC continues to fall over 2 weeks to 0.8/μL with a 16% relative and 128 absolute neutrophil count. After 1 more week, his WBC increases to 2.6/μL and returns to normal 1 week later—4 weeks after stopping clozapine

We then target Mr. S’ catatonia with intramuscular haloperidol, 100 mg/d for 4 weeks, and ziprasidone, 80 mg bid with food. He tolerates this combination but gradually develops tremor and rigidity. Six weeks later, we add levodopa/carbidopa, 25/250 mg bid for his movement problems.

Two weeks later, Mr. S is sweating profusely, disoriented, rigid, and febrile (104.6°F). We diagnose neuroleptic malignant syndrome (NMS), stop both antipsychotics, and admit him for treatment. We start lorazepam, 1 mg tid for catatonia; bromocriptine, 250 mg bid for rigidity; and continue levodopa/carbidopa at the same dosage. We also add dantrolene, 25 mg tid for 5 days for fever and rigidity, and provide a cooling blanket for hyperthermia.

Mr. S’ fever, autonomic changes, and diaphoresis diminish within 3 days. Rigidity and mental status improve gradually over 2 weeks. We discharge him after 10 days.

poll here

The author’s observations

Catatonia is a recognized risk factor for NMS. White and Robins⁸ described 17 patients with a catatonic syndrome that developed into NMS within 5 to 96 hours of starting a neuroleptic. Sachdev developed an NMS rating scale that includes catatonic symptoms.⁹

Northoff,¹⁰ however, associates NMS with D2 receptor blockage in the basal ganglia and relates catatonia to a frontocortical gamma-aminobutyric acid (GABA) dysfunction. Based on this theory, haloperidol—which offers a higher D2 blockade than do SGAs such as ziprasidone—might have contributed to Mr. S’ NMS.

Some evidence suggests that lorazepam—which works on gamma-aminobutyric acid ionotropic type A (GABAA) receptors—helps treat catatonia in NMS and improves rigidity, hyperthermia, and autonomic signs.¹¹

Treatment: which agents will work?

Three weeks after his discharge, we restart ziprasidone, 40 mg bid for Mr. S’ catatonic schizophrenia. He remains free of NMS symptoms but still has mannerisms (posturing, staring, immobility, stereotypic scratching on his face).

Over 1 year, Mr. S is hospitalized repeatedly because of persistent impulsivity and delusions. He has failed numerous antipsychotic regimens lasting 1 month or longer, including olanzapine, up to 30 mg/d; quetiapine, 300 mg tid; and risperidone, 2 mg tid. Adding a first-generation antipsychotic either does not help (as with perphenazine, 12 mg/d) or diminishes his memory (as with chlorpromazine, 250 mg/d). The anticholinergic benztropine, 2 mg bid, also is ineffective.

Combination quetiapine, 300 mg/d, and the antiviral amantadine, 100 mg tid, improve Mr. S’ stereotypy at first, but his delusions intensify within 1 week. His Bush-Francis Catatonia Rating Scale scores range from 9 (indicating moderate catatonia) to 16 (persistent catatonic features).¹²

poll here

The authors’ observations

Catatonic schizophrenia’s pathophysiology and response to medication might differ compared with other schizophrenia forms.¹³ Dopamine D2 hypoactivity, glutamate N-methyl-D-aspartate (NMDA) hyperactivity, or GABAA hypoactivity are believed to cause catatonia.^3,6,7 GABA agonists, anticonvulsants, dopamine agonists, SGAs, and NMDA antagonists target these pathophysiologies, but patients with a catatonia subtype often respond to only one type of medication.

Lorazepam exerts an anticatatonic effect by binding to GABAA receptors and increasing GABA activity. Lorazepam can help some patients with schizophrenia but has not shown benefit when added to an antipsychotic for chronic catatonia.^6,14

SGAs can provide marked improvement in patients with catatonic schizophrenia.⁵

Salokangas et al¹⁵ note that “atypicals” pass more dopamine to the D2 receptor when dopamine is low in the basal ganglia. This suggests that SGAs with low D2 binding—such as clozapine, olanzapine, and quetiapine—are more beneficial than other SGAs for catatonia. Serotonin binding or other mechanisms might add to these drugs’ anticatatonic effect.⁷

Anticonvulsants. Adjunctive anticonvulsant therapy might alleviate catatonia by increasing GABA activity or by causing a modest antiglutaminergic effect, as reported with carbamazepine or valproic acid.¹⁶ Anticholinergics also might help treat neuroleptic-induced catatonia.¹⁷

Amantadine—FDA-approved to treat Parkinson’s disease and extrapyramidal disease—can alleviate catatonia by blocking hyperglutamatergic excitotoxicity in neurons, thus blocking NMDA receptors.¹⁸ As with Mr. S, however, amantadine can worsen psychosis by increasing dopamine release.

Memantine—an NMDA receptor antagonist indicated for moderate to severe Alzheimer’s disease—also blocks hyperglutamatergic excitotoxicity in neurons. The medication has shown effectiveness for treating catatonic schizophrenia in case reports,^19-21 but 3 patients have reported memantine-induced psychosis and seizures.²¹

Some might argue that Mr. S’ delusions are predominant and more compelling than his catatonia, but these did not hamper his ability to live in a group home. His catatonia-related negativism, impulsivity, and inability to cooperate are what led to frequent hospitalization.

Follow-up: treatment change

We stop amantadine, add memantine, 10 mg bid, and titrate quetiapine over 2 weeks to 900 mg/d. Mr. S’ catatonia improves but some delusions persist. We add olanzapine, 7.5 mg bid, and within 2 weeks Mr. S is less delusional and more cooperative.

We discharge Mr. S on the above medications, plus:

• lorazepam, 1 mg each morning and 2 mg nightly, which he has been taking for catatonia for about 1 year
  
• trazodone, 150 mg bid, which we added 6 months ago to help him sleep and reduce psychomotor excitement
  
• ranitidine, 150 mg bid, for gastroesophageal reflux disorder
  
• and levothyroxine, 0.5 mg/d, for comobrid hypothyroidism. His thyroid-stimulating hormone level is normal.
  

We see Mr. S monthly. He is still impulsive at times, occasionally collecting his neighbors’ newspapers and mail despite instructions from group home staff not to do so. Yet his sponsors say Mr. S is “like a new person.” He talks spontaneously, interacts, and is cooperative. He has not been hospitalized for more than 1 year.

The authors’ observations

Mr. S responded favorably to clozapine but cannot tolerate it. With a combination of two other SGAs, a patient might gain the benefits of clozapine without the need for frequent blood draws or the risk of agranulocytosis, other side effects, or interactions between clozapine and other drugs. Adding memantine was necessary to improve the catatonic features that prevented his return to the group home.

Related resources

• World Federation of Societies of Biological Psychiatry. www.wfsbp.com.
  
• Neuroleptic Malignant Syndrome Information Service. www.nmsis.org.
  
• Mann SC, Caroff SN, Keck PE Jr, Lazarus A. Neuroleptic malignant syndrome and related conditions, 2nd ed. Arlington, VA: American Psychiatric Press; 2003:1-44.
  
• Ungvari GS (ed). Catatonia-an anthology of classical contributions. Hong Kong: Scientific Communications International; 2006.
  

• Amantadine • Symmetrel
  
• Benztropine • Cogentin
  
• Bromocriptine • Parlodel
  
• Carbamazepine • Equetro, others
  
• Chlorpromazine • Thorazine
  
• Clozapine • Clozaril
  
• Dantrolene • Dantrium
  
• Haloperidol • Haldol
  
• Levodopa/carbidopa • Sinemet
  
• Levothyroxine • Synthroid
  
• Lorazepam • Ativan
  
• Memantine • Namenda
  
• Olanzapine • Zyprexa
  
• Perphenazine • Trilafon
  
• Quetiapine • Seroquel
  
• Ranitidine • Zantac
  
• Risperidone • Risperdal
  
• Trazodone • Desyrel
  
• Valproic acid • Depakene
  
• Ziprasidone • Geodon
  

Dr. Carroll is a speaker for Abbott Laboratories, AstraZeneca Pharmaceuticals, Bristol Myers-Squibb Co., Forest Pharmaceuticals, Janssen Pharmaceutica, and Pfizer.

Dr. Thomas receives grant support from Pfizer and is a speaker for Abbott Laboratories, AstraZeneca Pharmaceuticals, and Pfizer.

Dr. Tugrul is a consultant to and speaker for Bristol Myers-Squibb Co. and Eli Lilly and Co.

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

Acknowledgment

The authors thank Francisco José Appiani, MD, chairman, psychiatry department, Military Hospital of Campo de Mayo, Buenos Aires, Argentina, and Vijay Jayanti, BS, medical student, The Ohio State University, Columbus, for their help with this article.

Few depressed older adults seek help from psychiatrists. Those who receive mental health treatment most likely do so in primary care settings. Yet primary care physicians (PCPs) often are ill-equipped to effectively treat depression while managing older patients’ numerous acute and chronic medical conditions.

If depressed older patients won’t go to a psychiatrist, why not bring the psychiatrist to the patients? This article describes a clinically tested approach called project IMPACT that links psychiatrists to primary care teams and dramatically improves depression treatment in older adults.

Untreated geriatric depression

Preference for primary care. Major depression is rare in community-living older adults (1% to 3% prevalence), but:

• 5% to 10% of older primary care patients meet DSM-IV-TR criteria for major depression¹
• approximately one-half of depressed older adults report a primary care visit when a mental health problem was addressed during the past year.²

Box 1

Box 2

Only 8% of depressed older adults visit a mental health specialist in a given year, compared with 25% of depressed younger adults.² Even when PCPs refer older patients to a mental health specialist, only 50% follow through (Box 1).^3,4

Barriers to effective care. Depression diagnosis and treatment by PCPs has improved, but a recent survey suggests that with usual treatment:

• only 1 in 5 depressed older adults treated in a primary care practice experiences substantial improvement over 12 months
• only 1 in 10 becomes symptom-free.⁵

Many depressed older adults do not realize they have depression and visit their PCPs complaining of physical symptoms (Box 2).^6-8 Their limited knowledge about depression or fear of being labeled “mentally ill” deters them from disclosing a depressed mood. They and their PCPs may think depression is inevitable with aging.

PCPs also may lack training to differentiate mood disorders, transitory reactions to life-events, or depression caused by medical illness. Their busy schedules limit time to address and prioritize patient concerns about acute and chronic medical problems (Box 3). Thus depression “falls through the cracks.”

Prescribing concerns. PCPs who feel uncomfortable prescribing antidepressants to older patients may be concerned about side effects and maintain dosages at low starting levels instead of titrating up to a therapeutic range.

Collaborative care

One way to overcome these barriers is to integrate mental health providers into primary care to support and augment PCP-prescribed depression treatment. Collaborative care can become an effective, efficient way to provide high-quality depression care to older patients who might otherwise go untreated.⁹

Box 3

Project IMPACT. One such model—project IMPACT (Improving Mood: Promoting Access to Collaborative Treatment for Late-life Depression)—was developed with support from the John A. Hartford Foundation and California Healthcare Foundation. At its heart is a depression care manager or depression clinical specialist—typically a nurse, social worker, or psychologist—who works in a primary care practice. Other team members include the patient’s PCP, the patient, and a consulting psychiatrist.

The care manager works closely with the PCP by:

• educating patients about depression
• coaching patients in pleasant events scheduling/behavioral activation
• supporting the PCP’s antidepressant management
• offering patients a brief course of evidence-based counseling, such as Problem Solving Treatment in Primary Care¹⁰
• measuring patients’ depressive symptoms at treatment onset and regularly thereafter with a tool such as the 9-item depression scale of the Patient Health Questionnaire (PHQ-9).¹¹

Consulting psychiatrist’s role. If a patient is not sufficiently improved after 8 to 10 weeks, the care manager works with the PCP and psychiatrist to change treatment according to an evidence-based algorithm (Figure).¹² In large health care systems, the psychiatrist meets weekly with the care manager to review treatment plans for approximately 100 depressed older adults, with particular attention to those who are not improving. Psychiatrists may see patients in person or facilitate other specialty mental health treatment, as indicated.

Patients are encouraged to choose an antidepressant prescribed by their PCP, psychotherapy provided by the care manager in the primary care setting, or both. The patient and PCP make treatment decisions with support from the care manager and psychiatrist.

After depressive symptoms remit, the care manager completes a relapse prevention plan with the patient. This includes:

• steps the patient can take to prevent a relapse
• identifying relapse warning signs
• an action plan if depressive symptoms recur.

Training for clinicians in the IMPACT model is available at workshops and on the Internet (see Related resources). Successful implementation requires addressing operational and financing issues, and potential funding sources for primary care-based management have been described.¹³

Figure 3-step IMPACT intervention for a typical older patient with depression

PST-PC: Problem Solving Treatment in Primary Care

ECT: Electroconvulsive therapy

Source: Adapted and reprinted with permission from Unützer J, Katon WJ, Williams JW, et al. Improving primary care for depression in late life: the design of a multi-center randomized trial. Medical Care 2001;39:785.

How impact was tested

The IMPACT team care model has been tested in a randomized trial with 1,801 depressed older adults from 18 primary care clinics in 8 U.S. health care organizations. These included feefor-service plans, health maintenance organizations, and Veterans Affairs clinics with more than 450 PCPs in rural and urban settings.

Patients were randomly assigned to receive IMPACT care for 12 months or care as usual (in usual care, the patient and provider were informed that the patient met diagnostic criteria for major depression or dysthymia). Patients assigned to usual care could engage in any depression treatment the provider normally used, including referral to specialty mental health services.¹⁴

Result: Better outcomes. IMPACT was more effective than usual care for late-life depression in all 8 organizations over 2 years.^15,16 Overall, IMPACT doubled the effect of usual care. IMPACT patients:

• experienced >100 additional depression-free days¹⁷
• showed substantial improvements in physical and social functioning^15,17 and quality of life,¹⁵ even 12 months after IMPACT resources were withdrawn¹⁸
• experienced less pain and pain-related functional impairment¹⁹
• had significantly less suicidal ideation.²⁰

IMPACT care also was more cost-effective than usual care for depression in older adults with and without comorbid medical illnesses.¹⁷

From research to practice

Based on the robust study outcomes, researchers received a grant from the John A. Hartford Foundation to provide materials, training, and technical assistance to organizations interested in adopting IMPACT. More than 20 health care organizations have used IMPACT, and several have completed program evaluations showing outcomes matching the original IMPACT trial’s.

Kaiser Permanente of Southern California serves 3 million members. Before adopting collaborative care for depression, Kaiser conducted a pilot study of the project IMPACT model modified to fit its health care system. Adaptations included:

• expanding the program to serve depressed adults of all ages
• adding medical assistants to the care team to help with patient follow-up
• adding a “depression class” to offer group-based patient education
• providing psychiatric consultation to the care manager and primary care providers by telephone.

Kaiser investigators compared the outcomes of 300 patients who experienced the adapted program with outcomes in 140 usual-care patients and 140 intervention patients in the original IMPACT study. The effects on depression symptoms were equal to those achieved in the original IMPACT study, with 68% of depressed older adults showing substantial improvement (at least a 50% reduction in depression symptoms) at 6 months.²¹

Related resources

• Project IMPACT. www.impact-uw.org.
• American Association of Geriatric Psychiatry. www.aagpgpa.org.
• Positive Aging Resource Center. http://positiveaging.org.

Acknowledgment

This paper is based on a presentation by Dr. Unützer for the Distinguished Scientist Award at the American Association of Geriatric Psychiatry annual meeting in Puerto Rico, March 10, 2006.

Few depressed older adults seek help from psychiatrists. Those who receive mental health treatment most likely do so in primary care settings. Yet primary care physicians (PCPs) often are ill-equipped to effectively treat depression while managing older patients’ numerous acute and chronic medical conditions.

If depressed older patients won’t go to a psychiatrist, why not bring the psychiatrist to the patients? This article describes a clinically tested approach called project IMPACT that links psychiatrists to primary care teams and dramatically improves depression treatment in older adults.

Untreated geriatric depression

Preference for primary care. Major depression is rare in community-living older adults (1% to 3% prevalence), but:

• 5% to 10% of older primary care patients meet DSM-IV-TR criteria for major depression¹
• approximately one-half of depressed older adults report a primary care visit when a mental health problem was addressed during the past year.²

Box 1

Box 2

Only 8% of depressed older adults visit a mental health specialist in a given year, compared with 25% of depressed younger adults.² Even when PCPs refer older patients to a mental health specialist, only 50% follow through (Box 1).^3,4

Barriers to effective care. Depression diagnosis and treatment by PCPs has improved, but a recent survey suggests that with usual treatment:

• only 1 in 5 depressed older adults treated in a primary care practice experiences substantial improvement over 12 months
• only 1 in 10 becomes symptom-free.⁵

Many depressed older adults do not realize they have depression and visit their PCPs complaining of physical symptoms (Box 2).^6-8 Their limited knowledge about depression or fear of being labeled “mentally ill” deters them from disclosing a depressed mood. They and their PCPs may think depression is inevitable with aging.

PCPs also may lack training to differentiate mood disorders, transitory reactions to life-events, or depression caused by medical illness. Their busy schedules limit time to address and prioritize patient concerns about acute and chronic medical problems (Box 3). Thus depression “falls through the cracks.”

Prescribing concerns. PCPs who feel uncomfortable prescribing antidepressants to older patients may be concerned about side effects and maintain dosages at low starting levels instead of titrating up to a therapeutic range.

Collaborative care

One way to overcome these barriers is to integrate mental health providers into primary care to support and augment PCP-prescribed depression treatment. Collaborative care can become an effective, efficient way to provide high-quality depression care to older patients who might otherwise go untreated.⁹

Box 3

Project IMPACT. One such model—project IMPACT (Improving Mood: Promoting Access to Collaborative Treatment for Late-life Depression)—was developed with support from the John A. Hartford Foundation and California Healthcare Foundation. At its heart is a depression care manager or depression clinical specialist—typically a nurse, social worker, or psychologist—who works in a primary care practice. Other team members include the patient’s PCP, the patient, and a consulting psychiatrist.

The care manager works closely with the PCP by:

• educating patients about depression
• coaching patients in pleasant events scheduling/behavioral activation
• supporting the PCP’s antidepressant management
• offering patients a brief course of evidence-based counseling, such as Problem Solving Treatment in Primary Care¹⁰
• measuring patients’ depressive symptoms at treatment onset and regularly thereafter with a tool such as the 9-item depression scale of the Patient Health Questionnaire (PHQ-9).¹¹

Consulting psychiatrist’s role. If a patient is not sufficiently improved after 8 to 10 weeks, the care manager works with the PCP and psychiatrist to change treatment according to an evidence-based algorithm (Figure).¹² In large health care systems, the psychiatrist meets weekly with the care manager to review treatment plans for approximately 100 depressed older adults, with particular attention to those who are not improving. Psychiatrists may see patients in person or facilitate other specialty mental health treatment, as indicated.

Patients are encouraged to choose an antidepressant prescribed by their PCP, psychotherapy provided by the care manager in the primary care setting, or both. The patient and PCP make treatment decisions with support from the care manager and psychiatrist.

After depressive symptoms remit, the care manager completes a relapse prevention plan with the patient. This includes:

• steps the patient can take to prevent a relapse
• identifying relapse warning signs
• an action plan if depressive symptoms recur.

Training for clinicians in the IMPACT model is available at workshops and on the Internet (see Related resources). Successful implementation requires addressing operational and financing issues, and potential funding sources for primary care-based management have been described.¹³

Figure 3-step IMPACT intervention for a typical older patient with depression

PST-PC: Problem Solving Treatment in Primary Care

ECT: Electroconvulsive therapy

Source: Adapted and reprinted with permission from Unützer J, Katon WJ, Williams JW, et al. Improving primary care for depression in late life: the design of a multi-center randomized trial. Medical Care 2001;39:785.

How impact was tested

The IMPACT team care model has been tested in a randomized trial with 1,801 depressed older adults from 18 primary care clinics in 8 U.S. health care organizations. These included feefor-service plans, health maintenance organizations, and Veterans Affairs clinics with more than 450 PCPs in rural and urban settings.

Patients were randomly assigned to receive IMPACT care for 12 months or care as usual (in usual care, the patient and provider were informed that the patient met diagnostic criteria for major depression or dysthymia). Patients assigned to usual care could engage in any depression treatment the provider normally used, including referral to specialty mental health services.¹⁴

Result: Better outcomes. IMPACT was more effective than usual care for late-life depression in all 8 organizations over 2 years.^15,16 Overall, IMPACT doubled the effect of usual care. IMPACT patients:

• experienced >100 additional depression-free days¹⁷
• showed substantial improvements in physical and social functioning^15,17 and quality of life,¹⁵ even 12 months after IMPACT resources were withdrawn¹⁸
• experienced less pain and pain-related functional impairment¹⁹
• had significantly less suicidal ideation.²⁰

IMPACT care also was more cost-effective than usual care for depression in older adults with and without comorbid medical illnesses.¹⁷

From research to practice

Based on the robust study outcomes, researchers received a grant from the John A. Hartford Foundation to provide materials, training, and technical assistance to organizations interested in adopting IMPACT. More than 20 health care organizations have used IMPACT, and several have completed program evaluations showing outcomes matching the original IMPACT trial’s.

Kaiser Permanente of Southern California serves 3 million members. Before adopting collaborative care for depression, Kaiser conducted a pilot study of the project IMPACT model modified to fit its health care system. Adaptations included:

• expanding the program to serve depressed adults of all ages
• adding medical assistants to the care team to help with patient follow-up
• adding a “depression class” to offer group-based patient education
• providing psychiatric consultation to the care manager and primary care providers by telephone.

Kaiser investigators compared the outcomes of 300 patients who experienced the adapted program with outcomes in 140 usual-care patients and 140 intervention patients in the original IMPACT study. The effects on depression symptoms were equal to those achieved in the original IMPACT study, with 68% of depressed older adults showing substantial improvement (at least a 50% reduction in depression symptoms) at 6 months.²¹

Related resources

• Project IMPACT. www.impact-uw.org.
• American Association of Geriatric Psychiatry. www.aagpgpa.org.
• Positive Aging Resource Center. http://positiveaging.org.

Acknowledgment

This paper is based on a presentation by Dr. Unützer for the Distinguished Scientist Award at the American Association of Geriatric Psychiatry annual meeting in Puerto Rico, March 10, 2006.

Few depressed older adults seek help from psychiatrists. Those who receive mental health treatment most likely do so in primary care settings. Yet primary care physicians (PCPs) often are ill-equipped to effectively treat depression while managing older patients’ numerous acute and chronic medical conditions.

If depressed older patients won’t go to a psychiatrist, why not bring the psychiatrist to the patients? This article describes a clinically tested approach called project IMPACT that links psychiatrists to primary care teams and dramatically improves depression treatment in older adults.

Untreated geriatric depression

Preference for primary care. Major depression is rare in community-living older adults (1% to 3% prevalence), but:

• 5% to 10% of older primary care patients meet DSM-IV-TR criteria for major depression¹
• approximately one-half of depressed older adults report a primary care visit when a mental health problem was addressed during the past year.²

Box 1

Box 2

Only 8% of depressed older adults visit a mental health specialist in a given year, compared with 25% of depressed younger adults.² Even when PCPs refer older patients to a mental health specialist, only 50% follow through (Box 1).^3,4

Barriers to effective care. Depression diagnosis and treatment by PCPs has improved, but a recent survey suggests that with usual treatment:

• only 1 in 5 depressed older adults treated in a primary care practice experiences substantial improvement over 12 months
• only 1 in 10 becomes symptom-free.⁵

Many depressed older adults do not realize they have depression and visit their PCPs complaining of physical symptoms (Box 2).^6-8 Their limited knowledge about depression or fear of being labeled “mentally ill” deters them from disclosing a depressed mood. They and their PCPs may think depression is inevitable with aging.

PCPs also may lack training to differentiate mood disorders, transitory reactions to life-events, or depression caused by medical illness. Their busy schedules limit time to address and prioritize patient concerns about acute and chronic medical problems (Box 3). Thus depression “falls through the cracks.”

Prescribing concerns. PCPs who feel uncomfortable prescribing antidepressants to older patients may be concerned about side effects and maintain dosages at low starting levels instead of titrating up to a therapeutic range.

Collaborative care

One way to overcome these barriers is to integrate mental health providers into primary care to support and augment PCP-prescribed depression treatment. Collaborative care can become an effective, efficient way to provide high-quality depression care to older patients who might otherwise go untreated.⁹

Box 3

Project IMPACT. One such model—project IMPACT (Improving Mood: Promoting Access to Collaborative Treatment for Late-life Depression)—was developed with support from the John A. Hartford Foundation and California Healthcare Foundation. At its heart is a depression care manager or depression clinical specialist—typically a nurse, social worker, or psychologist—who works in a primary care practice. Other team members include the patient’s PCP, the patient, and a consulting psychiatrist.

The care manager works closely with the PCP by:

• educating patients about depression
• coaching patients in pleasant events scheduling/behavioral activation
• supporting the PCP’s antidepressant management
• offering patients a brief course of evidence-based counseling, such as Problem Solving Treatment in Primary Care¹⁰
• measuring patients’ depressive symptoms at treatment onset and regularly thereafter with a tool such as the 9-item depression scale of the Patient Health Questionnaire (PHQ-9).¹¹

Consulting psychiatrist’s role. If a patient is not sufficiently improved after 8 to 10 weeks, the care manager works with the PCP and psychiatrist to change treatment according to an evidence-based algorithm (Figure).¹² In large health care systems, the psychiatrist meets weekly with the care manager to review treatment plans for approximately 100 depressed older adults, with particular attention to those who are not improving. Psychiatrists may see patients in person or facilitate other specialty mental health treatment, as indicated.

Patients are encouraged to choose an antidepressant prescribed by their PCP, psychotherapy provided by the care manager in the primary care setting, or both. The patient and PCP make treatment decisions with support from the care manager and psychiatrist.

After depressive symptoms remit, the care manager completes a relapse prevention plan with the patient. This includes:

• steps the patient can take to prevent a relapse
• identifying relapse warning signs
• an action plan if depressive symptoms recur.

Training for clinicians in the IMPACT model is available at workshops and on the Internet (see Related resources). Successful implementation requires addressing operational and financing issues, and potential funding sources for primary care-based management have been described.¹³

Figure 3-step IMPACT intervention for a typical older patient with depression

PST-PC: Problem Solving Treatment in Primary Care

ECT: Electroconvulsive therapy

Source: Adapted and reprinted with permission from Unützer J, Katon WJ, Williams JW, et al. Improving primary care for depression in late life: the design of a multi-center randomized trial. Medical Care 2001;39:785.

How impact was tested

The IMPACT team care model has been tested in a randomized trial with 1,801 depressed older adults from 18 primary care clinics in 8 U.S. health care organizations. These included feefor-service plans, health maintenance organizations, and Veterans Affairs clinics with more than 450 PCPs in rural and urban settings.

Patients were randomly assigned to receive IMPACT care for 12 months or care as usual (in usual care, the patient and provider were informed that the patient met diagnostic criteria for major depression or dysthymia). Patients assigned to usual care could engage in any depression treatment the provider normally used, including referral to specialty mental health services.¹⁴

Result: Better outcomes. IMPACT was more effective than usual care for late-life depression in all 8 organizations over 2 years.^15,16 Overall, IMPACT doubled the effect of usual care. IMPACT patients:

• experienced >100 additional depression-free days¹⁷
• showed substantial improvements in physical and social functioning^15,17 and quality of life,¹⁵ even 12 months after IMPACT resources were withdrawn¹⁸
• experienced less pain and pain-related functional impairment¹⁹
• had significantly less suicidal ideation.²⁰

IMPACT care also was more cost-effective than usual care for depression in older adults with and without comorbid medical illnesses.¹⁷

From research to practice

Based on the robust study outcomes, researchers received a grant from the John A. Hartford Foundation to provide materials, training, and technical assistance to organizations interested in adopting IMPACT. More than 20 health care organizations have used IMPACT, and several have completed program evaluations showing outcomes matching the original IMPACT trial’s.

Kaiser Permanente of Southern California serves 3 million members. Before adopting collaborative care for depression, Kaiser conducted a pilot study of the project IMPACT model modified to fit its health care system. Adaptations included:

• expanding the program to serve depressed adults of all ages
• adding medical assistants to the care team to help with patient follow-up
• adding a “depression class” to offer group-based patient education
• providing psychiatric consultation to the care manager and primary care providers by telephone.

Kaiser investigators compared the outcomes of 300 patients who experienced the adapted program with outcomes in 140 usual-care patients and 140 intervention patients in the original IMPACT study. The effects on depression symptoms were equal to those achieved in the original IMPACT study, with 68% of depressed older adults showing substantial improvement (at least a 50% reduction in depression symptoms) at 6 months.²¹

Related resources

• Project IMPACT. www.impact-uw.org.
• American Association of Geriatric Psychiatry. www.aagpgpa.org.
• Positive Aging Resource Center. http://positiveaging.org.

Acknowledgment

This paper is based on a presentation by Dr. Unützer for the Distinguished Scientist Award at the American Association of Geriatric Psychiatry annual meeting in Puerto Rico, March 10, 2006.

The Positive and Negative Syndrome Scale (PANSS) is moving from research into clinical practice as demand grows for objective rating scales. We see the PANSS becoming a treatment and planning tool for psychiatry, just as the electrocardiogram evolved into a measure of cardiac status in medical practice.

Based on our experience in co-authoring (LA Opler) and using the PANSS, we describe how you can use it to:

• identify psychotic symptoms for targeted treatment
  
• predict with greater accuracy how patients will respond to the treatment you provide.
  

Standardized Assessments

The PANSS first gained stature in studies that established the efficacy of second-generation antipsychotics (SGAs).^1-6 But its authors⁷ also envisioned the scale as a useful tool to help practicing clinicians treat patients with schizophrenia and other psychotic disorders.

Twenty years of experience has shown the PANSS to be a reliable and valid severity symptom scale for schizophrenia, bipolar disorder, and other serious mental illnesses. It is particularly useful to track changes in positive and negative symptoms.⁸

Traditionally, psychiatric evaluation has been impressionistic and subjective, but standardized tools provide a common language while introducing objective, empiric measures of clinical status. Because patients with mental disorders are treated by providers from psychiatry, psychology, social work, nursing, and other mental health disciplines, having standardized benchmarks to assess symptom severity can facilitate an integrated approach. And because the PANSS has been translated into some 40 languages and is being adopted in clinical settings worldwide, it provides a universal means of communicating information about a patient’s clinical status.

Panss Scoring System

The PANSS includes 30 items, each rated from 1 (absent) to 7 (extreme). In theory, a patient rated “absent” (or 1) on all items would receive a total score of 30, and a patient rated “extreme” (or 7) on all items would receive a total score of 210. In the real world, though, no one sees these extremes. Stable outpatients usually score 60 to 80. Inpatients’ scores rarely exceed 80 to 150, even in “treatment refractory” cases.

The 30 items are arranged as 7 positive symptom subscale items (P1-P7), 7 negative symptom subscale items (N1-N7), and 16 general psychopathology symptom items (G1-G16) (Table 1). Each item has a definition and a basis for rating. The first question you need to answer when rating a patient is whether the item is absent or present.

How it works. For example, the PANSS defines delusions as “beliefs that are unfounded, unrealistic, and idiosyncratic,” and the basis for rating is “thought content expressed during the interview and its influence on the patient’s social relations and behavior as reported from primary care workers or family.” If the definition does not apply to your patient, you rate this item 1 or absent. If the definition does apply, “anchoring points” for each level of severity are provided (Table 2), and you decide which anchoring point best describes the patient’s functioning during the interview and the preceding week.

Time required. In research, gathering informant information, conducting the interview, and generating reliable ratings takes 45 to 60 minutes. In clinical settings, if you know your patient and can function as informant and interviewer, you probably can obtain accurate ratings in 30 to 45 minutes.

Ideally, you would use the Structured Clinical Interview for the PANSS (SCI-PANSS), though clinicians who know this instrument well may prefer a less structured interview that covers all areas of inquiry. Accurate PANSS scores are easy to generate on all 30 items by combining information from the interview with information about how the patient has functioned in the past week.

PANSS ratings are not meant to be obtained after every patient contact but rather as often as needed to guide clinical treatment. For example, you might obtain a PANSS rating:

• when an inpatient is first admitted
  
• before starting a new medication
  
• weeks or months later to gauge the new treatment’s effect.
  

The PANSS manual—a complete individual kit costs approximately $200—or licenses to use multiple copies are available from the copyright holder, MultiHealth Systems, Inc. (see Related resources).

Table 1

Subscales of the 30-item Positive and Negative Syndrome Scale (PANSS)

7 levels of severity on the PANSS for characterizing delusions

Gauging Symptom Severity

Treatment planning. Clinicians at the Rochester (New York) Psychiatric Center use the PANSS to assess symptom severity in inpatients with schizophrenia and other psychotic disorders.

Within 1 week of admission, patients are evaluated on the 30 items by a team of experienced PANSS raters. Symptoms identified by the PANSS become targets in individualized treatment plans. Follow-up PANSS assessments help determine if treatment has improved the selected symptoms.

Tracking patient progress. Florida State Hospital uses the PANSS to track progress of patients with serious mental illnesses. Data collected over 8 years from >19,000 PANSS assessments in a multilingual, multicultural population suggests that the PANSS:

• aids in decision making for medical and nonmedical aspects of care for individual patients
  
• can help determine if changes in agency prescribing practices affect patient symptom profiles and severity, one indicator of how policy and guidelines translate into patient care.⁹
  

Predicting Outcomes

The PANSS has been shown to predict course of illness and treatment response, functional outcomes (including aggression), and long-term outcomes (including deterioration). Adjusting treatments to achieve optimal PANSS scores also can help clinicians achieve remission of their patients’ psychotic symptoms (Box).^11,12

Box

Functional outcomes. Steinert et al¹⁴ used the PANSS to rate 199 inpatients within 24 hours of admission into an acute psychiatric ward. After discharge, each patient was assessed retrospectively for aggressive behavior. The conceptual disorganization and hostility items from the positive sub-scale could predict violent behaviors during inpatient treatment with statistical significance.

Long-term outcomes. White et al¹⁵ assessed older schizophrenia inpatients, using the PANSS at baseline and after 1 year. The researchers looked specifically at the “activation factor”—six PANSS items including hostility, poor impulse control, excitement, uncooperativeness, poor rapport, and tension. Poor outcome and low discharge rates were directly correlated with high baseline scores on the PANSS activation factor (PANSS-AF).

Deterioration. Goetz et al¹⁶ showed that residual positive symptoms were significantly related to deteriorating course of illness, even when patients adhered to their medications. These results suggest that even subtle symptom elevations as measured by the PANSS can predict deterioration.

Related resources

• The PANSS Institute. Information on how to attain, maintain, and retain high reliability as a PANSS rater. www.panss.org.
  
• MultiHealth Systems, Inc. (publisher and copyright holder) to purchase the Positive and Negative Syndrome Scale (PANSS). www.mhs.com.
  
• Opler LA, Ramirez PM, Mougios VA. Measuring outcome in serious mental illness. In: IsHak WW, Burt T, Sederer L (eds). Outcome measurement in psychiatry: a critical review. Washington, DC: American Psychiatric Press; 2002.
  

Dr. Lewis A. Opler receives royalties from MultiHealth Systems, Inc. on sales of the Positive and Negative Syndrome Scale (PANSS) Manual, the Structured Clinical Interview for the PANSS (SCI-PANSS), and the Informant Questionnaire for the PANSS (IQ-PANSS).

Dr. Mark G. Opler is Executive Director of The PANSS Institute.

Acknowledgement

This work was supported in part by NIMH grant K24 MH01699 (DM).

The Positive and Negative Syndrome Scale (PANSS) is moving from research into clinical practice as demand grows for objective rating scales. We see the PANSS becoming a treatment and planning tool for psychiatry, just as the electrocardiogram evolved into a measure of cardiac status in medical practice.

Based on our experience in co-authoring (LA Opler) and using the PANSS, we describe how you can use it to:

• identify psychotic symptoms for targeted treatment
  
• predict with greater accuracy how patients will respond to the treatment you provide.
  

Standardized Assessments

The PANSS first gained stature in studies that established the efficacy of second-generation antipsychotics (SGAs).^1-6 But its authors⁷ also envisioned the scale as a useful tool to help practicing clinicians treat patients with schizophrenia and other psychotic disorders.

Twenty years of experience has shown the PANSS to be a reliable and valid severity symptom scale for schizophrenia, bipolar disorder, and other serious mental illnesses. It is particularly useful to track changes in positive and negative symptoms.⁸

Traditionally, psychiatric evaluation has been impressionistic and subjective, but standardized tools provide a common language while introducing objective, empiric measures of clinical status. Because patients with mental disorders are treated by providers from psychiatry, psychology, social work, nursing, and other mental health disciplines, having standardized benchmarks to assess symptom severity can facilitate an integrated approach. And because the PANSS has been translated into some 40 languages and is being adopted in clinical settings worldwide, it provides a universal means of communicating information about a patient’s clinical status.

Panss Scoring System

The PANSS includes 30 items, each rated from 1 (absent) to 7 (extreme). In theory, a patient rated “absent” (or 1) on all items would receive a total score of 30, and a patient rated “extreme” (or 7) on all items would receive a total score of 210. In the real world, though, no one sees these extremes. Stable outpatients usually score 60 to 80. Inpatients’ scores rarely exceed 80 to 150, even in “treatment refractory” cases.

The 30 items are arranged as 7 positive symptom subscale items (P1-P7), 7 negative symptom subscale items (N1-N7), and 16 general psychopathology symptom items (G1-G16) (Table 1). Each item has a definition and a basis for rating. The first question you need to answer when rating a patient is whether the item is absent or present.

How it works. For example, the PANSS defines delusions as “beliefs that are unfounded, unrealistic, and idiosyncratic,” and the basis for rating is “thought content expressed during the interview and its influence on the patient’s social relations and behavior as reported from primary care workers or family.” If the definition does not apply to your patient, you rate this item 1 or absent. If the definition does apply, “anchoring points” for each level of severity are provided (Table 2), and you decide which anchoring point best describes the patient’s functioning during the interview and the preceding week.

Time required. In research, gathering informant information, conducting the interview, and generating reliable ratings takes 45 to 60 minutes. In clinical settings, if you know your patient and can function as informant and interviewer, you probably can obtain accurate ratings in 30 to 45 minutes.

Ideally, you would use the Structured Clinical Interview for the PANSS (SCI-PANSS), though clinicians who know this instrument well may prefer a less structured interview that covers all areas of inquiry. Accurate PANSS scores are easy to generate on all 30 items by combining information from the interview with information about how the patient has functioned in the past week.

PANSS ratings are not meant to be obtained after every patient contact but rather as often as needed to guide clinical treatment. For example, you might obtain a PANSS rating:

• when an inpatient is first admitted
  
• before starting a new medication
  
• weeks or months later to gauge the new treatment’s effect.
  

The PANSS manual—a complete individual kit costs approximately $200—or licenses to use multiple copies are available from the copyright holder, MultiHealth Systems, Inc. (see Related resources).

Table 1

Subscales of the 30-item Positive and Negative Syndrome Scale (PANSS)

7 levels of severity on the PANSS for characterizing delusions

Gauging Symptom Severity

Treatment planning. Clinicians at the Rochester (New York) Psychiatric Center use the PANSS to assess symptom severity in inpatients with schizophrenia and other psychotic disorders.

Within 1 week of admission, patients are evaluated on the 30 items by a team of experienced PANSS raters. Symptoms identified by the PANSS become targets in individualized treatment plans. Follow-up PANSS assessments help determine if treatment has improved the selected symptoms.

Tracking patient progress. Florida State Hospital uses the PANSS to track progress of patients with serious mental illnesses. Data collected over 8 years from >19,000 PANSS assessments in a multilingual, multicultural population suggests that the PANSS:

• aids in decision making for medical and nonmedical aspects of care for individual patients
  
• can help determine if changes in agency prescribing practices affect patient symptom profiles and severity, one indicator of how policy and guidelines translate into patient care.⁹
  

Predicting Outcomes

The PANSS has been shown to predict course of illness and treatment response, functional outcomes (including aggression), and long-term outcomes (including deterioration). Adjusting treatments to achieve optimal PANSS scores also can help clinicians achieve remission of their patients’ psychotic symptoms (Box).^11,12

Box

Functional outcomes. Steinert et al¹⁴ used the PANSS to rate 199 inpatients within 24 hours of admission into an acute psychiatric ward. After discharge, each patient was assessed retrospectively for aggressive behavior. The conceptual disorganization and hostility items from the positive sub-scale could predict violent behaviors during inpatient treatment with statistical significance.

Long-term outcomes. White et al¹⁵ assessed older schizophrenia inpatients, using the PANSS at baseline and after 1 year. The researchers looked specifically at the “activation factor”—six PANSS items including hostility, poor impulse control, excitement, uncooperativeness, poor rapport, and tension. Poor outcome and low discharge rates were directly correlated with high baseline scores on the PANSS activation factor (PANSS-AF).

Deterioration. Goetz et al¹⁶ showed that residual positive symptoms were significantly related to deteriorating course of illness, even when patients adhered to their medications. These results suggest that even subtle symptom elevations as measured by the PANSS can predict deterioration.

Related resources

• The PANSS Institute. Information on how to attain, maintain, and retain high reliability as a PANSS rater. www.panss.org.
  
• MultiHealth Systems, Inc. (publisher and copyright holder) to purchase the Positive and Negative Syndrome Scale (PANSS). www.mhs.com.
  
• Opler LA, Ramirez PM, Mougios VA. Measuring outcome in serious mental illness. In: IsHak WW, Burt T, Sederer L (eds). Outcome measurement in psychiatry: a critical review. Washington, DC: American Psychiatric Press; 2002.
  

Dr. Lewis A. Opler receives royalties from MultiHealth Systems, Inc. on sales of the Positive and Negative Syndrome Scale (PANSS) Manual, the Structured Clinical Interview for the PANSS (SCI-PANSS), and the Informant Questionnaire for the PANSS (IQ-PANSS).

Dr. Mark G. Opler is Executive Director of The PANSS Institute.

Acknowledgement

This work was supported in part by NIMH grant K24 MH01699 (DM).

The Positive and Negative Syndrome Scale (PANSS) is moving from research into clinical practice as demand grows for objective rating scales. We see the PANSS becoming a treatment and planning tool for psychiatry, just as the electrocardiogram evolved into a measure of cardiac status in medical practice.

Based on our experience in co-authoring (LA Opler) and using the PANSS, we describe how you can use it to:

• identify psychotic symptoms for targeted treatment
  
• predict with greater accuracy how patients will respond to the treatment you provide.
  

Standardized Assessments

The PANSS first gained stature in studies that established the efficacy of second-generation antipsychotics (SGAs).^1-6 But its authors⁷ also envisioned the scale as a useful tool to help practicing clinicians treat patients with schizophrenia and other psychotic disorders.

Twenty years of experience has shown the PANSS to be a reliable and valid severity symptom scale for schizophrenia, bipolar disorder, and other serious mental illnesses. It is particularly useful to track changes in positive and negative symptoms.⁸

Traditionally, psychiatric evaluation has been impressionistic and subjective, but standardized tools provide a common language while introducing objective, empiric measures of clinical status. Because patients with mental disorders are treated by providers from psychiatry, psychology, social work, nursing, and other mental health disciplines, having standardized benchmarks to assess symptom severity can facilitate an integrated approach. And because the PANSS has been translated into some 40 languages and is being adopted in clinical settings worldwide, it provides a universal means of communicating information about a patient’s clinical status.

Panss Scoring System

The PANSS includes 30 items, each rated from 1 (absent) to 7 (extreme). In theory, a patient rated “absent” (or 1) on all items would receive a total score of 30, and a patient rated “extreme” (or 7) on all items would receive a total score of 210. In the real world, though, no one sees these extremes. Stable outpatients usually score 60 to 80. Inpatients’ scores rarely exceed 80 to 150, even in “treatment refractory” cases.

The 30 items are arranged as 7 positive symptom subscale items (P1-P7), 7 negative symptom subscale items (N1-N7), and 16 general psychopathology symptom items (G1-G16) (Table 1). Each item has a definition and a basis for rating. The first question you need to answer when rating a patient is whether the item is absent or present.

How it works. For example, the PANSS defines delusions as “beliefs that are unfounded, unrealistic, and idiosyncratic,” and the basis for rating is “thought content expressed during the interview and its influence on the patient’s social relations and behavior as reported from primary care workers or family.” If the definition does not apply to your patient, you rate this item 1 or absent. If the definition does apply, “anchoring points” for each level of severity are provided (Table 2), and you decide which anchoring point best describes the patient’s functioning during the interview and the preceding week.

Time required. In research, gathering informant information, conducting the interview, and generating reliable ratings takes 45 to 60 minutes. In clinical settings, if you know your patient and can function as informant and interviewer, you probably can obtain accurate ratings in 30 to 45 minutes.

Ideally, you would use the Structured Clinical Interview for the PANSS (SCI-PANSS), though clinicians who know this instrument well may prefer a less structured interview that covers all areas of inquiry. Accurate PANSS scores are easy to generate on all 30 items by combining information from the interview with information about how the patient has functioned in the past week.

PANSS ratings are not meant to be obtained after every patient contact but rather as often as needed to guide clinical treatment. For example, you might obtain a PANSS rating:

• when an inpatient is first admitted
  
• before starting a new medication
  
• weeks or months later to gauge the new treatment’s effect.
  

The PANSS manual—a complete individual kit costs approximately $200—or licenses to use multiple copies are available from the copyright holder, MultiHealth Systems, Inc. (see Related resources).

Table 1

Subscales of the 30-item Positive and Negative Syndrome Scale (PANSS)

7 levels of severity on the PANSS for characterizing delusions

Gauging Symptom Severity

Treatment planning. Clinicians at the Rochester (New York) Psychiatric Center use the PANSS to assess symptom severity in inpatients with schizophrenia and other psychotic disorders.

Within 1 week of admission, patients are evaluated on the 30 items by a team of experienced PANSS raters. Symptoms identified by the PANSS become targets in individualized treatment plans. Follow-up PANSS assessments help determine if treatment has improved the selected symptoms.

Tracking patient progress. Florida State Hospital uses the PANSS to track progress of patients with serious mental illnesses. Data collected over 8 years from >19,000 PANSS assessments in a multilingual, multicultural population suggests that the PANSS:

• aids in decision making for medical and nonmedical aspects of care for individual patients
  
• can help determine if changes in agency prescribing practices affect patient symptom profiles and severity, one indicator of how policy and guidelines translate into patient care.⁹
  

Predicting Outcomes

The PANSS has been shown to predict course of illness and treatment response, functional outcomes (including aggression), and long-term outcomes (including deterioration). Adjusting treatments to achieve optimal PANSS scores also can help clinicians achieve remission of their patients’ psychotic symptoms (Box).^11,12

Box

Functional outcomes. Steinert et al¹⁴ used the PANSS to rate 199 inpatients within 24 hours of admission into an acute psychiatric ward. After discharge, each patient was assessed retrospectively for aggressive behavior. The conceptual disorganization and hostility items from the positive sub-scale could predict violent behaviors during inpatient treatment with statistical significance.

Long-term outcomes. White et al¹⁵ assessed older schizophrenia inpatients, using the PANSS at baseline and after 1 year. The researchers looked specifically at the “activation factor”—six PANSS items including hostility, poor impulse control, excitement, uncooperativeness, poor rapport, and tension. Poor outcome and low discharge rates were directly correlated with high baseline scores on the PANSS activation factor (PANSS-AF).

Deterioration. Goetz et al¹⁶ showed that residual positive symptoms were significantly related to deteriorating course of illness, even when patients adhered to their medications. These results suggest that even subtle symptom elevations as measured by the PANSS can predict deterioration.

Related resources

• The PANSS Institute. Information on how to attain, maintain, and retain high reliability as a PANSS rater. www.panss.org.
  
• MultiHealth Systems, Inc. (publisher and copyright holder) to purchase the Positive and Negative Syndrome Scale (PANSS). www.mhs.com.
  
• Opler LA, Ramirez PM, Mougios VA. Measuring outcome in serious mental illness. In: IsHak WW, Burt T, Sederer L (eds). Outcome measurement in psychiatry: a critical review. Washington, DC: American Psychiatric Press; 2002.
  

Dr. Lewis A. Opler receives royalties from MultiHealth Systems, Inc. on sales of the Positive and Negative Syndrome Scale (PANSS) Manual, the Structured Clinical Interview for the PANSS (SCI-PANSS), and the Informant Questionnaire for the PANSS (IQ-PANSS).

Dr. Mark G. Opler is Executive Director of The PANSS Institute.

Acknowledgement

This work was supported in part by NIMH grant K24 MH01699 (DM).

Keeping track of outpatients with psychosis can be challenging when multiple clinicians provide care and social services are disjointed. To ensure continuity of care and achieve optimal functional outcome, answer these five questions periodically during long-term treatment.

1. Are you acquainted with the patient’s primary care physician? If not, introduce yourself by e-mail, letter, or telephone.

For some patients, you are the primary physician by default, so know some medical monitoring guidelines such as the Mount Sinai Consensus Conference Recommendations for Physical Health Monitoring in Schizophrenia.¹ Also encourage the patient to exercise, stop smoking, and lose weight, if necessary.²

2. When did you last request a written consultation from a specialist? Documenting consultations helps other clinicians follow the steps you took to care for the patient. Requesting and preparing the written consultation also forces you and the consultant to think about the exact nature of the patient’s problem.

3. How strong is the patient’s social network? To identify who can help with the patient’s care, draw a schematic of his or her family tree. Determine who lives with the patient or lives nearby and regularly sees him or her. Repeat this exercise every year because family networks change.

4. Do you have data regarding the patient’s complaints? Get as much information as possible to determine a cause. For example, if the patient has:

• Insomnia—Have the patient or caregiver complete a sleep log and count the number of caffeinated drinks and cigarettes the patient consumes daily.
  
• Overweight/obesity—Weigh the patient at each visit and suggest that the patient or caregiver keep a food diary.
  
• Worsening psychosis—Use a psychopathology rating scale such as the Brief Psychiatric Rating Scale. Count pills, and order antipsychotic blood levels and urine drug screens if necessary to test for medication non-adherence or unrecognized drug use.
  
• Possible cognitive problems—Order neuropsychological tests, particularly to assess memory, attention, and executive function.
  

5. How much progress can you expect in 1 year? To help you set feasible rehabilitation goals, make a schedule of a typical week in the patient’s life. This schedule will suggest how to engage the patient in work, family, and the community. Have the patient bring in a resume or work history as a starting point for discussion. Repeat this exercise annually.

Keeping track of outpatients with psychosis can be challenging when multiple clinicians provide care and social services are disjointed. To ensure continuity of care and achieve optimal functional outcome, answer these five questions periodically during long-term treatment.

1. Are you acquainted with the patient’s primary care physician? If not, introduce yourself by e-mail, letter, or telephone.

For some patients, you are the primary physician by default, so know some medical monitoring guidelines such as the Mount Sinai Consensus Conference Recommendations for Physical Health Monitoring in Schizophrenia.¹ Also encourage the patient to exercise, stop smoking, and lose weight, if necessary.²

2. When did you last request a written consultation from a specialist? Documenting consultations helps other clinicians follow the steps you took to care for the patient. Requesting and preparing the written consultation also forces you and the consultant to think about the exact nature of the patient’s problem.

3. How strong is the patient’s social network? To identify who can help with the patient’s care, draw a schematic of his or her family tree. Determine who lives with the patient or lives nearby and regularly sees him or her. Repeat this exercise every year because family networks change.

4. Do you have data regarding the patient’s complaints? Get as much information as possible to determine a cause. For example, if the patient has:

• Insomnia—Have the patient or caregiver complete a sleep log and count the number of caffeinated drinks and cigarettes the patient consumes daily.
  
• Overweight/obesity—Weigh the patient at each visit and suggest that the patient or caregiver keep a food diary.
  
• Worsening psychosis—Use a psychopathology rating scale such as the Brief Psychiatric Rating Scale. Count pills, and order antipsychotic blood levels and urine drug screens if necessary to test for medication non-adherence or unrecognized drug use.
  
• Possible cognitive problems—Order neuropsychological tests, particularly to assess memory, attention, and executive function.
  

5. How much progress can you expect in 1 year? To help you set feasible rehabilitation goals, make a schedule of a typical week in the patient’s life. This schedule will suggest how to engage the patient in work, family, and the community. Have the patient bring in a resume or work history as a starting point for discussion. Repeat this exercise annually.

Keeping track of outpatients with psychosis can be challenging when multiple clinicians provide care and social services are disjointed. To ensure continuity of care and achieve optimal functional outcome, answer these five questions periodically during long-term treatment.

1. Are you acquainted with the patient’s primary care physician? If not, introduce yourself by e-mail, letter, or telephone.

For some patients, you are the primary physician by default, so know some medical monitoring guidelines such as the Mount Sinai Consensus Conference Recommendations for Physical Health Monitoring in Schizophrenia.¹ Also encourage the patient to exercise, stop smoking, and lose weight, if necessary.²

2. When did you last request a written consultation from a specialist? Documenting consultations helps other clinicians follow the steps you took to care for the patient. Requesting and preparing the written consultation also forces you and the consultant to think about the exact nature of the patient’s problem.

3. How strong is the patient’s social network? To identify who can help with the patient’s care, draw a schematic of his or her family tree. Determine who lives with the patient or lives nearby and regularly sees him or her. Repeat this exercise every year because family networks change.

4. Do you have data regarding the patient’s complaints? Get as much information as possible to determine a cause. For example, if the patient has:

• Insomnia—Have the patient or caregiver complete a sleep log and count the number of caffeinated drinks and cigarettes the patient consumes daily.
  
• Overweight/obesity—Weigh the patient at each visit and suggest that the patient or caregiver keep a food diary.
  
• Worsening psychosis—Use a psychopathology rating scale such as the Brief Psychiatric Rating Scale. Count pills, and order antipsychotic blood levels and urine drug screens if necessary to test for medication non-adherence or unrecognized drug use.
  
• Possible cognitive problems—Order neuropsychological tests, particularly to assess memory, attention, and executive function.
  

5. How much progress can you expect in 1 year? To help you set feasible rehabilitation goals, make a schedule of a typical week in the patient’s life. This schedule will suggest how to engage the patient in work, family, and the community. Have the patient bring in a resume or work history as a starting point for discussion. Repeat this exercise annually.

Numerous medications and other substances can appear in a urine drug screen (UDS) as an illicit narcotic (Table). These false positives can:

• lead to incorrect diagnosis and inappropriate intervention, particularly if the result determines treatment
  
• endanger the therapeutic alliance by making the patient uncomfortable and defensive.
  

Table

Substances that may trigger a false urinary drug screen result

Why Drug Screens Are Sometimes Wrong

A UDS for recreational drug use is commonly performed when the patient presents to the ER with acute changes in mental or behavioral status.

Ms. A, age 57, presents to the ER with fluctuating consciousness. The cause is unknown.

Surgical removal of a pituitary tumor 39 years earlier caused hormone deficiencies, seizures, and excessive sleepiness. Symptoms of panhypopituitarism have been managed with medication, and her current regimen includes thyroxine, phenytoin, the proton pump inhibitor pantoprazole, and prednisone. Recently, comorbid depression caused her to skip doses.

ER physicians order a UDS because of Ms. A’s mental status changes. The enzyme-linked immunosorbent (ELISA) toxicology test for alcohol, amphetamines, barbiturates, benzodiazepines, cocaine, opiates, marijuana, and phencyclidine (PCP) is positive for marijuana. When the attending psychiatrist informs Ms. A of the result, she is shocked. She tells the psychiatrist she is active in church and opposes recreational use of narcotics. She adamantly denies using marijuana or other street drugs, alcohol, nicotine, or caffeine.

Eventually, physicians attributed Ms. A’s mental status changes to several underlying medical issues, including Addison’s disease. A thorough review of the case revealed that the proton pump inhibitor pantoprazole caused the false-positive UDS.

UDS tests are sensitive but not highly specific. A medication or other substance with a chemical structure similar to that of the suspected drug can cause a false positive.^1-3

The “WEED” mnemonic spells out steps for critically evaluating UDS test results to ensure appropriate care:

• Write out a list of the patient’s medications. This list may explain the symptoms or help interpret UDS results. If a narcotic dose was recently increased, for example, a UDS might not be needed to confirm what caused the change in mental status.
  
• Examine the patient carefully. Evaluate physical signs, take a thorough medical history, and consider the potential for drug use. Although not impossible, for example, PCP intoxication is not a likely cause of psychosis in nursing home patients.
  
• Equate UDS results with presenting complaints and symptoms. For example, if a patient with sudden syncope tests positive for marijuana, the syncopal symptoms demand further investigation because marijuana is not the likely cause.
  
• Duplicate the UDS screen with confirmatory tests if the result will determine treatment. When UDS results are ambiguous, use highly specific tests such as gas chromatography with mass spectrometry and high-performance liquid chromatography. Although expensive and time consuming, these tests confirm the presence or absence of substances with few false results.
  

Numerous medications and other substances can appear in a urine drug screen (UDS) as an illicit narcotic (Table). These false positives can:

• lead to incorrect diagnosis and inappropriate intervention, particularly if the result determines treatment
  
• endanger the therapeutic alliance by making the patient uncomfortable and defensive.
  

Table

Substances that may trigger a false urinary drug screen result

Why Drug Screens Are Sometimes Wrong

A UDS for recreational drug use is commonly performed when the patient presents to the ER with acute changes in mental or behavioral status.

Ms. A, age 57, presents to the ER with fluctuating consciousness. The cause is unknown.

Surgical removal of a pituitary tumor 39 years earlier caused hormone deficiencies, seizures, and excessive sleepiness. Symptoms of panhypopituitarism have been managed with medication, and her current regimen includes thyroxine, phenytoin, the proton pump inhibitor pantoprazole, and prednisone. Recently, comorbid depression caused her to skip doses.

ER physicians order a UDS because of Ms. A’s mental status changes. The enzyme-linked immunosorbent (ELISA) toxicology test for alcohol, amphetamines, barbiturates, benzodiazepines, cocaine, opiates, marijuana, and phencyclidine (PCP) is positive for marijuana. When the attending psychiatrist informs Ms. A of the result, she is shocked. She tells the psychiatrist she is active in church and opposes recreational use of narcotics. She adamantly denies using marijuana or other street drugs, alcohol, nicotine, or caffeine.

Eventually, physicians attributed Ms. A’s mental status changes to several underlying medical issues, including Addison’s disease. A thorough review of the case revealed that the proton pump inhibitor pantoprazole caused the false-positive UDS.

UDS tests are sensitive but not highly specific. A medication or other substance with a chemical structure similar to that of the suspected drug can cause a false positive.^1-3

The “WEED” mnemonic spells out steps for critically evaluating UDS test results to ensure appropriate care:

• Write out a list of the patient’s medications. This list may explain the symptoms or help interpret UDS results. If a narcotic dose was recently increased, for example, a UDS might not be needed to confirm what caused the change in mental status.
  
• Examine the patient carefully. Evaluate physical signs, take a thorough medical history, and consider the potential for drug use. Although not impossible, for example, PCP intoxication is not a likely cause of psychosis in nursing home patients.
  
• Equate UDS results with presenting complaints and symptoms. For example, if a patient with sudden syncope tests positive for marijuana, the syncopal symptoms demand further investigation because marijuana is not the likely cause.
  
• Duplicate the UDS screen with confirmatory tests if the result will determine treatment. When UDS results are ambiguous, use highly specific tests such as gas chromatography with mass spectrometry and high-performance liquid chromatography. Although expensive and time consuming, these tests confirm the presence or absence of substances with few false results.
  

Numerous medications and other substances can appear in a urine drug screen (UDS) as an illicit narcotic (Table). These false positives can:

• lead to incorrect diagnosis and inappropriate intervention, particularly if the result determines treatment
  
• endanger the therapeutic alliance by making the patient uncomfortable and defensive.
  

Table

Substances that may trigger a false urinary drug screen result

Why Drug Screens Are Sometimes Wrong

A UDS for recreational drug use is commonly performed when the patient presents to the ER with acute changes in mental or behavioral status.

Ms. A, age 57, presents to the ER with fluctuating consciousness. The cause is unknown.

Surgical removal of a pituitary tumor 39 years earlier caused hormone deficiencies, seizures, and excessive sleepiness. Symptoms of panhypopituitarism have been managed with medication, and her current regimen includes thyroxine, phenytoin, the proton pump inhibitor pantoprazole, and prednisone. Recently, comorbid depression caused her to skip doses.

ER physicians order a UDS because of Ms. A’s mental status changes. The enzyme-linked immunosorbent (ELISA) toxicology test for alcohol, amphetamines, barbiturates, benzodiazepines, cocaine, opiates, marijuana, and phencyclidine (PCP) is positive for marijuana. When the attending psychiatrist informs Ms. A of the result, she is shocked. She tells the psychiatrist she is active in church and opposes recreational use of narcotics. She adamantly denies using marijuana or other street drugs, alcohol, nicotine, or caffeine.

Eventually, physicians attributed Ms. A’s mental status changes to several underlying medical issues, including Addison’s disease. A thorough review of the case revealed that the proton pump inhibitor pantoprazole caused the false-positive UDS.

UDS tests are sensitive but not highly specific. A medication or other substance with a chemical structure similar to that of the suspected drug can cause a false positive.^1-3

The “WEED” mnemonic spells out steps for critically evaluating UDS test results to ensure appropriate care:

• Write out a list of the patient’s medications. This list may explain the symptoms or help interpret UDS results. If a narcotic dose was recently increased, for example, a UDS might not be needed to confirm what caused the change in mental status.
  
• Examine the patient carefully. Evaluate physical signs, take a thorough medical history, and consider the potential for drug use. Although not impossible, for example, PCP intoxication is not a likely cause of psychosis in nursing home patients.
  
• Equate UDS results with presenting complaints and symptoms. For example, if a patient with sudden syncope tests positive for marijuana, the syncopal symptoms demand further investigation because marijuana is not the likely cause.
  
• Duplicate the UDS screen with confirmatory tests if the result will determine treatment. When UDS results are ambiguous, use highly specific tests such as gas chromatography with mass spectrometry and high-performance liquid chromatography. Although expensive and time consuming, these tests confirm the presence or absence of substances with few false results.
  

Pegylated interferon-alpha with ribavirin is the most effective therapy for chronic hepatitis C infection,^1,2 but psychiatric patients often discontinue IFN-alpha because of its mood side effects.³ Most studies describe depressive states, although manic symptoms—irritability, aggression, anger, hostility, emotional lability, anxiety, panic attacks, and insomnia—also have been reported.^4-6

To help you manage adverse mood changes and prevent IFN-alpha treatment discontinuation in patients with hepatitis C, this article:

• reviews studies of patients with a history mood disorders who were treated with IFN-alpha
• explains how to recognize and treat IFN-alpha-induced mood disturbances when antidepressants are contraindicated.

Psychiatric Patients and IFN Therapy

Chronic hepatitis C infection is common among psychiatric patients. Routine screening among 1,556 patients admitted to a U.S. public psychiatric hospital across 3 years identified 133 patients (8.5%) who were positive for hepatitis C virus.⁷

Patients with psychopathologic symptoms before starting IFN therapy may suffer more-severe adverse psychiatric effects during treatment than those without psychopathology.⁸ In fact, mood disorders were considered an absolute contraindication to IFN therapy until recently.

Now that the National Institutes of Health (NIH) has recommended extending hepatitis C research and treatment to psychiatric patients, IFN-alpha-induced mental illness could become more common in clinical practice. Before the 2002 NIH consensus statement, patients with mental illness and substance use disorders—who represent >50% of candidates who need IFN therapy—were excluded from research protocols.

Safely using IFN. Some reports and studies suggest that patients with past or existing psychiatric disorders can be treated safely and effectively with IFN-alpha.

In a prospective open-label study, 29 of 31 patients with co-existing chronic hepatitis C and psychiatric illness completed 6 months of IFN therapy, 5 million units (MU) three times/week or 5 MU daily. Patients continued maintenance psychotropics during IFN treatment, and a psychiatrist monitored psychiatric symptoms.

Psychiatric illness worsened in four patients, and two discontinued IFN treatment. Serum alanine aminotransferase returned to normal in 22 patients (71%), and hepatitis C virus RNA cleared from the sera of 15 (48%).⁹

Another prospective study of 50 patients treated with IFN for chronic hepatitis found that those with a pre-existing mood or anxiety disorder were not more likely than others to discontinue IFN therapy.¹⁰

Unique to hepatitis therapy? IFN-induced mood disturbances are probably different in patients with chronic hepatitis C than in those receiving IFN for other diseases because of differences in regimens and effects of the underlying pathologies. For example, prescribing a preventative antidepressant before starting IFN treatment might help cancer patients but not patients with hepatitis C.¹¹

This distinction could be particularly important when giving IFN-alpha to patients who are vulnerable to psychiatric illness with impulsive features. To emphasize this point, we describe clinical features and treatment response in patients with hepatitis C who were treated in our department.

IFN-Alpha-Induced Moods

In our prospective study of 93 patients, 30 (32%) developed IFN-alpha-induced mood disorders during the first 12 weeks of treatment.¹² Contrary to previous studies focusing on depression, most of our patients had a mix of manic/hypomanic and depressive symptoms. Twenty-four (13 women and 11 men, mean age 43) accepted referral to a psychiatrist specializing in mood disorders to characterize their symptoms.

Mood characteristics. Using DSM-IV criteria, the psychiatrist determined that IFN-alpha induced both manic/hypomanic and depressive symptoms in many patients, and the manic/hypomanic features predominated. Five manic symptoms and three depressive symptoms were present in >50% of patients (Figure 1, Table 1).

Three patients presented with a manic episode,15 with hypomania with prevalent irritability and dysphoric features, and 6 with a mixed depressive state (major depressive episode with at least 3 hypomania symptoms).¹³ Four patients (17%) suffered a relapse of alcohol or cannabis abuse.

Nearly all (84%) reported paroxysmal anxiety, and all had emotional hyperreactivity that the psychiatrist described as a main symptom of a manic or mixed state.¹⁴ Most felt extremely impulsive and feared losing control. Two faced legal difficulties, and one was in prison.

The patients’ mean Montgomery-Åsburg Depression Rating Scale (MADRS) score was 16.12 (±1.6), indicating a mild to moderate depressive state. The highest-scoring items were inner tension, reduced sleep, reduced appetite, and lassitude (Figure 2).

Their mean Bech-Rafaelsen Mania Scale¹⁵ score was 14.33 (±1.3), indicating hypomanic or moderate manic symptoms. Hostility was by far the predominant manic symptom (mean score 3.2/4). Other manic symptom scores ranged from 0.2/4 to 2.1/4 (Figure 3).

Figure 1 Mood symptoms identified in 24 patients after 12 weeks of IFN-alpha treatment

IFN: Interferon

Source: Reference 12

Figure 2 Depressive symptoms in 24 patients with IFN-induced mood disorder

IFN: Interferon

Source: Reference 12

Figure 3 Manic symptoms in 24 patients with IFN-induced mood disorder

IFN: Interferon

Source: Reference 12Table 1

Most-prevalent IFN-induced mood symptoms in 24 patients treated for hepatitis C

Treatment. Given the predominance of manic/hypomanic symptoms, we treated the 24 patients with low-to-moderate dosages of an atypical antipsychotic (amisulpride, 100 to 600 mg/d). This medication—not available in the United States—would be similar to using risperidone, 1 to 6 mg/d. Low-dose benzodiazepines (clonazepam or alprazolam) were added as needed to manage insomnia. Antipsychotic treatment enabled 23 of 24 patients (96%) to continue antiviral therapy.

Five patients had received selective serotonin reuptake inhibitors (SSRIs) for 1 to 4 weeks before referral to the psychiatrist. Antidepressants worsened their mood symptoms, which included impulsivity, agitation, and insomnia. Emotional hyper-reactivity, irritability, hostility, and impulsiveness improved in all 5 patients within 1 to 2 weeks of stopping SSRIs and starting the atypical antipsychotic.

Discussion

Accurately characterizing IFN-induced mood states confirmed our published data showing a mix of manic/hypomanic and depressive symptoms in psychiatric patients treated for chronic hepatitis C. Irritability and hostility were the two most prominent symptoms.

These findings differ from those of investigations that identified depressive symptoms as the hallmark of IFN’s psychiatric side effects.⁶ Our data were thoroughly characterized according to DSM-IV-TR criteria, two mood-rating scales, and a psychiatrist experienced in mood disorders.

Why is mania missed? One possible explanation for these different findings is that IFN-alpha-induced fatigue and flu-like symptoms could be misinterpreted as depressive symptoms. Also, most researchers have used depression rating scales—but not mania scales—and have not considered other psychiatric diagnostics.¹⁶

Self-report questionnaires for evaluating depression also take into account somatic side effects, resulting in higher rating scores. Moreover, few studies have included clinical psychiatric interviews and even fewer diagnostic confirmation by a psychiatrist.

Finally, clinical experience indicates that patients who present with both manic/hypomanic and depressive symptoms are more likely to complain about depression than about manic symptoms. Therefore, clinicians may miss manic symptoms if they don’t actively seek them.

Irritability and hostility. Previous studies have described irritability, mood lability, and anger/hostility as frequent symptoms^4,5 but failed to consider them as possible manifestations of mania or hypomania. Many of our patients reported irritability severe enough to interfere with their work, social, and family relationships.

Hostility was the symptom with the highest score on the Bech-Rafaelsen Mania Scale. This objective evidence suggests that investigating the consequences to patients of increased irritability, impulsivity, or hostility might reveal some frightening behavior

How to treat these patients. Considerable evidence from characterizing IFN-induced side effects in patients with hepatitis C points to a syndrome of depressive and manic/hypomanic symptoms^13,17-19 that does not fulfill criteria for a mixed state. This disorder is not described in DSM-IV-TR and, unfortunately, usually is misdiagnosed as a depressive state.

Antidepressants can worsen depression by causing agitation and impulsivity and can increase risk of suicide.^17,18,20 By contrast, atypical antipsychotics have been shown to improve bipolar depression.^21,22

We used an atypical antipsychotic to treat patients with prevalent manic or hypomanic symptoms and those who did not respond to SSRIs. Their IFN-induced mood disorders improved rapidly on low dosages of amisulpride, and antiviral therapy discontinuation rates were low (1/24; 4%). By comparison, other studies have reported antiviral treatment discontinuation rates of 30% to 40% in patients taking antidepressants.^23,24

The atypical antipsychotic did not improve our patients’ fatigue and other neurovegetative symptoms, but antidepressants likewise do not improve these symptoms.⁸

Recommendations

This study leads us to warn clinicians that antidepressants can worsen IFN-alpha-induced mood disorders and to recommend that antipsychotics be considered:

• at least before you decide to discontinue a hepatitis C patient’s IFN-alpha therapy
• or in patients with high irritability and hostility.

To determine whether to use an antidepressant or antimanic agent as first-line treatment, carefully diagnose the patient’s symptoms as a manic/hypomanic state, depressive mixed state, or depression. Successful IFN therapy requires:

• psychological support
• medication for psychiatric adverse effects
• collaboration between the psychiatrist and hepatologist.

IFN-alpha-induced mood disorders are triggered exogenously, do not correspond to typical features of any classic psychiatric disorder, and require further study. IFN-induced impulsivity and hostility also need to be better characterized, particularly as more patients with pre-existing psychiatric disorders are treated for hepatitis C.

Related resources

• Lauer GM, Walker BD. Hepatitis C virus infection. N Engl J Med 2001;345(1):41-52.
• Onyike CU, Bonner JO, Lyketsos CG, Treisman GJ. Mania during treatment of chronic hepatitis C with pegylated interferon and ribavirin. Am J Psychiatry 2004;161:3.

Drug brand names

• Alprazolam • Xanax
• Amisulpride • (not available in the United States)
• Clonazepam • Klonopin
• Risperidone • Risperdal

Acknowledgment

This study was conducted with support from the Centre d’Investigation Clinique INSERM/CHU de Bordeaux.

Pegylated interferon-alpha with ribavirin is the most effective therapy for chronic hepatitis C infection,^1,2 but psychiatric patients often discontinue IFN-alpha because of its mood side effects.³ Most studies describe depressive states, although manic symptoms—irritability, aggression, anger, hostility, emotional lability, anxiety, panic attacks, and insomnia—also have been reported.^4-6

To help you manage adverse mood changes and prevent IFN-alpha treatment discontinuation in patients with hepatitis C, this article:

• reviews studies of patients with a history mood disorders who were treated with IFN-alpha
• explains how to recognize and treat IFN-alpha-induced mood disturbances when antidepressants are contraindicated.

Psychiatric Patients and IFN Therapy

Chronic hepatitis C infection is common among psychiatric patients. Routine screening among 1,556 patients admitted to a U.S. public psychiatric hospital across 3 years identified 133 patients (8.5%) who were positive for hepatitis C virus.⁷

Patients with psychopathologic symptoms before starting IFN therapy may suffer more-severe adverse psychiatric effects during treatment than those without psychopathology.⁸ In fact, mood disorders were considered an absolute contraindication to IFN therapy until recently.

Now that the National Institutes of Health (NIH) has recommended extending hepatitis C research and treatment to psychiatric patients, IFN-alpha-induced mental illness could become more common in clinical practice. Before the 2002 NIH consensus statement, patients with mental illness and substance use disorders—who represent >50% of candidates who need IFN therapy—were excluded from research protocols.

Safely using IFN. Some reports and studies suggest that patients with past or existing psychiatric disorders can be treated safely and effectively with IFN-alpha.

In a prospective open-label study, 29 of 31 patients with co-existing chronic hepatitis C and psychiatric illness completed 6 months of IFN therapy, 5 million units (MU) three times/week or 5 MU daily. Patients continued maintenance psychotropics during IFN treatment, and a psychiatrist monitored psychiatric symptoms.

Psychiatric illness worsened in four patients, and two discontinued IFN treatment. Serum alanine aminotransferase returned to normal in 22 patients (71%), and hepatitis C virus RNA cleared from the sera of 15 (48%).⁹

Another prospective study of 50 patients treated with IFN for chronic hepatitis found that those with a pre-existing mood or anxiety disorder were not more likely than others to discontinue IFN therapy.¹⁰

Unique to hepatitis therapy? IFN-induced mood disturbances are probably different in patients with chronic hepatitis C than in those receiving IFN for other diseases because of differences in regimens and effects of the underlying pathologies. For example, prescribing a preventative antidepressant before starting IFN treatment might help cancer patients but not patients with hepatitis C.¹¹

This distinction could be particularly important when giving IFN-alpha to patients who are vulnerable to psychiatric illness with impulsive features. To emphasize this point, we describe clinical features and treatment response in patients with hepatitis C who were treated in our department.

IFN-Alpha-Induced Moods

In our prospective study of 93 patients, 30 (32%) developed IFN-alpha-induced mood disorders during the first 12 weeks of treatment.¹² Contrary to previous studies focusing on depression, most of our patients had a mix of manic/hypomanic and depressive symptoms. Twenty-four (13 women and 11 men, mean age 43) accepted referral to a psychiatrist specializing in mood disorders to characterize their symptoms.

Mood characteristics. Using DSM-IV criteria, the psychiatrist determined that IFN-alpha induced both manic/hypomanic and depressive symptoms in many patients, and the manic/hypomanic features predominated. Five manic symptoms and three depressive symptoms were present in >50% of patients (Figure 1, Table 1).

Three patients presented with a manic episode,15 with hypomania with prevalent irritability and dysphoric features, and 6 with a mixed depressive state (major depressive episode with at least 3 hypomania symptoms).¹³ Four patients (17%) suffered a relapse of alcohol or cannabis abuse.

Nearly all (84%) reported paroxysmal anxiety, and all had emotional hyperreactivity that the psychiatrist described as a main symptom of a manic or mixed state.¹⁴ Most felt extremely impulsive and feared losing control. Two faced legal difficulties, and one was in prison.

The patients’ mean Montgomery-Åsburg Depression Rating Scale (MADRS) score was 16.12 (±1.6), indicating a mild to moderate depressive state. The highest-scoring items were inner tension, reduced sleep, reduced appetite, and lassitude (Figure 2).

Their mean Bech-Rafaelsen Mania Scale¹⁵ score was 14.33 (±1.3), indicating hypomanic or moderate manic symptoms. Hostility was by far the predominant manic symptom (mean score 3.2/4). Other manic symptom scores ranged from 0.2/4 to 2.1/4 (Figure 3).

Figure 1 Mood symptoms identified in 24 patients after 12 weeks of IFN-alpha treatment

IFN: Interferon

Source: Reference 12

Figure 2 Depressive symptoms in 24 patients with IFN-induced mood disorder

IFN: Interferon

Source: Reference 12

Figure 3 Manic symptoms in 24 patients with IFN-induced mood disorder

IFN: Interferon

Source: Reference 12Table 1

Most-prevalent IFN-induced mood symptoms in 24 patients treated for hepatitis C

Treatment. Given the predominance of manic/hypomanic symptoms, we treated the 24 patients with low-to-moderate dosages of an atypical antipsychotic (amisulpride, 100 to 600 mg/d). This medication—not available in the United States—would be similar to using risperidone, 1 to 6 mg/d. Low-dose benzodiazepines (clonazepam or alprazolam) were added as needed to manage insomnia. Antipsychotic treatment enabled 23 of 24 patients (96%) to continue antiviral therapy.

Five patients had received selective serotonin reuptake inhibitors (SSRIs) for 1 to 4 weeks before referral to the psychiatrist. Antidepressants worsened their mood symptoms, which included impulsivity, agitation, and insomnia. Emotional hyper-reactivity, irritability, hostility, and impulsiveness improved in all 5 patients within 1 to 2 weeks of stopping SSRIs and starting the atypical antipsychotic.

Discussion

Accurately characterizing IFN-induced mood states confirmed our published data showing a mix of manic/hypomanic and depressive symptoms in psychiatric patients treated for chronic hepatitis C. Irritability and hostility were the two most prominent symptoms.

These findings differ from those of investigations that identified depressive symptoms as the hallmark of IFN’s psychiatric side effects.⁶ Our data were thoroughly characterized according to DSM-IV-TR criteria, two mood-rating scales, and a psychiatrist experienced in mood disorders.

Why is mania missed? One possible explanation for these different findings is that IFN-alpha-induced fatigue and flu-like symptoms could be misinterpreted as depressive symptoms. Also, most researchers have used depression rating scales—but not mania scales—and have not considered other psychiatric diagnostics.¹⁶

Self-report questionnaires for evaluating depression also take into account somatic side effects, resulting in higher rating scores. Moreover, few studies have included clinical psychiatric interviews and even fewer diagnostic confirmation by a psychiatrist.

Finally, clinical experience indicates that patients who present with both manic/hypomanic and depressive symptoms are more likely to complain about depression than about manic symptoms. Therefore, clinicians may miss manic symptoms if they don’t actively seek them.

Irritability and hostility. Previous studies have described irritability, mood lability, and anger/hostility as frequent symptoms^4,5 but failed to consider them as possible manifestations of mania or hypomania. Many of our patients reported irritability severe enough to interfere with their work, social, and family relationships.

Hostility was the symptom with the highest score on the Bech-Rafaelsen Mania Scale. This objective evidence suggests that investigating the consequences to patients of increased irritability, impulsivity, or hostility might reveal some frightening behavior

How to treat these patients. Considerable evidence from characterizing IFN-induced side effects in patients with hepatitis C points to a syndrome of depressive and manic/hypomanic symptoms^13,17-19 that does not fulfill criteria for a mixed state. This disorder is not described in DSM-IV-TR and, unfortunately, usually is misdiagnosed as a depressive state.

Antidepressants can worsen depression by causing agitation and impulsivity and can increase risk of suicide.^17,18,20 By contrast, atypical antipsychotics have been shown to improve bipolar depression.^21,22

We used an atypical antipsychotic to treat patients with prevalent manic or hypomanic symptoms and those who did not respond to SSRIs. Their IFN-induced mood disorders improved rapidly on low dosages of amisulpride, and antiviral therapy discontinuation rates were low (1/24; 4%). By comparison, other studies have reported antiviral treatment discontinuation rates of 30% to 40% in patients taking antidepressants.^23,24

The atypical antipsychotic did not improve our patients’ fatigue and other neurovegetative symptoms, but antidepressants likewise do not improve these symptoms.⁸

Recommendations

This study leads us to warn clinicians that antidepressants can worsen IFN-alpha-induced mood disorders and to recommend that antipsychotics be considered:

• at least before you decide to discontinue a hepatitis C patient’s IFN-alpha therapy
• or in patients with high irritability and hostility.

To determine whether to use an antidepressant or antimanic agent as first-line treatment, carefully diagnose the patient’s symptoms as a manic/hypomanic state, depressive mixed state, or depression. Successful IFN therapy requires:

• psychological support
• medication for psychiatric adverse effects
• collaboration between the psychiatrist and hepatologist.

IFN-alpha-induced mood disorders are triggered exogenously, do not correspond to typical features of any classic psychiatric disorder, and require further study. IFN-induced impulsivity and hostility also need to be better characterized, particularly as more patients with pre-existing psychiatric disorders are treated for hepatitis C.

Related resources

• Lauer GM, Walker BD. Hepatitis C virus infection. N Engl J Med 2001;345(1):41-52.
• Onyike CU, Bonner JO, Lyketsos CG, Treisman GJ. Mania during treatment of chronic hepatitis C with pegylated interferon and ribavirin. Am J Psychiatry 2004;161:3.

Drug brand names

• Alprazolam • Xanax
• Amisulpride • (not available in the United States)
• Clonazepam • Klonopin
• Risperidone • Risperdal

Acknowledgment

This study was conducted with support from the Centre d’Investigation Clinique INSERM/CHU de Bordeaux.

Pegylated interferon-alpha with ribavirin is the most effective therapy for chronic hepatitis C infection,^1,2 but psychiatric patients often discontinue IFN-alpha because of its mood side effects.³ Most studies describe depressive states, although manic symptoms—irritability, aggression, anger, hostility, emotional lability, anxiety, panic attacks, and insomnia—also have been reported.^4-6

To help you manage adverse mood changes and prevent IFN-alpha treatment discontinuation in patients with hepatitis C, this article:

• reviews studies of patients with a history mood disorders who were treated with IFN-alpha
• explains how to recognize and treat IFN-alpha-induced mood disturbances when antidepressants are contraindicated.

Psychiatric Patients and IFN Therapy

Chronic hepatitis C infection is common among psychiatric patients. Routine screening among 1,556 patients admitted to a U.S. public psychiatric hospital across 3 years identified 133 patients (8.5%) who were positive for hepatitis C virus.⁷

Patients with psychopathologic symptoms before starting IFN therapy may suffer more-severe adverse psychiatric effects during treatment than those without psychopathology.⁸ In fact, mood disorders were considered an absolute contraindication to IFN therapy until recently.

Now that the National Institutes of Health (NIH) has recommended extending hepatitis C research and treatment to psychiatric patients, IFN-alpha-induced mental illness could become more common in clinical practice. Before the 2002 NIH consensus statement, patients with mental illness and substance use disorders—who represent >50% of candidates who need IFN therapy—were excluded from research protocols.

Safely using IFN. Some reports and studies suggest that patients with past or existing psychiatric disorders can be treated safely and effectively with IFN-alpha.

In a prospective open-label study, 29 of 31 patients with co-existing chronic hepatitis C and psychiatric illness completed 6 months of IFN therapy, 5 million units (MU) three times/week or 5 MU daily. Patients continued maintenance psychotropics during IFN treatment, and a psychiatrist monitored psychiatric symptoms.

Psychiatric illness worsened in four patients, and two discontinued IFN treatment. Serum alanine aminotransferase returned to normal in 22 patients (71%), and hepatitis C virus RNA cleared from the sera of 15 (48%).⁹

Another prospective study of 50 patients treated with IFN for chronic hepatitis found that those with a pre-existing mood or anxiety disorder were not more likely than others to discontinue IFN therapy.¹⁰

Unique to hepatitis therapy? IFN-induced mood disturbances are probably different in patients with chronic hepatitis C than in those receiving IFN for other diseases because of differences in regimens and effects of the underlying pathologies. For example, prescribing a preventative antidepressant before starting IFN treatment might help cancer patients but not patients with hepatitis C.¹¹

This distinction could be particularly important when giving IFN-alpha to patients who are vulnerable to psychiatric illness with impulsive features. To emphasize this point, we describe clinical features and treatment response in patients with hepatitis C who were treated in our department.

IFN-Alpha-Induced Moods

In our prospective study of 93 patients, 30 (32%) developed IFN-alpha-induced mood disorders during the first 12 weeks of treatment.¹² Contrary to previous studies focusing on depression, most of our patients had a mix of manic/hypomanic and depressive symptoms. Twenty-four (13 women and 11 men, mean age 43) accepted referral to a psychiatrist specializing in mood disorders to characterize their symptoms.

Mood characteristics. Using DSM-IV criteria, the psychiatrist determined that IFN-alpha induced both manic/hypomanic and depressive symptoms in many patients, and the manic/hypomanic features predominated. Five manic symptoms and three depressive symptoms were present in >50% of patients (Figure 1, Table 1).

Three patients presented with a manic episode,15 with hypomania with prevalent irritability and dysphoric features, and 6 with a mixed depressive state (major depressive episode with at least 3 hypomania symptoms).¹³ Four patients (17%) suffered a relapse of alcohol or cannabis abuse.

Nearly all (84%) reported paroxysmal anxiety, and all had emotional hyperreactivity that the psychiatrist described as a main symptom of a manic or mixed state.¹⁴ Most felt extremely impulsive and feared losing control. Two faced legal difficulties, and one was in prison.

The patients’ mean Montgomery-Åsburg Depression Rating Scale (MADRS) score was 16.12 (±1.6), indicating a mild to moderate depressive state. The highest-scoring items were inner tension, reduced sleep, reduced appetite, and lassitude (Figure 2).

Their mean Bech-Rafaelsen Mania Scale¹⁵ score was 14.33 (±1.3), indicating hypomanic or moderate manic symptoms. Hostility was by far the predominant manic symptom (mean score 3.2/4). Other manic symptom scores ranged from 0.2/4 to 2.1/4 (Figure 3).

Figure 1 Mood symptoms identified in 24 patients after 12 weeks of IFN-alpha treatment

IFN: Interferon

Source: Reference 12

Figure 2 Depressive symptoms in 24 patients with IFN-induced mood disorder

IFN: Interferon

Source: Reference 12

Figure 3 Manic symptoms in 24 patients with IFN-induced mood disorder

IFN: Interferon

Source: Reference 12Table 1

Most-prevalent IFN-induced mood symptoms in 24 patients treated for hepatitis C

Treatment. Given the predominance of manic/hypomanic symptoms, we treated the 24 patients with low-to-moderate dosages of an atypical antipsychotic (amisulpride, 100 to 600 mg/d). This medication—not available in the United States—would be similar to using risperidone, 1 to 6 mg/d. Low-dose benzodiazepines (clonazepam or alprazolam) were added as needed to manage insomnia. Antipsychotic treatment enabled 23 of 24 patients (96%) to continue antiviral therapy.

Five patients had received selective serotonin reuptake inhibitors (SSRIs) for 1 to 4 weeks before referral to the psychiatrist. Antidepressants worsened their mood symptoms, which included impulsivity, agitation, and insomnia. Emotional hyper-reactivity, irritability, hostility, and impulsiveness improved in all 5 patients within 1 to 2 weeks of stopping SSRIs and starting the atypical antipsychotic.

Discussion

Accurately characterizing IFN-induced mood states confirmed our published data showing a mix of manic/hypomanic and depressive symptoms in psychiatric patients treated for chronic hepatitis C. Irritability and hostility were the two most prominent symptoms.

These findings differ from those of investigations that identified depressive symptoms as the hallmark of IFN’s psychiatric side effects.⁶ Our data were thoroughly characterized according to DSM-IV-TR criteria, two mood-rating scales, and a psychiatrist experienced in mood disorders.

Why is mania missed? One possible explanation for these different findings is that IFN-alpha-induced fatigue and flu-like symptoms could be misinterpreted as depressive symptoms. Also, most researchers have used depression rating scales—but not mania scales—and have not considered other psychiatric diagnostics.¹⁶

Self-report questionnaires for evaluating depression also take into account somatic side effects, resulting in higher rating scores. Moreover, few studies have included clinical psychiatric interviews and even fewer diagnostic confirmation by a psychiatrist.

Finally, clinical experience indicates that patients who present with both manic/hypomanic and depressive symptoms are more likely to complain about depression than about manic symptoms. Therefore, clinicians may miss manic symptoms if they don’t actively seek them.

Irritability and hostility. Previous studies have described irritability, mood lability, and anger/hostility as frequent symptoms^4,5 but failed to consider them as possible manifestations of mania or hypomania. Many of our patients reported irritability severe enough to interfere with their work, social, and family relationships.

Hostility was the symptom with the highest score on the Bech-Rafaelsen Mania Scale. This objective evidence suggests that investigating the consequences to patients of increased irritability, impulsivity, or hostility might reveal some frightening behavior

How to treat these patients. Considerable evidence from characterizing IFN-induced side effects in patients with hepatitis C points to a syndrome of depressive and manic/hypomanic symptoms^13,17-19 that does not fulfill criteria for a mixed state. This disorder is not described in DSM-IV-TR and, unfortunately, usually is misdiagnosed as a depressive state.

Antidepressants can worsen depression by causing agitation and impulsivity and can increase risk of suicide.^17,18,20 By contrast, atypical antipsychotics have been shown to improve bipolar depression.^21,22

We used an atypical antipsychotic to treat patients with prevalent manic or hypomanic symptoms and those who did not respond to SSRIs. Their IFN-induced mood disorders improved rapidly on low dosages of amisulpride, and antiviral therapy discontinuation rates were low (1/24; 4%). By comparison, other studies have reported antiviral treatment discontinuation rates of 30% to 40% in patients taking antidepressants.^23,24

The atypical antipsychotic did not improve our patients’ fatigue and other neurovegetative symptoms, but antidepressants likewise do not improve these symptoms.⁸

Recommendations

This study leads us to warn clinicians that antidepressants can worsen IFN-alpha-induced mood disorders and to recommend that antipsychotics be considered:

• at least before you decide to discontinue a hepatitis C patient’s IFN-alpha therapy
• or in patients with high irritability and hostility.

To determine whether to use an antidepressant or antimanic agent as first-line treatment, carefully diagnose the patient’s symptoms as a manic/hypomanic state, depressive mixed state, or depression. Successful IFN therapy requires:

• psychological support
• medication for psychiatric adverse effects
• collaboration between the psychiatrist and hepatologist.

IFN-alpha-induced mood disorders are triggered exogenously, do not correspond to typical features of any classic psychiatric disorder, and require further study. IFN-induced impulsivity and hostility also need to be better characterized, particularly as more patients with pre-existing psychiatric disorders are treated for hepatitis C.

Related resources

• Lauer GM, Walker BD. Hepatitis C virus infection. N Engl J Med 2001;345(1):41-52.
• Onyike CU, Bonner JO, Lyketsos CG, Treisman GJ. Mania during treatment of chronic hepatitis C with pegylated interferon and ribavirin. Am J Psychiatry 2004;161:3.

Drug brand names

• Alprazolam • Xanax
• Amisulpride • (not available in the United States)
• Clonazepam • Klonopin
• Risperidone • Risperdal

Acknowledgment

This study was conducted with support from the Centre d’Investigation Clinique INSERM/CHU de Bordeaux.

The article “Female sexual dysfunction: don’t assume it’s a side effect” (Med/Psych Update, July 2006,) contained an error on pages 56-7. It should have stated that elevated thyroid-stimulating hormone (TSH) could point to hypothyroidism and low TSH could signal hyperthyroidism.

The article “Female sexual dysfunction: don’t assume it’s a side effect” (Med/Psych Update, July 2006,) contained an error on pages 56-7. It should have stated that elevated thyroid-stimulating hormone (TSH) could point to hypothyroidism and low TSH could signal hyperthyroidism.

The article “Female sexual dysfunction: don’t assume it’s a side effect” (Med/Psych Update, July 2006,) contained an error on pages 56-7. It should have stated that elevated thyroid-stimulating hormone (TSH) could point to hypothyroidism and low TSH could signal hyperthyroidism.

In their article on managing dementia-related behaviors (Current Psychiatry, May 2006), Dr. Bruce Sutor and colleagues advise against using opioids to manage pain in older patients (p. 88). I was disheartened by their comments.

American Geriatrics Society guidelines clearly state that opiate medications are an appropriate choice for treating moderate to severe pain in some elderly persons, including those with dementia.¹

Refusing to consider opiates for pain management can lead to severe suffering, especially for a patient who is incapable of communication. If an elder is at risk for falls, perhaps a full pain assessment is warranted and other CNS suppressants such as benzodiazepines can be eliminated. That way, we can treat pain and reduce the risk of falls.

Unfortunately, the myth that certain groups are not appropriate candidates for opiates lives on. We need to assess and treat all elders for pain, whether they are cognitively intact or impaired. As health care professionals, we are obligated to reduce pain and suffering in all patients, especially those who have trouble communicating.

Teresa Keane, RN, MSN, PMHNP
Psychiatric nurse practitioner, Kaiser Permanente Northwest
Adjunct faculty member, Oregon Health & Science University
Portland, OR

Dr. Sutor responds

Ms. Keane’s letter raises important points regarding managing pain in dementia patients.

We agree wholeheartedly that a full pain assessment is warranted when pain is suspected. All patients need and deserve to have their pain managed. There is, however, a danger that a dementia patient with behavior problems may be presumed to have pain when he or she does not and may be given narcotics reflexively.

Ms. Keane’s contention that CNS suppressants (such as benzodiazepines) can be eliminated is germane to our point about opiate analgesics and supports our argument that opiates should be used warily, if at all. Eliminating and avoiding opiates, when possible, reduces the risks of falls, fractures, agitation, delirium, and diminished cognitive function.

Our article should have stated clearly that opiates should be avoided if the cause of pain can be ameliorated or eliminated, or if other “rungs” on the pain ladder effectively manage pain.

It is no myth that some dementia patients are not appropriate candidates for opiates. Those who have suffered falls, fractures, delirium, confusion, and CNS obtundation support the contention that opiates can do more harm than good for some dementia patients.

Bruce Sutor, MD
Assistant professor of psychiatry
Department of psychiatry and psychology
Mayo Clinic College of Medicine
Rochester, MN

In their article on managing dementia-related behaviors (Current Psychiatry, May 2006), Dr. Bruce Sutor and colleagues advise against using opioids to manage pain in older patients (p. 88). I was disheartened by their comments.

American Geriatrics Society guidelines clearly state that opiate medications are an appropriate choice for treating moderate to severe pain in some elderly persons, including those with dementia.¹

Refusing to consider opiates for pain management can lead to severe suffering, especially for a patient who is incapable of communication. If an elder is at risk for falls, perhaps a full pain assessment is warranted and other CNS suppressants such as benzodiazepines can be eliminated. That way, we can treat pain and reduce the risk of falls.

Unfortunately, the myth that certain groups are not appropriate candidates for opiates lives on. We need to assess and treat all elders for pain, whether they are cognitively intact or impaired. As health care professionals, we are obligated to reduce pain and suffering in all patients, especially those who have trouble communicating.

Teresa Keane, RN, MSN, PMHNP
Psychiatric nurse practitioner, Kaiser Permanente Northwest
Adjunct faculty member, Oregon Health & Science University
Portland, OR

Dr. Sutor responds

Ms. Keane’s letter raises important points regarding managing pain in dementia patients.

We agree wholeheartedly that a full pain assessment is warranted when pain is suspected. All patients need and deserve to have their pain managed. There is, however, a danger that a dementia patient with behavior problems may be presumed to have pain when he or she does not and may be given narcotics reflexively.

Ms. Keane’s contention that CNS suppressants (such as benzodiazepines) can be eliminated is germane to our point about opiate analgesics and supports our argument that opiates should be used warily, if at all. Eliminating and avoiding opiates, when possible, reduces the risks of falls, fractures, agitation, delirium, and diminished cognitive function.

Our article should have stated clearly that opiates should be avoided if the cause of pain can be ameliorated or eliminated, or if other “rungs” on the pain ladder effectively manage pain.

It is no myth that some dementia patients are not appropriate candidates for opiates. Those who have suffered falls, fractures, delirium, confusion, and CNS obtundation support the contention that opiates can do more harm than good for some dementia patients.

Bruce Sutor, MD
Assistant professor of psychiatry
Department of psychiatry and psychology
Mayo Clinic College of Medicine
Rochester, MN

In their article on managing dementia-related behaviors (Current Psychiatry, May 2006), Dr. Bruce Sutor and colleagues advise against using opioids to manage pain in older patients (p. 88). I was disheartened by their comments.

American Geriatrics Society guidelines clearly state that opiate medications are an appropriate choice for treating moderate to severe pain in some elderly persons, including those with dementia.¹

Refusing to consider opiates for pain management can lead to severe suffering, especially for a patient who is incapable of communication. If an elder is at risk for falls, perhaps a full pain assessment is warranted and other CNS suppressants such as benzodiazepines can be eliminated. That way, we can treat pain and reduce the risk of falls.

Unfortunately, the myth that certain groups are not appropriate candidates for opiates lives on. We need to assess and treat all elders for pain, whether they are cognitively intact or impaired. As health care professionals, we are obligated to reduce pain and suffering in all patients, especially those who have trouble communicating.

Teresa Keane, RN, MSN, PMHNP
Psychiatric nurse practitioner, Kaiser Permanente Northwest
Adjunct faculty member, Oregon Health & Science University
Portland, OR

Dr. Sutor responds

Ms. Keane’s letter raises important points regarding managing pain in dementia patients.

We agree wholeheartedly that a full pain assessment is warranted when pain is suspected. All patients need and deserve to have their pain managed. There is, however, a danger that a dementia patient with behavior problems may be presumed to have pain when he or she does not and may be given narcotics reflexively.

Ms. Keane’s contention that CNS suppressants (such as benzodiazepines) can be eliminated is germane to our point about opiate analgesics and supports our argument that opiates should be used warily, if at all. Eliminating and avoiding opiates, when possible, reduces the risks of falls, fractures, agitation, delirium, and diminished cognitive function.

Our article should have stated clearly that opiates should be avoided if the cause of pain can be ameliorated or eliminated, or if other “rungs” on the pain ladder effectively manage pain.

It is no myth that some dementia patients are not appropriate candidates for opiates. Those who have suffered falls, fractures, delirium, confusion, and CNS obtundation support the contention that opiates can do more harm than good for some dementia patients.

Bruce Sutor, MD
Assistant professor of psychiatry
Department of psychiatry and psychology
Mayo Clinic College of Medicine
Rochester, MN

Valedictory: a statement of farewell or leave taking. From pp. stem of L. valedicere “bid farewell,” from vale, imperative of valere “be well” + dicere “to say”

Dear Readers,

I am graduating from my position as editor-in-chief of Current Psychiatry. Four years is enough for high school, college, medical school, or residency training (at least in psychiatry). It is also the right amount of time to have been editor of this journal (although it has taken me closer to 5 years to finish).

Since January 2002, Current Psychiatry has become one of the most widely read journals in its field. It is the only journal I read cover to cover, and I know that is the case for many of you as well. Wherever I go, I hear extremely positive comments from psychiatrists and psychiatric nurse practitioners. I like to think that Current Psychiatry—more than other journals—has put our readers at the center and has always tried to give you “news you can use this week” in your clinical practices.

I am very happy that our editor, Alice Luddington, and the rest of the excellent Quadrant HealthCom Inc. staff will continue to work with you on this journal. I am also happy that my good friend, Henry A. Nasrallah, MD, is assuming the role of editor-in-chief as of next month’s issue.

I plan to work on several exciting new projects this summer. I also plan to read Current Psychiatry for a long time to come. Be well!

Valedictory: a statement of farewell or leave taking. From pp. stem of L. valedicere “bid farewell,” from vale, imperative of valere “be well” + dicere “to say”

Dear Readers,

I am graduating from my position as editor-in-chief of Current Psychiatry. Four years is enough for high school, college, medical school, or residency training (at least in psychiatry). It is also the right amount of time to have been editor of this journal (although it has taken me closer to 5 years to finish).

Since January 2002, Current Psychiatry has become one of the most widely read journals in its field. It is the only journal I read cover to cover, and I know that is the case for many of you as well. Wherever I go, I hear extremely positive comments from psychiatrists and psychiatric nurse practitioners. I like to think that Current Psychiatry—more than other journals—has put our readers at the center and has always tried to give you “news you can use this week” in your clinical practices.

I am very happy that our editor, Alice Luddington, and the rest of the excellent Quadrant HealthCom Inc. staff will continue to work with you on this journal. I am also happy that my good friend, Henry A. Nasrallah, MD, is assuming the role of editor-in-chief as of next month’s issue.

I plan to work on several exciting new projects this summer. I also plan to read Current Psychiatry for a long time to come. Be well!

Valedictory: a statement of farewell or leave taking. From pp. stem of L. valedicere “bid farewell,” from vale, imperative of valere “be well” + dicere “to say”

Dear Readers,

I am graduating from my position as editor-in-chief of Current Psychiatry. Four years is enough for high school, college, medical school, or residency training (at least in psychiatry). It is also the right amount of time to have been editor of this journal (although it has taken me closer to 5 years to finish).

Since January 2002, Current Psychiatry has become one of the most widely read journals in its field. It is the only journal I read cover to cover, and I know that is the case for many of you as well. Wherever I go, I hear extremely positive comments from psychiatrists and psychiatric nurse practitioners. I like to think that Current Psychiatry—more than other journals—has put our readers at the center and has always tried to give you “news you can use this week” in your clinical practices.

I am very happy that our editor, Alice Luddington, and the rest of the excellent Quadrant HealthCom Inc. staff will continue to work with you on this journal. I am also happy that my good friend, Henry A. Nasrallah, MD, is assuming the role of editor-in-chief as of next month’s issue.