Richard Balon, MD

Ms. P, a 63-year-old woman with a history of schizophrenia whose symptoms have been stable on haloperidol 10 mg/d and ziprasidone 40 mg twice daily, presents to the outpatient clinic for a medication review. She mentions that she has noticed problems with her “memory.” She says she has had difficulty remembering names of people and places as well as difficulty concentrating while reading and writing, which she did months ago with ease. A Montreal Cognitive Assessment (MoCA) is conducted, and Ms. P scores 13/30, indicating moderate cognitive impairment. Visuospatial tasks and clock drawing are intact, but she exhibits impairments in working memory, attention, and concentration. One year ago, Ms. P’s MoCA score was 27/30. She agrees to a neurologic assessment and is referred to neurology for work-up.

Ms. P’s physical examination and routine laboratory tests are all within normal limits. The neurologic exam reveals deficits in working memory, concentration, and attention, but is otherwise unremarkable. MRI reveals mild chronic microvascular changes. The neurology service does not rule out cognitive impairment but recommends adjusting the dosage of Ms. P’s psychiatric medications to elucidate if her impairment of memory and attention is due to medications. However, Ms. P had been managed on her current regimen for several years and had not been hospitalized in many years. Previous attempts to taper her antipsychotics had resulted in worsening symptoms. Ms. P is reluctant to attempt a taper of her antipsychotics because she fears decompensation of her chronic illness. The treating team reviews Ms. P’s medication regimen, and notes that she is receiving benztropine 1 mg twice daily for prophylaxis of extrapyramidal symptoms (EPS). Ms. P denies past or present symptoms of drug-induced parkinsonism, dystonia, or akathisia as well as constipation, sialorrhea, blurry vision, palpitations, or urinary retention.

Benztropine is a tropane alkaloid that was synthetized by combining the tropine portion of atropine with the benzhydryl portion of diphenhydramine hydrochloride. It has anticholinergic and antihistaminic properties¹ and seems to inhibit the dopamine transporter. Benztropine is indicated for all forms of parkinsonism, including antipsychotic-induced parkinsonism, but is also prescribed for many off-label uses, including sialorrhea and akathisia (although many authors do not recommend anticholinergics for this purpose^2,3), and for prophylaxis of EPS. Benztropine can be administered intravenously, intramuscularly, or orally. Given orally, the typical dosing is twice daily with a maximum dose of 6 mg/d. Benztropine is preferred over diphenhydramine and trihexyphenidyl due to adverse effects of sedation or potential for misuse of the medication.¹

Second-generation antipsychotics (SGAs) have been associated with lower rates of neurologic adverse effects compared with first-generation antipsychotics (FGAs). Because SGAs are increasingly prescribed, the use of benztropine (along with other agents such as trihexyphenidyl) for EPS prophylaxis is not an evidence-based practice. However, despite a movement away from prophylactic management of movement disorders, benztropine continues to be prescribed for EPS and/or cholinergic symptoms, despite the peripheral and cognitive adverse effects of this agent and, in many instances, the lack of clear indication for its use.

According to the most recent edition of the American Psychiatric Association’s (APA) Practice Guideline for the Treatment of Patients with Schizophrenia,⁴ anticholinergics should only be used for preventing acute dystonia in conjunction with a long-acting injectable antipsychotic. Furthermore, the APA Guideline states anticholinergics may be used for drug-induced parkinsonism when the dose of an antipsychotic cannot be reduced and an alternative agent is required. However, the first-line agent for drug-induced parkinsonism is amantadine, and benztropine should only be considered if amantadine is contraindicated.⁴ The rationale for this guideline and for judicious use of anticholinergics is that like any pharmacologic treatment, anticholinergics (including benztropine) carry the potential for adverse effects. For benztropine, these range from mild effects such as tachycardia and constipation to paralytic ileus, increased falls, worsening of tardive dyskinesia (TD), and potential cognitive impairment. Literature suggests that the first step in managing cognitive concerns in a patient with schizophrenia should be a close review of medications, and avoidance of agents with anticholinergic properties.⁵

Prescribing benztropine for EPS

EPS, which include dystonia, akathisia, drug-induced parkinsonism, and TD, are very frequent adverse effects noted with antipsychotics. Benztropine has demonstrated benefit in managing acute dystonia and the APA Guideline recommends IM administration of either benztropine 1 mg or diphenhydramine 25 mg for this purpose.⁴ However, in our experience, the most frequent indication for long-term prescribing of benztropine is prophylaxis of antipsychoticinduced dystonia. This use was suggested by some older studies. In a 1987 study by Boyer et al,⁶ patients who were administered benztropine with haloperidol did not develop acute dystonia, while patients who received haloperidol alone developed dystonia. However, this was a small retrospective study with methodological issues. Boyer et al⁶ suggested discontinuing prophylaxis with benztropine within 1 week, as acute dystonia occurred within 2.5 days. Other researchers^7,8 have argued that short-term prophylaxis with benztropine for 1 week may work, especially during treatment with high-potency antipsychotics. However, in a review of the use of anticholinergics in conjunction with antipsychotics, Desmarais et al⁵ concluded that there is no need for prophylaxis and recommended alternative treatments. As we have noticed in Ms. P and other patients treated in our facilities, benztropine is frequently continued indefinitely without a clinical indication for its continuous use. Assessment and indication for continued use of benztropine should be considered regularly, and it should be discontinued when there is no clear indication for its use or when adverse effects emerge.

Prescribing benztropine for TD

TD is a subtype of tardive syndromes associated with the use of antipsychotics. It is characterized by repetitive involuntary movements such as lip smacking, puckering, chewing, or tongue protrusion. Proposed pathophysiological mechanisms include dopamine receptor hypersensitivity, N-methyl-D-aspartate (NMDA) receptor excitotoxicity, and gamma-aminobutyric acid (GABA)-containing neuron activity.

According to the APA Guideline, evidence of benztropine’s efficacy for the prevention of TD is lacking.⁴ A 2018 Cochrane systematic review⁹ was unable to provide a definitive conclusion regarding the effectiveness of benztropine and other anticholinergics for the treatment of antipsychotic-induced TD. While many clinicians believe that benztropine can be used to treat all types of EPS, there are no clear instances in reviewed literature where the efficacy of benztropine for treating TD could be reliably demonstrated. Furthermore, some literature suggests that anticholinergics such as benztropine increase the risk of developing TD.^5,10 The mechanism underlying benztropine’s ability to precipitate or exacerbate abnormal movements is unclear, though it is theorized that anticholinergic medications may inhibit dopamine reuptake into neurons, thus leading to an excess of dopamine in the synaptic cleft that manifests as dyskinesias.¹⁰ Some authors also recommend that the first step in the management of TD should be to gradually discontinue anticholinergics, as this has been associated with improvement in TD.¹¹

Continue to: Prescribing anticholinergics in specific patient populations...

In addition to the adverse effects described above, benztropine can affect cognition, as we observed in Ms. P. The cholinergic system plays a role in human cognition, and blockade of muscarinic receptors has been associated with impairments in working memory and prefrontal tasks.¹² These adverse cognitive effects are more pronounced in certain populations, including patients with schizophrenia and older adults.

Schizophrenia is associated with declining cognitive function, and the cognitive faculties of patients with schizophrenia may be worsened by anticholinergics. In patients with schizophrenia, social interactions and social integration are often impacted by profound negative symptoms such as social withdrawal and poverty of thought and speech.¹³ In a double-blind study by Baker et al,¹⁴ benztropine was found to have an impact on attention and concentration in patients with chronic schizophrenia. Baker et al¹⁴ found that patients with schizophrenia who were switched from benztropine to placebo increased their overall Wechsler Memory Scale scores compared to those maintained on benztropine. One crosssectional analysis found that a higher anticholinergic burden was associated with impairments across all cognitive domains, including memory, attention/control, executive and visuospatial functioning, and motor speed domains.¹⁵ Importantly, a higher anticholinergic medication burden was associated with worse cognitive performance.¹⁵ In addition to impairments in cognitive processing, anticholinergics have been associated with a decreased ability to benefit from psychosocial programs and impaired abilities to manage activities of daily living.⁴ In another study exploring the effects of discontinuing anticholinergics and the impact on movement disorders, Desmarais et al¹⁶ found patients experienced a significant improvement in scores on the Brief Assessment of Cognition in Schizophrenia after discontinuing anticholinergics. Vinogradov et al¹⁷ noted that “serum anticholinergic activity in schizophrenia patients shows a significant association with impaired performance in measures of verbal working memory and verbal learning memory and was significantly associated with a lowered response to an intensive course of computerized cognitive training.” They felt their findings underscored the cognitive cost of medications with high anticholinergic burden.

Geriatric patients. Careful consideration should be given before starting benztropine in patients age ≥65. The 2019 American Geriatric Society’s Beers Criteria¹⁸ recommend avoiding benztropine in geriatric patients; the level of recommendation is strong. Furthermore, the American Geriatric Society designates benztropine as a medication that should be avoided, and a nondrug approach or alternative medication be prescribed independent of the patient’s condition or diagnosis. In a recently published case report, Esang et al¹⁹ highlighted several salient findings from previous studies on the risks associated with anticholinergic use:

• any medications a patient takes with anticholinergic properties contribute to the overall anticholinergic load of a patient’s medication regimen
• the higher the anticholinergic burden, the greater the cognitive deficits
• switching from an FGA to an SGA may decrease the risk of EPS and may limit the need for anticholinergic medications such as benztropine for a particular patient.

One must also consider that the effects of multiple medications with anticholinergic properties is probably cumulative.

Alternatives for treating drug-induced parkinsonism

Antipsychotics exert their effects through antagonism of the D2 receptor, and this is the same mechanism that leads to parkinsonism. Specifically, the mechanism is believed to be D2 receptor antagonism in the striatum leading to disinhibition of striatal neurons containing GABA.¹¹ This disinhibition of medium spiny neurons is propagated when acetylcholine is released from cholinergic interneurons. Anticholinergics such as benztropine can remedy symptoms by blocking the signal of acetylcholine on the M1 receptors on medium spiny neurons. However, benztropine also has the propensity to decrease cholinergic transmission, thereby impairing storage of new information into long-term memory as well as impair perception of time—similar to effects seen with (for instance) diphenhydramine.²⁰

The first step in managing drug-induced parkinsonism is to monitor symptoms. The APA Guideline recommends monitoring for acute-onset EPS at weekly intervals when beginning treatment and until stable for 2 weeks, and then monitoring at every follow-up visit thereafter.⁴ The next recommendation for long-term management of drug-induced parkinsonism is reducing the antipsychotic dose, or replacing the patient’s antipsychotic with an antipsychotic that is less likely to precipitate parkinsonism,4 such as quetiapine, iloperidone, or clozapine.¹¹ If dose reduction is not possible, and the patient’s symptoms are severe, pharmacologic management is indicated. The APA Guideline recommends amantadine as a first-line agent because it is associated with fewer peripheral adverse effects and less impairment in cognition compared with benztropine.⁴ In a small (N = 60) doubleblind crossover trial, Gelenberg et al²⁰ found benztropine 4 mg/d—but not amantadine 200 mg/d—impaired free recall and perception of time, and participants’ perception of their own memory impairment was significantly greater with benztropine. Amantadine has also been compared to biperiden, a relatively selective M1 muscarinic receptor muscarinic agent. In a separate double-blind crossover study of 26 patients with chronic schizophrenia, Silver and Geraisy²¹ found that compared to amantadine, biperiden was associated with worse memory performance. The recommended starting dose of amantadine for parkinsonism is 100 mg in the morning, increased to 100 mg twice a day and titrated to a maximum daily dose of 300 mg/d in divided doses.⁴

Continue to: Alternatives for treating drug-induced akathisia...

Akathisia remains a relatively common adverse effect of SGAs, and the profound physical distress and impaired functioning caused by akathisia necessitates pharmacologic treatment. Despite frequent use in practice for presumed benefit in akathisia, benztropine is not effective for the treatment of akathisia and the APA Guideline recommends that long-term management should begin with an antipsychotic dose reduction, followed by a switch to an agent with less propensity to incite akathisia.⁴ Acute manifestations of akathisia must be treated, and mirtazapine, propranolol, or clonazepam may be considered as alternatives.⁴ Mirtazapine is dosed 7.5 mg to 10 mg nightly for akathisia, though it should be used in caution in patients at risk for mania.⁴ Mirtazapine’s potent 5-HT2A blockade at low doses may contribute to its utility in treating akathisia.² Propranolol, a nonselective lipophilic beta-adrenergic antagonist, also has demonstrated efficacy in managing akathisia, with recommended dosing of 40 mg to 80 mg twice daily.² Benzodiazepines such as clonazepam require judicious use for akathisia because they may also precipitate or exacerbate cognitive impairment.⁴

Alternatives for treating TD

As mentioned above, benztropine is not recommended for the treatment of TD.¹ The Box^4,22,23 outlines potential treatment options for TD.

Box

Discontinuing benztropine

Benztropine is recommended as a firstline agent for the management of acute dystonia, and it may be used temporarily for drug-induced parkinsonism, but it is not recommended to prevent EPS or TD. Given the multitude of adverse effects and cognitive impairment noted with anticholinergics, tapering should be considered for patients receiving an anticholinergic agent such as benztropine. Based on their review of earlier studies, Desmarais et al⁵ suggest a gradual 3-month discontinuation of benztropine. Multiple studies have demonstrated an ability to taper anticholinergics in days to months.⁴ However, gradual discontinuation is advisable to avoid cholinergic rebound and the reemergence of EPS, and to decrease the risk of neuroleptic malignant syndrome associated with sudden discontinuation.⁵ One suggested taper regimen is a decrease of 0.5 mg benztropine every week. Amantadine may be considered if parkinsonism is noted during the taper. Patients on benztropine may develop rebound symptoms, such as vivid dreams/nightmares; if this occurs, the taper rate can be slowed to a decrease of 0.5 mg every 2 weeks.⁴

Continue to: First do no harm...

Psychiatrists commonly prescribe benztropine to prevent EPS and TD, but available literature does not support the efficacy of benztropine for mitigating drug-induced parkinsonism, and studies report benztropine may significantly worsen cognitive processes and exacerbate TD.¹⁶ In addition, benztropine misuse has been correlated with euphoria and psychosis.¹⁶ More than 3 decades ago, the World Health Organization Heads of Centres Collaborating in WHO-Coordinated Studies on Biological Aspects of Mental Illness issued a consensus statement²⁴ discouraging the prophylactic use of anticholinergics for patients receiving antipsychotics, yet we still see patients on an indefinite regimen of benztropine.

As clinicians, our goals should be to optimize a patient’s functioning and quality of life, and to use the lowest dose of medication along with the fewest medications necessary to avoid adverse effects such as EPS. Benztropine is recommended as a first-line agent for the management of acute dystonia, but its continued or indefinite use to prevent antipsychotic-induced adverse effects is not recommended. While all pharmacologic interventions carry a risk of adverse effects, weighing the risk of those effects against the clinical benefits is the prerogative of a skilled clinician. Benztropine and other anticholinergics prescribed for prophylactic purposes have numerous adverse effects, limited clinical utility, and a deleterious effect on quality of life. Furthermore, benztropine prophylaxis of drug-induced parkinsonism does not seem to be warranted, and the risks do not seem to outweigh the harm benztropine may cause, with the possible exception of “prophylactic” treatment of dystonia that is discontinued in a few days, as some researchers have suggested.^6-8 The preventive value of benztropine has not been demonstrated. It is time we took inventory of medications that might cause more harm than good, rely on current treatment guidelines instead of habit, and use these agents judiciously while considering replacement with novel, safer medications whenever possible.

CASE CONTINUED

The clinical team considers benztropine’s ability to cause cognitive effects, and decides to taper and discontinue it over 1 month. Ms. P is seen in an outpatient clinic within 1 month of discontinuing benztropine. She reports that her difficulty remembering words and details has improved. She also says that she is now able to concentrate on writing and reading. The consulting neurologist also notes improvement. Ms. P continues to report improvement in symptoms over the next 2 months of follow-up, and says that her mood improved and she has less apathy.

Bottom Line

Benztropine is a first-line medication for acute dystonia, but its continued or indefinite use for preventing antipsychotic-induced adverse effects is not recommended. Given the multitude of adverse effects and cognitive impairment noted with anticholinergics, tapering should be considered for patients receiving an anticholinergic medication such as benztropine.

Ms. P, a 63-year-old woman with a history of schizophrenia whose symptoms have been stable on haloperidol 10 mg/d and ziprasidone 40 mg twice daily, presents to the outpatient clinic for a medication review. She mentions that she has noticed problems with her “memory.” She says she has had difficulty remembering names of people and places as well as difficulty concentrating while reading and writing, which she did months ago with ease. A Montreal Cognitive Assessment (MoCA) is conducted, and Ms. P scores 13/30, indicating moderate cognitive impairment. Visuospatial tasks and clock drawing are intact, but she exhibits impairments in working memory, attention, and concentration. One year ago, Ms. P’s MoCA score was 27/30. She agrees to a neurologic assessment and is referred to neurology for work-up.

Ms. P’s physical examination and routine laboratory tests are all within normal limits. The neurologic exam reveals deficits in working memory, concentration, and attention, but is otherwise unremarkable. MRI reveals mild chronic microvascular changes. The neurology service does not rule out cognitive impairment but recommends adjusting the dosage of Ms. P’s psychiatric medications to elucidate if her impairment of memory and attention is due to medications. However, Ms. P had been managed on her current regimen for several years and had not been hospitalized in many years. Previous attempts to taper her antipsychotics had resulted in worsening symptoms. Ms. P is reluctant to attempt a taper of her antipsychotics because she fears decompensation of her chronic illness. The treating team reviews Ms. P’s medication regimen, and notes that she is receiving benztropine 1 mg twice daily for prophylaxis of extrapyramidal symptoms (EPS). Ms. P denies past or present symptoms of drug-induced parkinsonism, dystonia, or akathisia as well as constipation, sialorrhea, blurry vision, palpitations, or urinary retention.

Benztropine is a tropane alkaloid that was synthetized by combining the tropine portion of atropine with the benzhydryl portion of diphenhydramine hydrochloride. It has anticholinergic and antihistaminic properties¹ and seems to inhibit the dopamine transporter. Benztropine is indicated for all forms of parkinsonism, including antipsychotic-induced parkinsonism, but is also prescribed for many off-label uses, including sialorrhea and akathisia (although many authors do not recommend anticholinergics for this purpose^2,3), and for prophylaxis of EPS. Benztropine can be administered intravenously, intramuscularly, or orally. Given orally, the typical dosing is twice daily with a maximum dose of 6 mg/d. Benztropine is preferred over diphenhydramine and trihexyphenidyl due to adverse effects of sedation or potential for misuse of the medication.¹

Second-generation antipsychotics (SGAs) have been associated with lower rates of neurologic adverse effects compared with first-generation antipsychotics (FGAs). Because SGAs are increasingly prescribed, the use of benztropine (along with other agents such as trihexyphenidyl) for EPS prophylaxis is not an evidence-based practice. However, despite a movement away from prophylactic management of movement disorders, benztropine continues to be prescribed for EPS and/or cholinergic symptoms, despite the peripheral and cognitive adverse effects of this agent and, in many instances, the lack of clear indication for its use.

According to the most recent edition of the American Psychiatric Association’s (APA) Practice Guideline for the Treatment of Patients with Schizophrenia,⁴ anticholinergics should only be used for preventing acute dystonia in conjunction with a long-acting injectable antipsychotic. Furthermore, the APA Guideline states anticholinergics may be used for drug-induced parkinsonism when the dose of an antipsychotic cannot be reduced and an alternative agent is required. However, the first-line agent for drug-induced parkinsonism is amantadine, and benztropine should only be considered if amantadine is contraindicated.⁴ The rationale for this guideline and for judicious use of anticholinergics is that like any pharmacologic treatment, anticholinergics (including benztropine) carry the potential for adverse effects. For benztropine, these range from mild effects such as tachycardia and constipation to paralytic ileus, increased falls, worsening of tardive dyskinesia (TD), and potential cognitive impairment. Literature suggests that the first step in managing cognitive concerns in a patient with schizophrenia should be a close review of medications, and avoidance of agents with anticholinergic properties.⁵

Prescribing benztropine for EPS

EPS, which include dystonia, akathisia, drug-induced parkinsonism, and TD, are very frequent adverse effects noted with antipsychotics. Benztropine has demonstrated benefit in managing acute dystonia and the APA Guideline recommends IM administration of either benztropine 1 mg or diphenhydramine 25 mg for this purpose.⁴ However, in our experience, the most frequent indication for long-term prescribing of benztropine is prophylaxis of antipsychoticinduced dystonia. This use was suggested by some older studies. In a 1987 study by Boyer et al,⁶ patients who were administered benztropine with haloperidol did not develop acute dystonia, while patients who received haloperidol alone developed dystonia. However, this was a small retrospective study with methodological issues. Boyer et al⁶ suggested discontinuing prophylaxis with benztropine within 1 week, as acute dystonia occurred within 2.5 days. Other researchers^7,8 have argued that short-term prophylaxis with benztropine for 1 week may work, especially during treatment with high-potency antipsychotics. However, in a review of the use of anticholinergics in conjunction with antipsychotics, Desmarais et al⁵ concluded that there is no need for prophylaxis and recommended alternative treatments. As we have noticed in Ms. P and other patients treated in our facilities, benztropine is frequently continued indefinitely without a clinical indication for its continuous use. Assessment and indication for continued use of benztropine should be considered regularly, and it should be discontinued when there is no clear indication for its use or when adverse effects emerge.

Prescribing benztropine for TD

TD is a subtype of tardive syndromes associated with the use of antipsychotics. It is characterized by repetitive involuntary movements such as lip smacking, puckering, chewing, or tongue protrusion. Proposed pathophysiological mechanisms include dopamine receptor hypersensitivity, N-methyl-D-aspartate (NMDA) receptor excitotoxicity, and gamma-aminobutyric acid (GABA)-containing neuron activity.

According to the APA Guideline, evidence of benztropine’s efficacy for the prevention of TD is lacking.⁴ A 2018 Cochrane systematic review⁹ was unable to provide a definitive conclusion regarding the effectiveness of benztropine and other anticholinergics for the treatment of antipsychotic-induced TD. While many clinicians believe that benztropine can be used to treat all types of EPS, there are no clear instances in reviewed literature where the efficacy of benztropine for treating TD could be reliably demonstrated. Furthermore, some literature suggests that anticholinergics such as benztropine increase the risk of developing TD.^5,10 The mechanism underlying benztropine’s ability to precipitate or exacerbate abnormal movements is unclear, though it is theorized that anticholinergic medications may inhibit dopamine reuptake into neurons, thus leading to an excess of dopamine in the synaptic cleft that manifests as dyskinesias.¹⁰ Some authors also recommend that the first step in the management of TD should be to gradually discontinue anticholinergics, as this has been associated with improvement in TD.¹¹

Continue to: Prescribing anticholinergics in specific patient populations...

In addition to the adverse effects described above, benztropine can affect cognition, as we observed in Ms. P. The cholinergic system plays a role in human cognition, and blockade of muscarinic receptors has been associated with impairments in working memory and prefrontal tasks.¹² These adverse cognitive effects are more pronounced in certain populations, including patients with schizophrenia and older adults.

Schizophrenia is associated with declining cognitive function, and the cognitive faculties of patients with schizophrenia may be worsened by anticholinergics. In patients with schizophrenia, social interactions and social integration are often impacted by profound negative symptoms such as social withdrawal and poverty of thought and speech.¹³ In a double-blind study by Baker et al,¹⁴ benztropine was found to have an impact on attention and concentration in patients with chronic schizophrenia. Baker et al¹⁴ found that patients with schizophrenia who were switched from benztropine to placebo increased their overall Wechsler Memory Scale scores compared to those maintained on benztropine. One crosssectional analysis found that a higher anticholinergic burden was associated with impairments across all cognitive domains, including memory, attention/control, executive and visuospatial functioning, and motor speed domains.¹⁵ Importantly, a higher anticholinergic medication burden was associated with worse cognitive performance.¹⁵ In addition to impairments in cognitive processing, anticholinergics have been associated with a decreased ability to benefit from psychosocial programs and impaired abilities to manage activities of daily living.⁴ In another study exploring the effects of discontinuing anticholinergics and the impact on movement disorders, Desmarais et al¹⁶ found patients experienced a significant improvement in scores on the Brief Assessment of Cognition in Schizophrenia after discontinuing anticholinergics. Vinogradov et al¹⁷ noted that “serum anticholinergic activity in schizophrenia patients shows a significant association with impaired performance in measures of verbal working memory and verbal learning memory and was significantly associated with a lowered response to an intensive course of computerized cognitive training.” They felt their findings underscored the cognitive cost of medications with high anticholinergic burden.

Geriatric patients. Careful consideration should be given before starting benztropine in patients age ≥65. The 2019 American Geriatric Society’s Beers Criteria¹⁸ recommend avoiding benztropine in geriatric patients; the level of recommendation is strong. Furthermore, the American Geriatric Society designates benztropine as a medication that should be avoided, and a nondrug approach or alternative medication be prescribed independent of the patient’s condition or diagnosis. In a recently published case report, Esang et al¹⁹ highlighted several salient findings from previous studies on the risks associated with anticholinergic use:

• any medications a patient takes with anticholinergic properties contribute to the overall anticholinergic load of a patient’s medication regimen
• the higher the anticholinergic burden, the greater the cognitive deficits
• switching from an FGA to an SGA may decrease the risk of EPS and may limit the need for anticholinergic medications such as benztropine for a particular patient.

One must also consider that the effects of multiple medications with anticholinergic properties is probably cumulative.

Alternatives for treating drug-induced parkinsonism

Antipsychotics exert their effects through antagonism of the D2 receptor, and this is the same mechanism that leads to parkinsonism. Specifically, the mechanism is believed to be D2 receptor antagonism in the striatum leading to disinhibition of striatal neurons containing GABA.¹¹ This disinhibition of medium spiny neurons is propagated when acetylcholine is released from cholinergic interneurons. Anticholinergics such as benztropine can remedy symptoms by blocking the signal of acetylcholine on the M1 receptors on medium spiny neurons. However, benztropine also has the propensity to decrease cholinergic transmission, thereby impairing storage of new information into long-term memory as well as impair perception of time—similar to effects seen with (for instance) diphenhydramine.²⁰

The first step in managing drug-induced parkinsonism is to monitor symptoms. The APA Guideline recommends monitoring for acute-onset EPS at weekly intervals when beginning treatment and until stable for 2 weeks, and then monitoring at every follow-up visit thereafter.⁴ The next recommendation for long-term management of drug-induced parkinsonism is reducing the antipsychotic dose, or replacing the patient’s antipsychotic with an antipsychotic that is less likely to precipitate parkinsonism,4 such as quetiapine, iloperidone, or clozapine.¹¹ If dose reduction is not possible, and the patient’s symptoms are severe, pharmacologic management is indicated. The APA Guideline recommends amantadine as a first-line agent because it is associated with fewer peripheral adverse effects and less impairment in cognition compared with benztropine.⁴ In a small (N = 60) doubleblind crossover trial, Gelenberg et al²⁰ found benztropine 4 mg/d—but not amantadine 200 mg/d—impaired free recall and perception of time, and participants’ perception of their own memory impairment was significantly greater with benztropine. Amantadine has also been compared to biperiden, a relatively selective M1 muscarinic receptor muscarinic agent. In a separate double-blind crossover study of 26 patients with chronic schizophrenia, Silver and Geraisy²¹ found that compared to amantadine, biperiden was associated with worse memory performance. The recommended starting dose of amantadine for parkinsonism is 100 mg in the morning, increased to 100 mg twice a day and titrated to a maximum daily dose of 300 mg/d in divided doses.⁴

Continue to: Alternatives for treating drug-induced akathisia...

Akathisia remains a relatively common adverse effect of SGAs, and the profound physical distress and impaired functioning caused by akathisia necessitates pharmacologic treatment. Despite frequent use in practice for presumed benefit in akathisia, benztropine is not effective for the treatment of akathisia and the APA Guideline recommends that long-term management should begin with an antipsychotic dose reduction, followed by a switch to an agent with less propensity to incite akathisia.⁴ Acute manifestations of akathisia must be treated, and mirtazapine, propranolol, or clonazepam may be considered as alternatives.⁴ Mirtazapine is dosed 7.5 mg to 10 mg nightly for akathisia, though it should be used in caution in patients at risk for mania.⁴ Mirtazapine’s potent 5-HT2A blockade at low doses may contribute to its utility in treating akathisia.² Propranolol, a nonselective lipophilic beta-adrenergic antagonist, also has demonstrated efficacy in managing akathisia, with recommended dosing of 40 mg to 80 mg twice daily.² Benzodiazepines such as clonazepam require judicious use for akathisia because they may also precipitate or exacerbate cognitive impairment.⁴

Alternatives for treating TD

As mentioned above, benztropine is not recommended for the treatment of TD.¹ The Box^4,22,23 outlines potential treatment options for TD.

Box

Discontinuing benztropine

Benztropine is recommended as a firstline agent for the management of acute dystonia, and it may be used temporarily for drug-induced parkinsonism, but it is not recommended to prevent EPS or TD. Given the multitude of adverse effects and cognitive impairment noted with anticholinergics, tapering should be considered for patients receiving an anticholinergic agent such as benztropine. Based on their review of earlier studies, Desmarais et al⁵ suggest a gradual 3-month discontinuation of benztropine. Multiple studies have demonstrated an ability to taper anticholinergics in days to months.⁴ However, gradual discontinuation is advisable to avoid cholinergic rebound and the reemergence of EPS, and to decrease the risk of neuroleptic malignant syndrome associated with sudden discontinuation.⁵ One suggested taper regimen is a decrease of 0.5 mg benztropine every week. Amantadine may be considered if parkinsonism is noted during the taper. Patients on benztropine may develop rebound symptoms, such as vivid dreams/nightmares; if this occurs, the taper rate can be slowed to a decrease of 0.5 mg every 2 weeks.⁴

Continue to: First do no harm...

Psychiatrists commonly prescribe benztropine to prevent EPS and TD, but available literature does not support the efficacy of benztropine for mitigating drug-induced parkinsonism, and studies report benztropine may significantly worsen cognitive processes and exacerbate TD.¹⁶ In addition, benztropine misuse has been correlated with euphoria and psychosis.¹⁶ More than 3 decades ago, the World Health Organization Heads of Centres Collaborating in WHO-Coordinated Studies on Biological Aspects of Mental Illness issued a consensus statement²⁴ discouraging the prophylactic use of anticholinergics for patients receiving antipsychotics, yet we still see patients on an indefinite regimen of benztropine.

As clinicians, our goals should be to optimize a patient’s functioning and quality of life, and to use the lowest dose of medication along with the fewest medications necessary to avoid adverse effects such as EPS. Benztropine is recommended as a first-line agent for the management of acute dystonia, but its continued or indefinite use to prevent antipsychotic-induced adverse effects is not recommended. While all pharmacologic interventions carry a risk of adverse effects, weighing the risk of those effects against the clinical benefits is the prerogative of a skilled clinician. Benztropine and other anticholinergics prescribed for prophylactic purposes have numerous adverse effects, limited clinical utility, and a deleterious effect on quality of life. Furthermore, benztropine prophylaxis of drug-induced parkinsonism does not seem to be warranted, and the risks do not seem to outweigh the harm benztropine may cause, with the possible exception of “prophylactic” treatment of dystonia that is discontinued in a few days, as some researchers have suggested.^6-8 The preventive value of benztropine has not been demonstrated. It is time we took inventory of medications that might cause more harm than good, rely on current treatment guidelines instead of habit, and use these agents judiciously while considering replacement with novel, safer medications whenever possible.

CASE CONTINUED

The clinical team considers benztropine’s ability to cause cognitive effects, and decides to taper and discontinue it over 1 month. Ms. P is seen in an outpatient clinic within 1 month of discontinuing benztropine. She reports that her difficulty remembering words and details has improved. She also says that she is now able to concentrate on writing and reading. The consulting neurologist also notes improvement. Ms. P continues to report improvement in symptoms over the next 2 months of follow-up, and says that her mood improved and she has less apathy.

Bottom Line

Benztropine is a first-line medication for acute dystonia, but its continued or indefinite use for preventing antipsychotic-induced adverse effects is not recommended. Given the multitude of adverse effects and cognitive impairment noted with anticholinergics, tapering should be considered for patients receiving an anticholinergic medication such as benztropine.

Ms. P, a 63-year-old woman with a history of schizophrenia whose symptoms have been stable on haloperidol 10 mg/d and ziprasidone 40 mg twice daily, presents to the outpatient clinic for a medication review. She mentions that she has noticed problems with her “memory.” She says she has had difficulty remembering names of people and places as well as difficulty concentrating while reading and writing, which she did months ago with ease. A Montreal Cognitive Assessment (MoCA) is conducted, and Ms. P scores 13/30, indicating moderate cognitive impairment. Visuospatial tasks and clock drawing are intact, but she exhibits impairments in working memory, attention, and concentration. One year ago, Ms. P’s MoCA score was 27/30. She agrees to a neurologic assessment and is referred to neurology for work-up.

Ms. P’s physical examination and routine laboratory tests are all within normal limits. The neurologic exam reveals deficits in working memory, concentration, and attention, but is otherwise unremarkable. MRI reveals mild chronic microvascular changes. The neurology service does not rule out cognitive impairment but recommends adjusting the dosage of Ms. P’s psychiatric medications to elucidate if her impairment of memory and attention is due to medications. However, Ms. P had been managed on her current regimen for several years and had not been hospitalized in many years. Previous attempts to taper her antipsychotics had resulted in worsening symptoms. Ms. P is reluctant to attempt a taper of her antipsychotics because she fears decompensation of her chronic illness. The treating team reviews Ms. P’s medication regimen, and notes that she is receiving benztropine 1 mg twice daily for prophylaxis of extrapyramidal symptoms (EPS). Ms. P denies past or present symptoms of drug-induced parkinsonism, dystonia, or akathisia as well as constipation, sialorrhea, blurry vision, palpitations, or urinary retention.

Benztropine is a tropane alkaloid that was synthetized by combining the tropine portion of atropine with the benzhydryl portion of diphenhydramine hydrochloride. It has anticholinergic and antihistaminic properties¹ and seems to inhibit the dopamine transporter. Benztropine is indicated for all forms of parkinsonism, including antipsychotic-induced parkinsonism, but is also prescribed for many off-label uses, including sialorrhea and akathisia (although many authors do not recommend anticholinergics for this purpose^2,3), and for prophylaxis of EPS. Benztropine can be administered intravenously, intramuscularly, or orally. Given orally, the typical dosing is twice daily with a maximum dose of 6 mg/d. Benztropine is preferred over diphenhydramine and trihexyphenidyl due to adverse effects of sedation or potential for misuse of the medication.¹

Second-generation antipsychotics (SGAs) have been associated with lower rates of neurologic adverse effects compared with first-generation antipsychotics (FGAs). Because SGAs are increasingly prescribed, the use of benztropine (along with other agents such as trihexyphenidyl) for EPS prophylaxis is not an evidence-based practice. However, despite a movement away from prophylactic management of movement disorders, benztropine continues to be prescribed for EPS and/or cholinergic symptoms, despite the peripheral and cognitive adverse effects of this agent and, in many instances, the lack of clear indication for its use.

According to the most recent edition of the American Psychiatric Association’s (APA) Practice Guideline for the Treatment of Patients with Schizophrenia,⁴ anticholinergics should only be used for preventing acute dystonia in conjunction with a long-acting injectable antipsychotic. Furthermore, the APA Guideline states anticholinergics may be used for drug-induced parkinsonism when the dose of an antipsychotic cannot be reduced and an alternative agent is required. However, the first-line agent for drug-induced parkinsonism is amantadine, and benztropine should only be considered if amantadine is contraindicated.⁴ The rationale for this guideline and for judicious use of anticholinergics is that like any pharmacologic treatment, anticholinergics (including benztropine) carry the potential for adverse effects. For benztropine, these range from mild effects such as tachycardia and constipation to paralytic ileus, increased falls, worsening of tardive dyskinesia (TD), and potential cognitive impairment. Literature suggests that the first step in managing cognitive concerns in a patient with schizophrenia should be a close review of medications, and avoidance of agents with anticholinergic properties.⁵

Prescribing benztropine for EPS

EPS, which include dystonia, akathisia, drug-induced parkinsonism, and TD, are very frequent adverse effects noted with antipsychotics. Benztropine has demonstrated benefit in managing acute dystonia and the APA Guideline recommends IM administration of either benztropine 1 mg or diphenhydramine 25 mg for this purpose.⁴ However, in our experience, the most frequent indication for long-term prescribing of benztropine is prophylaxis of antipsychoticinduced dystonia. This use was suggested by some older studies. In a 1987 study by Boyer et al,⁶ patients who were administered benztropine with haloperidol did not develop acute dystonia, while patients who received haloperidol alone developed dystonia. However, this was a small retrospective study with methodological issues. Boyer et al⁶ suggested discontinuing prophylaxis with benztropine within 1 week, as acute dystonia occurred within 2.5 days. Other researchers^7,8 have argued that short-term prophylaxis with benztropine for 1 week may work, especially during treatment with high-potency antipsychotics. However, in a review of the use of anticholinergics in conjunction with antipsychotics, Desmarais et al⁵ concluded that there is no need for prophylaxis and recommended alternative treatments. As we have noticed in Ms. P and other patients treated in our facilities, benztropine is frequently continued indefinitely without a clinical indication for its continuous use. Assessment and indication for continued use of benztropine should be considered regularly, and it should be discontinued when there is no clear indication for its use or when adverse effects emerge.

Prescribing benztropine for TD

TD is a subtype of tardive syndromes associated with the use of antipsychotics. It is characterized by repetitive involuntary movements such as lip smacking, puckering, chewing, or tongue protrusion. Proposed pathophysiological mechanisms include dopamine receptor hypersensitivity, N-methyl-D-aspartate (NMDA) receptor excitotoxicity, and gamma-aminobutyric acid (GABA)-containing neuron activity.

According to the APA Guideline, evidence of benztropine’s efficacy for the prevention of TD is lacking.⁴ A 2018 Cochrane systematic review⁹ was unable to provide a definitive conclusion regarding the effectiveness of benztropine and other anticholinergics for the treatment of antipsychotic-induced TD. While many clinicians believe that benztropine can be used to treat all types of EPS, there are no clear instances in reviewed literature where the efficacy of benztropine for treating TD could be reliably demonstrated. Furthermore, some literature suggests that anticholinergics such as benztropine increase the risk of developing TD.^5,10 The mechanism underlying benztropine’s ability to precipitate or exacerbate abnormal movements is unclear, though it is theorized that anticholinergic medications may inhibit dopamine reuptake into neurons, thus leading to an excess of dopamine in the synaptic cleft that manifests as dyskinesias.¹⁰ Some authors also recommend that the first step in the management of TD should be to gradually discontinue anticholinergics, as this has been associated with improvement in TD.¹¹

Continue to: Prescribing anticholinergics in specific patient populations...

In addition to the adverse effects described above, benztropine can affect cognition, as we observed in Ms. P. The cholinergic system plays a role in human cognition, and blockade of muscarinic receptors has been associated with impairments in working memory and prefrontal tasks.¹² These adverse cognitive effects are more pronounced in certain populations, including patients with schizophrenia and older adults.

Schizophrenia is associated with declining cognitive function, and the cognitive faculties of patients with schizophrenia may be worsened by anticholinergics. In patients with schizophrenia, social interactions and social integration are often impacted by profound negative symptoms such as social withdrawal and poverty of thought and speech.¹³ In a double-blind study by Baker et al,¹⁴ benztropine was found to have an impact on attention and concentration in patients with chronic schizophrenia. Baker et al¹⁴ found that patients with schizophrenia who were switched from benztropine to placebo increased their overall Wechsler Memory Scale scores compared to those maintained on benztropine. One crosssectional analysis found that a higher anticholinergic burden was associated with impairments across all cognitive domains, including memory, attention/control, executive and visuospatial functioning, and motor speed domains.¹⁵ Importantly, a higher anticholinergic medication burden was associated with worse cognitive performance.¹⁵ In addition to impairments in cognitive processing, anticholinergics have been associated with a decreased ability to benefit from psychosocial programs and impaired abilities to manage activities of daily living.⁴ In another study exploring the effects of discontinuing anticholinergics and the impact on movement disorders, Desmarais et al¹⁶ found patients experienced a significant improvement in scores on the Brief Assessment of Cognition in Schizophrenia after discontinuing anticholinergics. Vinogradov et al¹⁷ noted that “serum anticholinergic activity in schizophrenia patients shows a significant association with impaired performance in measures of verbal working memory and verbal learning memory and was significantly associated with a lowered response to an intensive course of computerized cognitive training.” They felt their findings underscored the cognitive cost of medications with high anticholinergic burden.

Geriatric patients. Careful consideration should be given before starting benztropine in patients age ≥65. The 2019 American Geriatric Society’s Beers Criteria¹⁸ recommend avoiding benztropine in geriatric patients; the level of recommendation is strong. Furthermore, the American Geriatric Society designates benztropine as a medication that should be avoided, and a nondrug approach or alternative medication be prescribed independent of the patient’s condition or diagnosis. In a recently published case report, Esang et al¹⁹ highlighted several salient findings from previous studies on the risks associated with anticholinergic use:

• any medications a patient takes with anticholinergic properties contribute to the overall anticholinergic load of a patient’s medication regimen
• the higher the anticholinergic burden, the greater the cognitive deficits
• switching from an FGA to an SGA may decrease the risk of EPS and may limit the need for anticholinergic medications such as benztropine for a particular patient.

One must also consider that the effects of multiple medications with anticholinergic properties is probably cumulative.

Alternatives for treating drug-induced parkinsonism

Antipsychotics exert their effects through antagonism of the D2 receptor, and this is the same mechanism that leads to parkinsonism. Specifically, the mechanism is believed to be D2 receptor antagonism in the striatum leading to disinhibition of striatal neurons containing GABA.¹¹ This disinhibition of medium spiny neurons is propagated when acetylcholine is released from cholinergic interneurons. Anticholinergics such as benztropine can remedy symptoms by blocking the signal of acetylcholine on the M1 receptors on medium spiny neurons. However, benztropine also has the propensity to decrease cholinergic transmission, thereby impairing storage of new information into long-term memory as well as impair perception of time—similar to effects seen with (for instance) diphenhydramine.²⁰

The first step in managing drug-induced parkinsonism is to monitor symptoms. The APA Guideline recommends monitoring for acute-onset EPS at weekly intervals when beginning treatment and until stable for 2 weeks, and then monitoring at every follow-up visit thereafter.⁴ The next recommendation for long-term management of drug-induced parkinsonism is reducing the antipsychotic dose, or replacing the patient’s antipsychotic with an antipsychotic that is less likely to precipitate parkinsonism,4 such as quetiapine, iloperidone, or clozapine.¹¹ If dose reduction is not possible, and the patient’s symptoms are severe, pharmacologic management is indicated. The APA Guideline recommends amantadine as a first-line agent because it is associated with fewer peripheral adverse effects and less impairment in cognition compared with benztropine.⁴ In a small (N = 60) doubleblind crossover trial, Gelenberg et al²⁰ found benztropine 4 mg/d—but not amantadine 200 mg/d—impaired free recall and perception of time, and participants’ perception of their own memory impairment was significantly greater with benztropine. Amantadine has also been compared to biperiden, a relatively selective M1 muscarinic receptor muscarinic agent. In a separate double-blind crossover study of 26 patients with chronic schizophrenia, Silver and Geraisy²¹ found that compared to amantadine, biperiden was associated with worse memory performance. The recommended starting dose of amantadine for parkinsonism is 100 mg in the morning, increased to 100 mg twice a day and titrated to a maximum daily dose of 300 mg/d in divided doses.⁴

Continue to: Alternatives for treating drug-induced akathisia...

Akathisia remains a relatively common adverse effect of SGAs, and the profound physical distress and impaired functioning caused by akathisia necessitates pharmacologic treatment. Despite frequent use in practice for presumed benefit in akathisia, benztropine is not effective for the treatment of akathisia and the APA Guideline recommends that long-term management should begin with an antipsychotic dose reduction, followed by a switch to an agent with less propensity to incite akathisia.⁴ Acute manifestations of akathisia must be treated, and mirtazapine, propranolol, or clonazepam may be considered as alternatives.⁴ Mirtazapine is dosed 7.5 mg to 10 mg nightly for akathisia, though it should be used in caution in patients at risk for mania.⁴ Mirtazapine’s potent 5-HT2A blockade at low doses may contribute to its utility in treating akathisia.² Propranolol, a nonselective lipophilic beta-adrenergic antagonist, also has demonstrated efficacy in managing akathisia, with recommended dosing of 40 mg to 80 mg twice daily.² Benzodiazepines such as clonazepam require judicious use for akathisia because they may also precipitate or exacerbate cognitive impairment.⁴

Alternatives for treating TD

As mentioned above, benztropine is not recommended for the treatment of TD.¹ The Box^4,22,23 outlines potential treatment options for TD.

Box

Discontinuing benztropine

Benztropine is recommended as a firstline agent for the management of acute dystonia, and it may be used temporarily for drug-induced parkinsonism, but it is not recommended to prevent EPS or TD. Given the multitude of adverse effects and cognitive impairment noted with anticholinergics, tapering should be considered for patients receiving an anticholinergic agent such as benztropine. Based on their review of earlier studies, Desmarais et al⁵ suggest a gradual 3-month discontinuation of benztropine. Multiple studies have demonstrated an ability to taper anticholinergics in days to months.⁴ However, gradual discontinuation is advisable to avoid cholinergic rebound and the reemergence of EPS, and to decrease the risk of neuroleptic malignant syndrome associated with sudden discontinuation.⁵ One suggested taper regimen is a decrease of 0.5 mg benztropine every week. Amantadine may be considered if parkinsonism is noted during the taper. Patients on benztropine may develop rebound symptoms, such as vivid dreams/nightmares; if this occurs, the taper rate can be slowed to a decrease of 0.5 mg every 2 weeks.⁴

Continue to: First do no harm...

Psychiatrists commonly prescribe benztropine to prevent EPS and TD, but available literature does not support the efficacy of benztropine for mitigating drug-induced parkinsonism, and studies report benztropine may significantly worsen cognitive processes and exacerbate TD.¹⁶ In addition, benztropine misuse has been correlated with euphoria and psychosis.¹⁶ More than 3 decades ago, the World Health Organization Heads of Centres Collaborating in WHO-Coordinated Studies on Biological Aspects of Mental Illness issued a consensus statement²⁴ discouraging the prophylactic use of anticholinergics for patients receiving antipsychotics, yet we still see patients on an indefinite regimen of benztropine.

As clinicians, our goals should be to optimize a patient’s functioning and quality of life, and to use the lowest dose of medication along with the fewest medications necessary to avoid adverse effects such as EPS. Benztropine is recommended as a first-line agent for the management of acute dystonia, but its continued or indefinite use to prevent antipsychotic-induced adverse effects is not recommended. While all pharmacologic interventions carry a risk of adverse effects, weighing the risk of those effects against the clinical benefits is the prerogative of a skilled clinician. Benztropine and other anticholinergics prescribed for prophylactic purposes have numerous adverse effects, limited clinical utility, and a deleterious effect on quality of life. Furthermore, benztropine prophylaxis of drug-induced parkinsonism does not seem to be warranted, and the risks do not seem to outweigh the harm benztropine may cause, with the possible exception of “prophylactic” treatment of dystonia that is discontinued in a few days, as some researchers have suggested.^6-8 The preventive value of benztropine has not been demonstrated. It is time we took inventory of medications that might cause more harm than good, rely on current treatment guidelines instead of habit, and use these agents judiciously while considering replacement with novel, safer medications whenever possible.

CASE CONTINUED

The clinical team considers benztropine’s ability to cause cognitive effects, and decides to taper and discontinue it over 1 month. Ms. P is seen in an outpatient clinic within 1 month of discontinuing benztropine. She reports that her difficulty remembering words and details has improved. She also says that she is now able to concentrate on writing and reading. The consulting neurologist also notes improvement. Ms. P continues to report improvement in symptoms over the next 2 months of follow-up, and says that her mood improved and she has less apathy.

Bottom Line

Benztropine is a first-line medication for acute dystonia, but its continued or indefinite use for preventing antipsychotic-induced adverse effects is not recommended. Given the multitude of adverse effects and cognitive impairment noted with anticholinergics, tapering should be considered for patients receiving an anticholinergic medication such as benztropine.

In our experience, most psychiatrists are uncomfortable with prescribing a medication when they feel that doing so would be outside their scope of practice. But there are many situations when prescribing a nonpsychotropic medication would be the correct choice. In this article, we discuss the scope of psychiatric practice, and present 4 case studies that illustrate situations in which psychiatrists should feel comfortable prescribing nonpsychotropic medications.

Defining the scope of practice

What is the scope of a psychiatrist’s practice? Scope of practice usually describes activities that a health care practitioner is allowed to undertake as defined by the terms of his/her license. A license to practice medicine does not include any stipulation restricting practice to a specific medical specialty. However, a local entity may delineate scope of practice within its organization. For instance, local practice standards held by the Detroit Wayne Mental Health Authority (DWMHA) state “Psychiatrists…shall not exceed their scope of practice as per DWMHA credentialing and privileging. For example, a Psychiatrist…who [has] not been appropriately privileged to deliver services to children shall not treat children, excepting crisis situations.”¹

Like physicians in other specialties, psychiatrists are not limited to prescribing only a subset of medications commonly associated with their specialty. But for many psychiatrists, prescribing nonpsychotropic medications is complicated by individual and local factors. On one hand, some psychiatrists do not feel it is their role to prescribe nonpsychotropic medications,² or even some psychotropic medications that may be more complex to prescribe, such as lithium, clozapine, or monoamine oxidase inhibitors.^3-5 However, many feel comfortable prescribing complex combinations of psychotropic medications, or prescribing in a way that does not necessarily make sense (eg, prescribing benztropine as prophylaxis for dystonia when starting an antipsychotic).

Reviewing an average day at one urban psychiatric clinic, these questions seem to come up in half of the patient population, especially in patients with chronic mental illness, multiple medical comorbidities, and limited access to health care. When a young patient walks in without an appointment with an acute dystonic reaction secondary to the initiation of antipsychotics a couple of days ago, there is no hesitation to swiftly and appropriately prescribe an IM anticholinergic medication. But why are psychiatrists often hesitant to prescribe nonpsychotropic medications to treat other adverse effects of medications? Lack of knowledge? Lack of training?

Psychiatrists who practice in hospital systems often have immediate access to consultants, and this availability may encourage them to defer to the consultant for treatment of certain adverse effects. We have seen psychiatrists consult Neurology regarding the prescription of donepezil for mild neurocognitive disorder due to Alzheimer’s disease, or Endocrinology regarding prescription of levothyroxine for lithium-induced hypothyroidism.

However, there are numerous scenarios in which psychiatrists should feel comfortable prescribing nonpsychotropic medications or managing medication adverse effects, regardless of whether they consider it to be within or outside their scope of practice. The following case examples illustrate several such situations.

CASE 1

Ms. W, age 30, has been diagnosed with schizophrenia. She requests a refill of quetiapine, 800 mg/d. This medication has been clearly beneficial in alleviating her psychotic symptoms. However, since her last visit 3 months ago, her face appears more round, and she has gained 9 kg. Further evaluation indicates that she has developed metabolic syndrome and pre-diabetes.

Continue to: Metabolic adverse effects

Metabolic adverse effects, such as metabolic syndrome, diabetic ketoacidosis, and cardiovascular disease, are well-known risks of prescribing second-generation antipsychotics.⁶ In such situations, psychiatrists often advise patients to modify their diet, increase physical activity, and follow up with their primary care physician to determine if other medications are needed. However, getting a patient with a serious mental illness to exercise and modify her/his diet is difficult, and many of these patients do not have a primary care physician.

For patients such as Ms. W, a psychiatrist should consider prescribing metformin. Wu et al⁷ found that in addition to lifestyle modifications, metformin had the greatest effect on antipsychotic-induced weight gain. In this study, metformin alone had more impact on reversing weight gain and increasing insulin sensitivity than lifestyle modifications alone.⁷ This is crucial because these patients are especially vulnerable to cardiac disease.⁸ Metformin is well tolerated and has a low risk of causing hypoglycemia. Concerns regarding lactic acidosis have abated to the extent that the estimated glomerular filtration rate (eGFR) limits for using metformin have been lowered significantly. After reviewing the contraindications, the only knowledge needed to prescribe metformin is the patient’s kidney function and a brief understanding of the titration needed to minimize gastrointestinal adverse effects.⁹ Thus, prescribing metformin would be a fairly logical and easy first step for managing metabolic syndrome, especially in a patient whose motivation for increasing physical activity and modifying his/her diet is doubtful.

CASE 2

Mr. B, age 45, has major depressive disorder that has been well-controlled on paroxetine, 40 mg/d, for the past 2 years. He has no history of physical illness. On his most recent visit, he appears uncomfortable and nervous. After a long discussion, he discloses that his sex life isn’t what it used to be since starting paroxetine. He is bothered by erectile problems and asks whether he can “get some Viagra.”

Sexual adverse effects, such as erectile dysfunction, are frequently associated with the use of selective serotonin reuptake inhibitors.¹⁰ Although managing these adverse effects requires careful evaluation, in most cases, psychiatrists should be able to treat them.¹⁰ The logical choice in this case would be to prescribe one of the 4 FDA-approved phosphodiesterase-5 inhibitors (sildenafil [Viagra], tadalafil [Cialis], vardenafil [Levitra], and avanafil [Stendra]. However, Balon et al¹¹ found that few psychiatrists prescribe phosphodiesterase-5 inhibitors, although they believed that they should be prescribing to treat their patients’ sexual dysfunction. Managing these adverse effects is important not only for the patient’s quality of life and relationship with his/her partner, but also for the therapeutic alliance. In a systematic review of 23 trials, Taylor et al¹² examined >1,800 patients who were prescribed a medication to address sexual dysfunction secondary to antidepressants. They found that for men, adding a phosphodiesterase-5 inhibitor was appropriate and effective, and for women, adding bupropion at high doses should be considered.¹² Like many other adverse effects, sexual adverse effects surely play a role in medication compliance. Dording et al¹³ found that the addition of sildenafil, 50 to 100 mg as needed, resulted in increased treatment satisfaction and overall contentment in 102 patients who complained of sexual dysfunction in the follow-up phase of the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) antidepressant trials. In most cases, with proper psychoeducation, prescription of phosphodiesterase-5 inhibitors is fairly straightforward.

CASE 3

Ms. G, age 22, was recently discharged from an inpatient psychiatric unit after an episode of mania. She was prescribed carbamazepine, 600 mg/d, and ziprasidone, 40 mg twice a day, and appears to be doing well on this regimen. When asked about what led to her admission, she recalls having an elevated mood, increased energy, hypersexuality, impulsivity, and poor judgment. She reveals that she had several sexual partners during her manic episode, and worries that if such behavior occurs again, she may get pregnant. Yet Ms. G was not prescribed birth control upon discharge.

Continue to: Contraception

Contraception. We believe that psychiatrists have an obligation to protect patients from consequences of mental illness. Much the same way that psychiatrists hope to prevent suicide in a patient who has depression, patients should be protected from risks encountered in the manic phase of bipolar disorder. Another reason to prescribe contraceptives in such patients is the teratogenic effects of mood stabilizers. Pagano et al¹⁴ reviewed 6 studies that examined common forms of hormonal birth control to determine their safety in patients with depression or bipolar disorder. They found that overall, use of hormonal contraception was not associated with a worse clinical course of disease.

Many available forms of birth control are available. When prescribing in an outpatient setting, a daily oral medication or a monthly depot injection are convenient options.

CASE 4

Mr. P, age 65, has bipolar I disorder and is stable on risperidone long-acting injection, 37.7 mg bimonthly, and lithium, 1,200 mg/d. He reports that he is doing well but has noticed a recent decrease in energy and weight gain without any change in mood. Laboratory testing conducted prior to this visit revealed a thyroid-stimulating hormone (TSH) level of 4 mU/L (normal range: 0.4 to 4.0 mU/L). Six months ago, Mr. P’s TSH level was 2.8 mU/L. The resident supervisor suggests discussing the case with an endocrinologist.

Thyroid function. The impact of lithium on the thyroid gland is well established; however, psychiatrists’ response to such changes are not.¹⁵ Gitlin¹⁶ reviewed the many adverse effects of lithium and presented various management strategies to address findings such as Mr. P’s. Two important points are that lithium should not be discontinued in light of hypothyroidism, and synthetic thyroxine (levothyroxine) can be initiated and titrated to return TSH levels to a normal range.¹⁶ Levothyroxine can be started at low doses (eg, 25 to 50 mcg/d) and increased every 6 weeks until a normal TSH level is achieved.¹⁷ Managing lithium-induced clinical or subclinical hypothyroidism can prevent further pathology and possible relapse to depression.

Incorporating integrated care

In all these cases, the prescription of a medication with which some psychiatrists are not comfortable prescribing would have been the logical, easiest, and preferable choice. Of course, when initiating any medication, boxed warnings, contraindications, and drug–drug interactions should be reviewed. Initial dosages and titration schedules can be found in every medication’s FDA-approved prescribing information document (package insert), as well as in numerous reference books and articles.

Continue to: We acknowledge...

We acknowledge that prescribing a nonpsychotropic medication is not always a psychiatrist’s best choice, and that in patients with multiple medical comorbidities and drug–drug interactions that are not clearly defined, referring to or consulting a specialist is appropriate. We in no way support reckless prescribing, but instead present an opportunity to expand the perception of what should be considered within a psychiatrist’s scope of practice, and call for further education of psychiatrists so that they are more comfortable managing these adverse effects and/or prescribing at least some nonpsychotropic medications. For example, metabolic syndrome, uncomplicated hypertension, and hypothyroidism (not necessarily lithium-induced) could be managed by psychiatrists practicing integrated care (Table).

We exhort integrated medical care during this time of a physician shortage; however, we do not practice this way. Interestingly, physicians in primary care, such as those in family medicine or obstetrics and gynecology, frequently attempt to treat patients with psychiatric conditions in an attempt to provide integrated care. Numerous articles have discussed these efforts.^18-20 However, this type of integrated care seems less frequent in psychiatry, even though the practice of modern psychiatry in the United States shows substantial overlap with the practice of physicians in primary care specialties.²¹ There are few articles or practical guidelines for psychiatrists who wish to treat patients’ physical illnesses, particularly patients with severe mental illness (see Related Resources, page 56). If we practice in an integrated manner to treat one of the simple conditions we described above, we can eliminate the need for a patient to visit a second physician, pay another co-pay, pay another bus fare, and take another day off work. This can be particularly helpful for patients who at times have to decide between paying for groceries or for medications. Having one clinician manage a patient’s medications also can decrease the risk of polypharmacy.

In addition to the case scenarios described in this article, there are more clinical situations and nonpsychotropic medications that psychiatrists could manage. Considering them outside the scope of psychiatric practice and being uncomfortable or ambivalent about them is not an excuse. We hope that psychiatrists can increase their expertise in this area, and can start to practice as the primary care physicians they claim they are, and should be.

Bottom Line

Many psychiatrists are uncomfortable prescribing nonpsychotropic medications, but there are numerous clinical scenarios in which the practice would make sense. This could include cases of metabolic syndrome, sexual dysfunction secondary to antidepressant use, or other adverse effects of commonly prescribed psychotropic medications.

Related Resources

• McCarron RM, Xiong GL, Keenan CR, et al. Preventive medical care in psychiatry. A practical guide for clinicians. Arlington, VA: American Psychiatric Association Publishing; 2015.
• McCarron RM, Xiong GL, Keenan CR, et al. Study guide to preventive medical care in psychiatry. Arlington, VA: American Psychiatric Association Publishing; 2017.
• Goldberg JF, Ernst CL. Managing the side effects of psychotropic medications. Washington, DC: American Psychiatric Association Publishing; 2019.

Drug Brand Names

Avanafil • Stendra
Benztropine • Cogentin
Bupropion • Wellbutrin, Zyban
Carbamazepine • Carbatrol, Tegretol
Clozapine • Clozaril
Donepezil • Aricept
Levothyroxine • Levoxyl, Synthroid
Lithium • Eskalith, Lithobid
Metformin • Fortamet, Glucophage
Paroxetine • Paxil
Quetiapine • Seroquel
Risperidone long-acting injection • Risperdal Consta
Sildenafil • Viagra
Tadalafil • Cialis
Vardenafil • Levitra
Ziprasidone • Geodon

In our experience, most psychiatrists are uncomfortable with prescribing a medication when they feel that doing so would be outside their scope of practice. But there are many situations when prescribing a nonpsychotropic medication would be the correct choice. In this article, we discuss the scope of psychiatric practice, and present 4 case studies that illustrate situations in which psychiatrists should feel comfortable prescribing nonpsychotropic medications.

Defining the scope of practice

What is the scope of a psychiatrist’s practice? Scope of practice usually describes activities that a health care practitioner is allowed to undertake as defined by the terms of his/her license. A license to practice medicine does not include any stipulation restricting practice to a specific medical specialty. However, a local entity may delineate scope of practice within its organization. For instance, local practice standards held by the Detroit Wayne Mental Health Authority (DWMHA) state “Psychiatrists…shall not exceed their scope of practice as per DWMHA credentialing and privileging. For example, a Psychiatrist…who [has] not been appropriately privileged to deliver services to children shall not treat children, excepting crisis situations.”¹

Like physicians in other specialties, psychiatrists are not limited to prescribing only a subset of medications commonly associated with their specialty. But for many psychiatrists, prescribing nonpsychotropic medications is complicated by individual and local factors. On one hand, some psychiatrists do not feel it is their role to prescribe nonpsychotropic medications,² or even some psychotropic medications that may be more complex to prescribe, such as lithium, clozapine, or monoamine oxidase inhibitors.^3-5 However, many feel comfortable prescribing complex combinations of psychotropic medications, or prescribing in a way that does not necessarily make sense (eg, prescribing benztropine as prophylaxis for dystonia when starting an antipsychotic).

Reviewing an average day at one urban psychiatric clinic, these questions seem to come up in half of the patient population, especially in patients with chronic mental illness, multiple medical comorbidities, and limited access to health care. When a young patient walks in without an appointment with an acute dystonic reaction secondary to the initiation of antipsychotics a couple of days ago, there is no hesitation to swiftly and appropriately prescribe an IM anticholinergic medication. But why are psychiatrists often hesitant to prescribe nonpsychotropic medications to treat other adverse effects of medications? Lack of knowledge? Lack of training?

Psychiatrists who practice in hospital systems often have immediate access to consultants, and this availability may encourage them to defer to the consultant for treatment of certain adverse effects. We have seen psychiatrists consult Neurology regarding the prescription of donepezil for mild neurocognitive disorder due to Alzheimer’s disease, or Endocrinology regarding prescription of levothyroxine for lithium-induced hypothyroidism.

However, there are numerous scenarios in which psychiatrists should feel comfortable prescribing nonpsychotropic medications or managing medication adverse effects, regardless of whether they consider it to be within or outside their scope of practice. The following case examples illustrate several such situations.

CASE 1

Ms. W, age 30, has been diagnosed with schizophrenia. She requests a refill of quetiapine, 800 mg/d. This medication has been clearly beneficial in alleviating her psychotic symptoms. However, since her last visit 3 months ago, her face appears more round, and she has gained 9 kg. Further evaluation indicates that she has developed metabolic syndrome and pre-diabetes.

Continue to: Metabolic adverse effects

Metabolic adverse effects, such as metabolic syndrome, diabetic ketoacidosis, and cardiovascular disease, are well-known risks of prescribing second-generation antipsychotics.⁶ In such situations, psychiatrists often advise patients to modify their diet, increase physical activity, and follow up with their primary care physician to determine if other medications are needed. However, getting a patient with a serious mental illness to exercise and modify her/his diet is difficult, and many of these patients do not have a primary care physician.

For patients such as Ms. W, a psychiatrist should consider prescribing metformin. Wu et al⁷ found that in addition to lifestyle modifications, metformin had the greatest effect on antipsychotic-induced weight gain. In this study, metformin alone had more impact on reversing weight gain and increasing insulin sensitivity than lifestyle modifications alone.⁷ This is crucial because these patients are especially vulnerable to cardiac disease.⁸ Metformin is well tolerated and has a low risk of causing hypoglycemia. Concerns regarding lactic acidosis have abated to the extent that the estimated glomerular filtration rate (eGFR) limits for using metformin have been lowered significantly. After reviewing the contraindications, the only knowledge needed to prescribe metformin is the patient’s kidney function and a brief understanding of the titration needed to minimize gastrointestinal adverse effects.⁹ Thus, prescribing metformin would be a fairly logical and easy first step for managing metabolic syndrome, especially in a patient whose motivation for increasing physical activity and modifying his/her diet is doubtful.

CASE 2

Mr. B, age 45, has major depressive disorder that has been well-controlled on paroxetine, 40 mg/d, for the past 2 years. He has no history of physical illness. On his most recent visit, he appears uncomfortable and nervous. After a long discussion, he discloses that his sex life isn’t what it used to be since starting paroxetine. He is bothered by erectile problems and asks whether he can “get some Viagra.”

Sexual adverse effects, such as erectile dysfunction, are frequently associated with the use of selective serotonin reuptake inhibitors.¹⁰ Although managing these adverse effects requires careful evaluation, in most cases, psychiatrists should be able to treat them.¹⁰ The logical choice in this case would be to prescribe one of the 4 FDA-approved phosphodiesterase-5 inhibitors (sildenafil [Viagra], tadalafil [Cialis], vardenafil [Levitra], and avanafil [Stendra]. However, Balon et al¹¹ found that few psychiatrists prescribe phosphodiesterase-5 inhibitors, although they believed that they should be prescribing to treat their patients’ sexual dysfunction. Managing these adverse effects is important not only for the patient’s quality of life and relationship with his/her partner, but also for the therapeutic alliance. In a systematic review of 23 trials, Taylor et al¹² examined >1,800 patients who were prescribed a medication to address sexual dysfunction secondary to antidepressants. They found that for men, adding a phosphodiesterase-5 inhibitor was appropriate and effective, and for women, adding bupropion at high doses should be considered.¹² Like many other adverse effects, sexual adverse effects surely play a role in medication compliance. Dording et al¹³ found that the addition of sildenafil, 50 to 100 mg as needed, resulted in increased treatment satisfaction and overall contentment in 102 patients who complained of sexual dysfunction in the follow-up phase of the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) antidepressant trials. In most cases, with proper psychoeducation, prescription of phosphodiesterase-5 inhibitors is fairly straightforward.

CASE 3

Ms. G, age 22, was recently discharged from an inpatient psychiatric unit after an episode of mania. She was prescribed carbamazepine, 600 mg/d, and ziprasidone, 40 mg twice a day, and appears to be doing well on this regimen. When asked about what led to her admission, she recalls having an elevated mood, increased energy, hypersexuality, impulsivity, and poor judgment. She reveals that she had several sexual partners during her manic episode, and worries that if such behavior occurs again, she may get pregnant. Yet Ms. G was not prescribed birth control upon discharge.

Continue to: Contraception

Contraception. We believe that psychiatrists have an obligation to protect patients from consequences of mental illness. Much the same way that psychiatrists hope to prevent suicide in a patient who has depression, patients should be protected from risks encountered in the manic phase of bipolar disorder. Another reason to prescribe contraceptives in such patients is the teratogenic effects of mood stabilizers. Pagano et al¹⁴ reviewed 6 studies that examined common forms of hormonal birth control to determine their safety in patients with depression or bipolar disorder. They found that overall, use of hormonal contraception was not associated with a worse clinical course of disease.

Many available forms of birth control are available. When prescribing in an outpatient setting, a daily oral medication or a monthly depot injection are convenient options.

CASE 4

Mr. P, age 65, has bipolar I disorder and is stable on risperidone long-acting injection, 37.7 mg bimonthly, and lithium, 1,200 mg/d. He reports that he is doing well but has noticed a recent decrease in energy and weight gain without any change in mood. Laboratory testing conducted prior to this visit revealed a thyroid-stimulating hormone (TSH) level of 4 mU/L (normal range: 0.4 to 4.0 mU/L). Six months ago, Mr. P’s TSH level was 2.8 mU/L. The resident supervisor suggests discussing the case with an endocrinologist.

Thyroid function. The impact of lithium on the thyroid gland is well established; however, psychiatrists’ response to such changes are not.¹⁵ Gitlin¹⁶ reviewed the many adverse effects of lithium and presented various management strategies to address findings such as Mr. P’s. Two important points are that lithium should not be discontinued in light of hypothyroidism, and synthetic thyroxine (levothyroxine) can be initiated and titrated to return TSH levels to a normal range.¹⁶ Levothyroxine can be started at low doses (eg, 25 to 50 mcg/d) and increased every 6 weeks until a normal TSH level is achieved.¹⁷ Managing lithium-induced clinical or subclinical hypothyroidism can prevent further pathology and possible relapse to depression.

Incorporating integrated care

In all these cases, the prescription of a medication with which some psychiatrists are not comfortable prescribing would have been the logical, easiest, and preferable choice. Of course, when initiating any medication, boxed warnings, contraindications, and drug–drug interactions should be reviewed. Initial dosages and titration schedules can be found in every medication’s FDA-approved prescribing information document (package insert), as well as in numerous reference books and articles.

Continue to: We acknowledge...

We acknowledge that prescribing a nonpsychotropic medication is not always a psychiatrist’s best choice, and that in patients with multiple medical comorbidities and drug–drug interactions that are not clearly defined, referring to or consulting a specialist is appropriate. We in no way support reckless prescribing, but instead present an opportunity to expand the perception of what should be considered within a psychiatrist’s scope of practice, and call for further education of psychiatrists so that they are more comfortable managing these adverse effects and/or prescribing at least some nonpsychotropic medications. For example, metabolic syndrome, uncomplicated hypertension, and hypothyroidism (not necessarily lithium-induced) could be managed by psychiatrists practicing integrated care (Table).

We exhort integrated medical care during this time of a physician shortage; however, we do not practice this way. Interestingly, physicians in primary care, such as those in family medicine or obstetrics and gynecology, frequently attempt to treat patients with psychiatric conditions in an attempt to provide integrated care. Numerous articles have discussed these efforts.^18-20 However, this type of integrated care seems less frequent in psychiatry, even though the practice of modern psychiatry in the United States shows substantial overlap with the practice of physicians in primary care specialties.²¹ There are few articles or practical guidelines for psychiatrists who wish to treat patients’ physical illnesses, particularly patients with severe mental illness (see Related Resources, page 56). If we practice in an integrated manner to treat one of the simple conditions we described above, we can eliminate the need for a patient to visit a second physician, pay another co-pay, pay another bus fare, and take another day off work. This can be particularly helpful for patients who at times have to decide between paying for groceries or for medications. Having one clinician manage a patient’s medications also can decrease the risk of polypharmacy.

In addition to the case scenarios described in this article, there are more clinical situations and nonpsychotropic medications that psychiatrists could manage. Considering them outside the scope of psychiatric practice and being uncomfortable or ambivalent about them is not an excuse. We hope that psychiatrists can increase their expertise in this area, and can start to practice as the primary care physicians they claim they are, and should be.

Bottom Line

Many psychiatrists are uncomfortable prescribing nonpsychotropic medications, but there are numerous clinical scenarios in which the practice would make sense. This could include cases of metabolic syndrome, sexual dysfunction secondary to antidepressant use, or other adverse effects of commonly prescribed psychotropic medications.

Related Resources

• McCarron RM, Xiong GL, Keenan CR, et al. Preventive medical care in psychiatry. A practical guide for clinicians. Arlington, VA: American Psychiatric Association Publishing; 2015.
• McCarron RM, Xiong GL, Keenan CR, et al. Study guide to preventive medical care in psychiatry. Arlington, VA: American Psychiatric Association Publishing; 2017.
• Goldberg JF, Ernst CL. Managing the side effects of psychotropic medications. Washington, DC: American Psychiatric Association Publishing; 2019.

Drug Brand Names

Avanafil • Stendra
Benztropine • Cogentin
Bupropion • Wellbutrin, Zyban
Carbamazepine • Carbatrol, Tegretol
Clozapine • Clozaril
Donepezil • Aricept
Levothyroxine • Levoxyl, Synthroid
Lithium • Eskalith, Lithobid
Metformin • Fortamet, Glucophage
Paroxetine • Paxil
Quetiapine • Seroquel
Risperidone long-acting injection • Risperdal Consta
Sildenafil • Viagra
Tadalafil • Cialis
Vardenafil • Levitra
Ziprasidone • Geodon

In our experience, most psychiatrists are uncomfortable with prescribing a medication when they feel that doing so would be outside their scope of practice. But there are many situations when prescribing a nonpsychotropic medication would be the correct choice. In this article, we discuss the scope of psychiatric practice, and present 4 case studies that illustrate situations in which psychiatrists should feel comfortable prescribing nonpsychotropic medications.

Defining the scope of practice

What is the scope of a psychiatrist’s practice? Scope of practice usually describes activities that a health care practitioner is allowed to undertake as defined by the terms of his/her license. A license to practice medicine does not include any stipulation restricting practice to a specific medical specialty. However, a local entity may delineate scope of practice within its organization. For instance, local practice standards held by the Detroit Wayne Mental Health Authority (DWMHA) state “Psychiatrists…shall not exceed their scope of practice as per DWMHA credentialing and privileging. For example, a Psychiatrist…who [has] not been appropriately privileged to deliver services to children shall not treat children, excepting crisis situations.”¹

Like physicians in other specialties, psychiatrists are not limited to prescribing only a subset of medications commonly associated with their specialty. But for many psychiatrists, prescribing nonpsychotropic medications is complicated by individual and local factors. On one hand, some psychiatrists do not feel it is their role to prescribe nonpsychotropic medications,² or even some psychotropic medications that may be more complex to prescribe, such as lithium, clozapine, or monoamine oxidase inhibitors.^3-5 However, many feel comfortable prescribing complex combinations of psychotropic medications, or prescribing in a way that does not necessarily make sense (eg, prescribing benztropine as prophylaxis for dystonia when starting an antipsychotic).

Reviewing an average day at one urban psychiatric clinic, these questions seem to come up in half of the patient population, especially in patients with chronic mental illness, multiple medical comorbidities, and limited access to health care. When a young patient walks in without an appointment with an acute dystonic reaction secondary to the initiation of antipsychotics a couple of days ago, there is no hesitation to swiftly and appropriately prescribe an IM anticholinergic medication. But why are psychiatrists often hesitant to prescribe nonpsychotropic medications to treat other adverse effects of medications? Lack of knowledge? Lack of training?

Psychiatrists who practice in hospital systems often have immediate access to consultants, and this availability may encourage them to defer to the consultant for treatment of certain adverse effects. We have seen psychiatrists consult Neurology regarding the prescription of donepezil for mild neurocognitive disorder due to Alzheimer’s disease, or Endocrinology regarding prescription of levothyroxine for lithium-induced hypothyroidism.

However, there are numerous scenarios in which psychiatrists should feel comfortable prescribing nonpsychotropic medications or managing medication adverse effects, regardless of whether they consider it to be within or outside their scope of practice. The following case examples illustrate several such situations.

CASE 1

Ms. W, age 30, has been diagnosed with schizophrenia. She requests a refill of quetiapine, 800 mg/d. This medication has been clearly beneficial in alleviating her psychotic symptoms. However, since her last visit 3 months ago, her face appears more round, and she has gained 9 kg. Further evaluation indicates that she has developed metabolic syndrome and pre-diabetes.

Continue to: Metabolic adverse effects

Metabolic adverse effects, such as metabolic syndrome, diabetic ketoacidosis, and cardiovascular disease, are well-known risks of prescribing second-generation antipsychotics.⁶ In such situations, psychiatrists often advise patients to modify their diet, increase physical activity, and follow up with their primary care physician to determine if other medications are needed. However, getting a patient with a serious mental illness to exercise and modify her/his diet is difficult, and many of these patients do not have a primary care physician.

For patients such as Ms. W, a psychiatrist should consider prescribing metformin. Wu et al⁷ found that in addition to lifestyle modifications, metformin had the greatest effect on antipsychotic-induced weight gain. In this study, metformin alone had more impact on reversing weight gain and increasing insulin sensitivity than lifestyle modifications alone.⁷ This is crucial because these patients are especially vulnerable to cardiac disease.⁸ Metformin is well tolerated and has a low risk of causing hypoglycemia. Concerns regarding lactic acidosis have abated to the extent that the estimated glomerular filtration rate (eGFR) limits for using metformin have been lowered significantly. After reviewing the contraindications, the only knowledge needed to prescribe metformin is the patient’s kidney function and a brief understanding of the titration needed to minimize gastrointestinal adverse effects.⁹ Thus, prescribing metformin would be a fairly logical and easy first step for managing metabolic syndrome, especially in a patient whose motivation for increasing physical activity and modifying his/her diet is doubtful.

CASE 2

Mr. B, age 45, has major depressive disorder that has been well-controlled on paroxetine, 40 mg/d, for the past 2 years. He has no history of physical illness. On his most recent visit, he appears uncomfortable and nervous. After a long discussion, he discloses that his sex life isn’t what it used to be since starting paroxetine. He is bothered by erectile problems and asks whether he can “get some Viagra.”

Sexual adverse effects, such as erectile dysfunction, are frequently associated with the use of selective serotonin reuptake inhibitors.¹⁰ Although managing these adverse effects requires careful evaluation, in most cases, psychiatrists should be able to treat them.¹⁰ The logical choice in this case would be to prescribe one of the 4 FDA-approved phosphodiesterase-5 inhibitors (sildenafil [Viagra], tadalafil [Cialis], vardenafil [Levitra], and avanafil [Stendra]. However, Balon et al¹¹ found that few psychiatrists prescribe phosphodiesterase-5 inhibitors, although they believed that they should be prescribing to treat their patients’ sexual dysfunction. Managing these adverse effects is important not only for the patient’s quality of life and relationship with his/her partner, but also for the therapeutic alliance. In a systematic review of 23 trials, Taylor et al¹² examined >1,800 patients who were prescribed a medication to address sexual dysfunction secondary to antidepressants. They found that for men, adding a phosphodiesterase-5 inhibitor was appropriate and effective, and for women, adding bupropion at high doses should be considered.¹² Like many other adverse effects, sexual adverse effects surely play a role in medication compliance. Dording et al¹³ found that the addition of sildenafil, 50 to 100 mg as needed, resulted in increased treatment satisfaction and overall contentment in 102 patients who complained of sexual dysfunction in the follow-up phase of the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) antidepressant trials. In most cases, with proper psychoeducation, prescription of phosphodiesterase-5 inhibitors is fairly straightforward.

CASE 3

Ms. G, age 22, was recently discharged from an inpatient psychiatric unit after an episode of mania. She was prescribed carbamazepine, 600 mg/d, and ziprasidone, 40 mg twice a day, and appears to be doing well on this regimen. When asked about what led to her admission, she recalls having an elevated mood, increased energy, hypersexuality, impulsivity, and poor judgment. She reveals that she had several sexual partners during her manic episode, and worries that if such behavior occurs again, she may get pregnant. Yet Ms. G was not prescribed birth control upon discharge.

Continue to: Contraception

Contraception. We believe that psychiatrists have an obligation to protect patients from consequences of mental illness. Much the same way that psychiatrists hope to prevent suicide in a patient who has depression, patients should be protected from risks encountered in the manic phase of bipolar disorder. Another reason to prescribe contraceptives in such patients is the teratogenic effects of mood stabilizers. Pagano et al¹⁴ reviewed 6 studies that examined common forms of hormonal birth control to determine their safety in patients with depression or bipolar disorder. They found that overall, use of hormonal contraception was not associated with a worse clinical course of disease.

Many available forms of birth control are available. When prescribing in an outpatient setting, a daily oral medication or a monthly depot injection are convenient options.

CASE 4

Mr. P, age 65, has bipolar I disorder and is stable on risperidone long-acting injection, 37.7 mg bimonthly, and lithium, 1,200 mg/d. He reports that he is doing well but has noticed a recent decrease in energy and weight gain without any change in mood. Laboratory testing conducted prior to this visit revealed a thyroid-stimulating hormone (TSH) level of 4 mU/L (normal range: 0.4 to 4.0 mU/L). Six months ago, Mr. P’s TSH level was 2.8 mU/L. The resident supervisor suggests discussing the case with an endocrinologist.

Thyroid function. The impact of lithium on the thyroid gland is well established; however, psychiatrists’ response to such changes are not.¹⁵ Gitlin¹⁶ reviewed the many adverse effects of lithium and presented various management strategies to address findings such as Mr. P’s. Two important points are that lithium should not be discontinued in light of hypothyroidism, and synthetic thyroxine (levothyroxine) can be initiated and titrated to return TSH levels to a normal range.¹⁶ Levothyroxine can be started at low doses (eg, 25 to 50 mcg/d) and increased every 6 weeks until a normal TSH level is achieved.¹⁷ Managing lithium-induced clinical or subclinical hypothyroidism can prevent further pathology and possible relapse to depression.

Incorporating integrated care

In all these cases, the prescription of a medication with which some psychiatrists are not comfortable prescribing would have been the logical, easiest, and preferable choice. Of course, when initiating any medication, boxed warnings, contraindications, and drug–drug interactions should be reviewed. Initial dosages and titration schedules can be found in every medication’s FDA-approved prescribing information document (package insert), as well as in numerous reference books and articles.

Continue to: We acknowledge...

We acknowledge that prescribing a nonpsychotropic medication is not always a psychiatrist’s best choice, and that in patients with multiple medical comorbidities and drug–drug interactions that are not clearly defined, referring to or consulting a specialist is appropriate. We in no way support reckless prescribing, but instead present an opportunity to expand the perception of what should be considered within a psychiatrist’s scope of practice, and call for further education of psychiatrists so that they are more comfortable managing these adverse effects and/or prescribing at least some nonpsychotropic medications. For example, metabolic syndrome, uncomplicated hypertension, and hypothyroidism (not necessarily lithium-induced) could be managed by psychiatrists practicing integrated care (Table).

We exhort integrated medical care during this time of a physician shortage; however, we do not practice this way. Interestingly, physicians in primary care, such as those in family medicine or obstetrics and gynecology, frequently attempt to treat patients with psychiatric conditions in an attempt to provide integrated care. Numerous articles have discussed these efforts.^18-20 However, this type of integrated care seems less frequent in psychiatry, even though the practice of modern psychiatry in the United States shows substantial overlap with the practice of physicians in primary care specialties.²¹ There are few articles or practical guidelines for psychiatrists who wish to treat patients’ physical illnesses, particularly patients with severe mental illness (see Related Resources, page 56). If we practice in an integrated manner to treat one of the simple conditions we described above, we can eliminate the need for a patient to visit a second physician, pay another co-pay, pay another bus fare, and take another day off work. This can be particularly helpful for patients who at times have to decide between paying for groceries or for medications. Having one clinician manage a patient’s medications also can decrease the risk of polypharmacy.

In addition to the case scenarios described in this article, there are more clinical situations and nonpsychotropic medications that psychiatrists could manage. Considering them outside the scope of psychiatric practice and being uncomfortable or ambivalent about them is not an excuse. We hope that psychiatrists can increase their expertise in this area, and can start to practice as the primary care physicians they claim they are, and should be.

Bottom Line

Many psychiatrists are uncomfortable prescribing nonpsychotropic medications, but there are numerous clinical scenarios in which the practice would make sense. This could include cases of metabolic syndrome, sexual dysfunction secondary to antidepressant use, or other adverse effects of commonly prescribed psychotropic medications.

Related Resources

• McCarron RM, Xiong GL, Keenan CR, et al. Preventive medical care in psychiatry. A practical guide for clinicians. Arlington, VA: American Psychiatric Association Publishing; 2015.
• McCarron RM, Xiong GL, Keenan CR, et al. Study guide to preventive medical care in psychiatry. Arlington, VA: American Psychiatric Association Publishing; 2017.
• Goldberg JF, Ernst CL. Managing the side effects of psychotropic medications. Washington, DC: American Psychiatric Association Publishing; 2019.

Drug Brand Names

Avanafil • Stendra
Benztropine • Cogentin
Bupropion • Wellbutrin, Zyban
Carbamazepine • Carbatrol, Tegretol
Clozapine • Clozaril
Donepezil • Aricept
Levothyroxine • Levoxyl, Synthroid
Lithium • Eskalith, Lithobid
Metformin • Fortamet, Glucophage
Paroxetine • Paxil
Quetiapine • Seroquel
Risperidone long-acting injection • Risperdal Consta
Sildenafil • Viagra
Tadalafil • Cialis
Vardenafil • Levitra
Ziprasidone • Geodon

Benzodiazepines’ potential antidepressant properties and their role in the treatment of depression were fairly extensively examined during the 1980s and early 1990s. There were various reasons for this investigation—from the adverse effects of available antidepressants (tricyclic antidepressants [TCAs] and monoamine oxidase inhibitors) to the delay of action of the existing antidepressants and treatment resistance of a significant portion of depressed patients. Benzodiazepines had already been used in the treatment of depressive disorders for decades, but not as monotherapy or main treatment agents, but rather in combination with existing antidepressants to alleviate initial or persistent anxiety, and to help with insomnia. Some authors¹ felt that specific benzodiazepines, such as alprazolam, were effective in mild and moderate depression, although not as effective as TCAs for patients with endogenous or melancholic depression. Others² proposed that benzodiazepines, particularly alprazolam, may be a useful treatment option for patients for whom antidepressants are contraindicated, poorly tolerated, or ineffective. Petty et al² suggested that the antidepressant efficacy of benzodiazepines was consistent with the then-entertained γ-aminobutyric acid theory of depression.

A shift from benzodiazepines to antidepressants

The evidence for using benzodiazepines in anxious depression was based on results of several studies, but it has not been adequately analyzed, summarized, and promoted. Then, after the arrival of the selective serotonin reuptake inhibitors (SSRIs) (fluoxetine arrived in the United States in 1987, and paroxetine and sertraline arrived in 1992), interest in benzodiazepines gradually waned. Within a few years, the SSRIs were also approved for various anxiety disorders. The SSRIs were heavily promoted not only for the treatment of depressive disorders, but also anxiety disorders, and were touted as well-tolerated medications without abuse potential. Benzodiazepines, on the other hand, were frequently described as less effective and having a substantial abuse potential.

Looking back, these claims were not properly substantiated. Berney et al³ concluded in a systematic review that comparative data of a high level of proof for using newer antidepressants in anxiety disorders rather that benzodiazepines were not available. Then, 5 years later, Offidani et al⁴ demonstrated in a systematic review and meta-analysis that benzodiazepines were more effective and better tolerated in the treatment of various anxiety disorders than TCAs. In addition, in a few studies comparing benzodiazepines with newer antidepressants such as paroxetine and venlafaxine, benzodiazepines were either comparable or showed greater improvement and fewer adverse effects that these antidepressants. Similarly to Berney et al,³ Offidani et al⁴ concluded that the change in the prescribing pattern favoring newer antidepressants over benzodiazepines for the treatment of anxiety disorders occurred without supporting evidence.

As far as abuse potential, the American Psychiatric Association Task Force on Benzodiazepine Dependency concluded that benzodiazepines do not strongly reinforce their own use and are not widely abused.⁵ When abuse occurs, it is almost always in the context of abusing other substances. The Task Force also noted that physiological dependence develops when benzodiazepines are used chronically; dependence being defined mostly in terms of symptoms of discontinuance.⁵ Thus, benzodiazepines need to be used appropriately, not in extremely high doses, and under medical supervision.

Nevertheless, the judgment, right or wrong, was out—benzodiazepines were deemed problematic and to be avoided. This has become, unfortunately, a pattern of many prescribing psychiatrists’ practice.

What about benzodiazepines for anxious depression?

Recently Benasi et al⁶ filled the void by investigating data from studies using benzodiazepines as monotherapy in depressive disorders (I was one of the co-authors of this study). They conducted a systematic review of 38 published randomized controlled trials that used benzodiazepines as a monotherapy vs placebo, antidepressants, or both. Patients in these trials were primarily diagnosed with depressive disorder or anxious depression. The majority of these studies used alprazolam as the benzodiazepine (other benzodiazepines used were adinazolam, bromazepam, chlordiazepoxide, and lorazepam) and imipramine or amitriptyline as the antidepressant comparator (other antidepressants used were desipramine, dothiepin, doxepin, and only one newer antidepressant, fluvoxamine, in one study). There was a lack of significant differences in response rate between benzodiazepines and placebo, and between benzodiazepines and TCAs.

In more than half of the studies comparing benzodiazepines with TCAs and/or placebo, benzodiazepines were significantly more effective than placebo and as effective as TCAs. In 11 studies, TCAs were better than benzodiazepines, while benzodiazepines were better than TCAs in one study. In 12 studies, benzodiazepines were associated with a faster onset of action than TCAs. Adverse effects occurred more frequently with TCAs, with the exception of drowsiness and cognitive impairment, which occurred more frequently with benzodiazepines. The findings of the meta-analysis (22 studies) confirmed the low response of anxious depression to psychotropic medications, whether TCAs or benzodiazepines. There was no demonstrated superiority of antidepressants over benzodiazepines for anxious depression. Thus, clearly, benzodiazepines are a bona fide therapeutic option for anxious depression and so far, there is no indication that antidepressants are preferable for this indication.

Continue to: However, it is important to note...

However, it is important to note that there are almost no studies comparing benzodiazepines to newer antidepressants for anxious depression. One double-blind 6-week study of 112 patients⁷ compared fluvoxamine with lorazepam for mixed anxiety and depression in general practice. There were no significant differences between treatments at any point in the study. Lorazepam produced more sedation, while fluvoxamine produced more nausea and vomiting.

We clearly need randomized controlled trials comparing benzodiazepines with newer antidepressants in anxious depression. However, as in the case with anxiety disorders, these types of trials are strikingly missing.

Any clinical wisdom?

Anxiety could be a serious clinical problem in the treatment of patients with depressive disorder(s). We have not always paid enough attention to anxiety and related issues in depressed patients. Interestingly, anxiety has not been listed among symptoms of major depression disorder (MDD) in several editions of the Diagnostic and Statistical Manual of Mental Disorders (DSM). Only and finally did DSM-5⁸ add a specifier “with anxious distress” for both MDD and persistent depressive disorder (dysthymia), although this specifier still avoids the word “anxiety” in the description of its symptomatology.

It is difficult to disentangle whether the anxiety is part of depressive disorder symptomatology or whether it is a comorbid anxiety disorder. As I noted in a previous article,⁹ psychiatric comorbidity is a confusing phenomenon. Nevertheless, anxiety and depression are highly comorbid or co-symptomatologic. In a study by Kessler et al,¹⁰ 45.7% of survey responders with lifetime MDD had ≥1 lifetime anxiety disorder. Similarly, in a STAR*D study,¹¹ in Level 1, 53.2% of patients had anxious depression.

Kessler et al¹⁰ raised an interesting question about the importance of temporally primary anxiety disorders as risk markers vs causal risk factors for the onset and persistence of subsequent MDD, including the possibility that anxiety disorders might primarily be risk markers for MDD onset and causal risk factors for MDD persistence. As is well-known, mood disorders should be treated as soon as possible after they are diagnosed, and should be treated vigorously, addressing the major symptomatology.

Continue to: These findings emphasize the need to...

These findings emphasize the need to pay more attention to anxiety in depressed patients (especially those newly diagnosed) and for forceful treatment of anxious depression. Importantly, in the STAR*D study,¹¹ remission in anxious Level 1 (treated with citalopram) depressed patients was significantly less likely and took longer to occur than in patients with nonanxious depression. In addition, ratings of adverse effects frequency, intensity, and burden, as well as the number of serious adverse events, were significantly greater in the anxious depression group. Similarly, in Level 2 (either switched to bupropion, sertraline or venlafaxine, or citalopram augmented with bupropion or buspirone), patients with anxious depression fared significantly worse in both the switching and augmentation options. One wonders if Level 1 patients treated with benzodiazepines, and Level 2 patients switched to benzodiazepines or offered augmentation with them would not have fared better, especially in view of the fact that many old and new antidepressants have significant adverse effects and are difficult to discontinue due to withdrawal symptoms such as dizziness, vertigo, and, in case of newer antidepressants, brain “zaps.” Benzodiazepines certainly have serious withdrawal symptoms, including anxiety, rebound insomnia, and withdrawal seizures, especially when discontinued abruptly and when the dose was high. Thus, as is the case for many other medications (eg, steroids, anticoagulants, and some antidepressants), benzodiazepines must be tapered carefully in order to avoid discontinuance signs and symptoms. Because benzodiazepines have been involved in nearly one-third of overdose-related deaths (either separately or in combination with opioids), and the FDA strongly warns against co-prescribing benzodiazepines and opioids, they need to be prescribed appropriately, carefully weighing their risks and benefits.¹²

Because the analysis by Benasi et al⁶ demonstrated that benzodiazepines seem comparably effective as antidepressants in anxious depression, we should be considering using benzodiazepines as monotherapy for this indication more frequently and vigorously, considering their similar efficacy, faster onset of action, and better tolerability, while also considering their risks. Clinicians use them in combinations anyway. We also need rigorous trials comparing benzodiazepines with newer antidepressants for anxious depression.

Benzodiazepines’ potential antidepressant properties and their role in the treatment of depression were fairly extensively examined during the 1980s and early 1990s. There were various reasons for this investigation—from the adverse effects of available antidepressants (tricyclic antidepressants [TCAs] and monoamine oxidase inhibitors) to the delay of action of the existing antidepressants and treatment resistance of a significant portion of depressed patients. Benzodiazepines had already been used in the treatment of depressive disorders for decades, but not as monotherapy or main treatment agents, but rather in combination with existing antidepressants to alleviate initial or persistent anxiety, and to help with insomnia. Some authors¹ felt that specific benzodiazepines, such as alprazolam, were effective in mild and moderate depression, although not as effective as TCAs for patients with endogenous or melancholic depression. Others² proposed that benzodiazepines, particularly alprazolam, may be a useful treatment option for patients for whom antidepressants are contraindicated, poorly tolerated, or ineffective. Petty et al² suggested that the antidepressant efficacy of benzodiazepines was consistent with the then-entertained γ-aminobutyric acid theory of depression.

A shift from benzodiazepines to antidepressants

The evidence for using benzodiazepines in anxious depression was based on results of several studies, but it has not been adequately analyzed, summarized, and promoted. Then, after the arrival of the selective serotonin reuptake inhibitors (SSRIs) (fluoxetine arrived in the United States in 1987, and paroxetine and sertraline arrived in 1992), interest in benzodiazepines gradually waned. Within a few years, the SSRIs were also approved for various anxiety disorders. The SSRIs were heavily promoted not only for the treatment of depressive disorders, but also anxiety disorders, and were touted as well-tolerated medications without abuse potential. Benzodiazepines, on the other hand, were frequently described as less effective and having a substantial abuse potential.

Looking back, these claims were not properly substantiated. Berney et al³ concluded in a systematic review that comparative data of a high level of proof for using newer antidepressants in anxiety disorders rather that benzodiazepines were not available. Then, 5 years later, Offidani et al⁴ demonstrated in a systematic review and meta-analysis that benzodiazepines were more effective and better tolerated in the treatment of various anxiety disorders than TCAs. In addition, in a few studies comparing benzodiazepines with newer antidepressants such as paroxetine and venlafaxine, benzodiazepines were either comparable or showed greater improvement and fewer adverse effects that these antidepressants. Similarly to Berney et al,³ Offidani et al⁴ concluded that the change in the prescribing pattern favoring newer antidepressants over benzodiazepines for the treatment of anxiety disorders occurred without supporting evidence.

As far as abuse potential, the American Psychiatric Association Task Force on Benzodiazepine Dependency concluded that benzodiazepines do not strongly reinforce their own use and are not widely abused.⁵ When abuse occurs, it is almost always in the context of abusing other substances. The Task Force also noted that physiological dependence develops when benzodiazepines are used chronically; dependence being defined mostly in terms of symptoms of discontinuance.⁵ Thus, benzodiazepines need to be used appropriately, not in extremely high doses, and under medical supervision.

Nevertheless, the judgment, right or wrong, was out—benzodiazepines were deemed problematic and to be avoided. This has become, unfortunately, a pattern of many prescribing psychiatrists’ practice.

What about benzodiazepines for anxious depression?

Recently Benasi et al⁶ filled the void by investigating data from studies using benzodiazepines as monotherapy in depressive disorders (I was one of the co-authors of this study). They conducted a systematic review of 38 published randomized controlled trials that used benzodiazepines as a monotherapy vs placebo, antidepressants, or both. Patients in these trials were primarily diagnosed with depressive disorder or anxious depression. The majority of these studies used alprazolam as the benzodiazepine (other benzodiazepines used were adinazolam, bromazepam, chlordiazepoxide, and lorazepam) and imipramine or amitriptyline as the antidepressant comparator (other antidepressants used were desipramine, dothiepin, doxepin, and only one newer antidepressant, fluvoxamine, in one study). There was a lack of significant differences in response rate between benzodiazepines and placebo, and between benzodiazepines and TCAs.

In more than half of the studies comparing benzodiazepines with TCAs and/or placebo, benzodiazepines were significantly more effective than placebo and as effective as TCAs. In 11 studies, TCAs were better than benzodiazepines, while benzodiazepines were better than TCAs in one study. In 12 studies, benzodiazepines were associated with a faster onset of action than TCAs. Adverse effects occurred more frequently with TCAs, with the exception of drowsiness and cognitive impairment, which occurred more frequently with benzodiazepines. The findings of the meta-analysis (22 studies) confirmed the low response of anxious depression to psychotropic medications, whether TCAs or benzodiazepines. There was no demonstrated superiority of antidepressants over benzodiazepines for anxious depression. Thus, clearly, benzodiazepines are a bona fide therapeutic option for anxious depression and so far, there is no indication that antidepressants are preferable for this indication.

Continue to: However, it is important to note...

However, it is important to note that there are almost no studies comparing benzodiazepines to newer antidepressants for anxious depression. One double-blind 6-week study of 112 patients⁷ compared fluvoxamine with lorazepam for mixed anxiety and depression in general practice. There were no significant differences between treatments at any point in the study. Lorazepam produced more sedation, while fluvoxamine produced more nausea and vomiting.

We clearly need randomized controlled trials comparing benzodiazepines with newer antidepressants in anxious depression. However, as in the case with anxiety disorders, these types of trials are strikingly missing.

Any clinical wisdom?

Anxiety could be a serious clinical problem in the treatment of patients with depressive disorder(s). We have not always paid enough attention to anxiety and related issues in depressed patients. Interestingly, anxiety has not been listed among symptoms of major depression disorder (MDD) in several editions of the Diagnostic and Statistical Manual of Mental Disorders (DSM). Only and finally did DSM-5⁸ add a specifier “with anxious distress” for both MDD and persistent depressive disorder (dysthymia), although this specifier still avoids the word “anxiety” in the description of its symptomatology.

It is difficult to disentangle whether the anxiety is part of depressive disorder symptomatology or whether it is a comorbid anxiety disorder. As I noted in a previous article,⁹ psychiatric comorbidity is a confusing phenomenon. Nevertheless, anxiety and depression are highly comorbid or co-symptomatologic. In a study by Kessler et al,¹⁰ 45.7% of survey responders with lifetime MDD had ≥1 lifetime anxiety disorder. Similarly, in a STAR*D study,¹¹ in Level 1, 53.2% of patients had anxious depression.

Kessler et al¹⁰ raised an interesting question about the importance of temporally primary anxiety disorders as risk markers vs causal risk factors for the onset and persistence of subsequent MDD, including the possibility that anxiety disorders might primarily be risk markers for MDD onset and causal risk factors for MDD persistence. As is well-known, mood disorders should be treated as soon as possible after they are diagnosed, and should be treated vigorously, addressing the major symptomatology.

Continue to: These findings emphasize the need to...

These findings emphasize the need to pay more attention to anxiety in depressed patients (especially those newly diagnosed) and for forceful treatment of anxious depression. Importantly, in the STAR*D study,¹¹ remission in anxious Level 1 (treated with citalopram) depressed patients was significantly less likely and took longer to occur than in patients with nonanxious depression. In addition, ratings of adverse effects frequency, intensity, and burden, as well as the number of serious adverse events, were significantly greater in the anxious depression group. Similarly, in Level 2 (either switched to bupropion, sertraline or venlafaxine, or citalopram augmented with bupropion or buspirone), patients with anxious depression fared significantly worse in both the switching and augmentation options. One wonders if Level 1 patients treated with benzodiazepines, and Level 2 patients switched to benzodiazepines or offered augmentation with them would not have fared better, especially in view of the fact that many old and new antidepressants have significant adverse effects and are difficult to discontinue due to withdrawal symptoms such as dizziness, vertigo, and, in case of newer antidepressants, brain “zaps.” Benzodiazepines certainly have serious withdrawal symptoms, including anxiety, rebound insomnia, and withdrawal seizures, especially when discontinued abruptly and when the dose was high. Thus, as is the case for many other medications (eg, steroids, anticoagulants, and some antidepressants), benzodiazepines must be tapered carefully in order to avoid discontinuance signs and symptoms. Because benzodiazepines have been involved in nearly one-third of overdose-related deaths (either separately or in combination with opioids), and the FDA strongly warns against co-prescribing benzodiazepines and opioids, they need to be prescribed appropriately, carefully weighing their risks and benefits.¹²

Because the analysis by Benasi et al⁶ demonstrated that benzodiazepines seem comparably effective as antidepressants in anxious depression, we should be considering using benzodiazepines as monotherapy for this indication more frequently and vigorously, considering their similar efficacy, faster onset of action, and better tolerability, while also considering their risks. Clinicians use them in combinations anyway. We also need rigorous trials comparing benzodiazepines with newer antidepressants for anxious depression.

Benzodiazepines’ potential antidepressant properties and their role in the treatment of depression were fairly extensively examined during the 1980s and early 1990s. There were various reasons for this investigation—from the adverse effects of available antidepressants (tricyclic antidepressants [TCAs] and monoamine oxidase inhibitors) to the delay of action of the existing antidepressants and treatment resistance of a significant portion of depressed patients. Benzodiazepines had already been used in the treatment of depressive disorders for decades, but not as monotherapy or main treatment agents, but rather in combination with existing antidepressants to alleviate initial or persistent anxiety, and to help with insomnia. Some authors¹ felt that specific benzodiazepines, such as alprazolam, were effective in mild and moderate depression, although not as effective as TCAs for patients with endogenous or melancholic depression. Others² proposed that benzodiazepines, particularly alprazolam, may be a useful treatment option for patients for whom antidepressants are contraindicated, poorly tolerated, or ineffective. Petty et al² suggested that the antidepressant efficacy of benzodiazepines was consistent with the then-entertained γ-aminobutyric acid theory of depression.

A shift from benzodiazepines to antidepressants

The evidence for using benzodiazepines in anxious depression was based on results of several studies, but it has not been adequately analyzed, summarized, and promoted. Then, after the arrival of the selective serotonin reuptake inhibitors (SSRIs) (fluoxetine arrived in the United States in 1987, and paroxetine and sertraline arrived in 1992), interest in benzodiazepines gradually waned. Within a few years, the SSRIs were also approved for various anxiety disorders. The SSRIs were heavily promoted not only for the treatment of depressive disorders, but also anxiety disorders, and were touted as well-tolerated medications without abuse potential. Benzodiazepines, on the other hand, were frequently described as less effective and having a substantial abuse potential.

Looking back, these claims were not properly substantiated. Berney et al³ concluded in a systematic review that comparative data of a high level of proof for using newer antidepressants in anxiety disorders rather that benzodiazepines were not available. Then, 5 years later, Offidani et al⁴ demonstrated in a systematic review and meta-analysis that benzodiazepines were more effective and better tolerated in the treatment of various anxiety disorders than TCAs. In addition, in a few studies comparing benzodiazepines with newer antidepressants such as paroxetine and venlafaxine, benzodiazepines were either comparable or showed greater improvement and fewer adverse effects that these antidepressants. Similarly to Berney et al,³ Offidani et al⁴ concluded that the change in the prescribing pattern favoring newer antidepressants over benzodiazepines for the treatment of anxiety disorders occurred without supporting evidence.

As far as abuse potential, the American Psychiatric Association Task Force on Benzodiazepine Dependency concluded that benzodiazepines do not strongly reinforce their own use and are not widely abused.⁵ When abuse occurs, it is almost always in the context of abusing other substances. The Task Force also noted that physiological dependence develops when benzodiazepines are used chronically; dependence being defined mostly in terms of symptoms of discontinuance.⁵ Thus, benzodiazepines need to be used appropriately, not in extremely high doses, and under medical supervision.

Nevertheless, the judgment, right or wrong, was out—benzodiazepines were deemed problematic and to be avoided. This has become, unfortunately, a pattern of many prescribing psychiatrists’ practice.

What about benzodiazepines for anxious depression?

Recently Benasi et al⁶ filled the void by investigating data from studies using benzodiazepines as monotherapy in depressive disorders (I was one of the co-authors of this study). They conducted a systematic review of 38 published randomized controlled trials that used benzodiazepines as a monotherapy vs placebo, antidepressants, or both. Patients in these trials were primarily diagnosed with depressive disorder or anxious depression. The majority of these studies used alprazolam as the benzodiazepine (other benzodiazepines used were adinazolam, bromazepam, chlordiazepoxide, and lorazepam) and imipramine or amitriptyline as the antidepressant comparator (other antidepressants used were desipramine, dothiepin, doxepin, and only one newer antidepressant, fluvoxamine, in one study). There was a lack of significant differences in response rate between benzodiazepines and placebo, and between benzodiazepines and TCAs.

In more than half of the studies comparing benzodiazepines with TCAs and/or placebo, benzodiazepines were significantly more effective than placebo and as effective as TCAs. In 11 studies, TCAs were better than benzodiazepines, while benzodiazepines were better than TCAs in one study. In 12 studies, benzodiazepines were associated with a faster onset of action than TCAs. Adverse effects occurred more frequently with TCAs, with the exception of drowsiness and cognitive impairment, which occurred more frequently with benzodiazepines. The findings of the meta-analysis (22 studies) confirmed the low response of anxious depression to psychotropic medications, whether TCAs or benzodiazepines. There was no demonstrated superiority of antidepressants over benzodiazepines for anxious depression. Thus, clearly, benzodiazepines are a bona fide therapeutic option for anxious depression and so far, there is no indication that antidepressants are preferable for this indication.

Continue to: However, it is important to note...

However, it is important to note that there are almost no studies comparing benzodiazepines to newer antidepressants for anxious depression. One double-blind 6-week study of 112 patients⁷ compared fluvoxamine with lorazepam for mixed anxiety and depression in general practice. There were no significant differences between treatments at any point in the study. Lorazepam produced more sedation, while fluvoxamine produced more nausea and vomiting.

We clearly need randomized controlled trials comparing benzodiazepines with newer antidepressants in anxious depression. However, as in the case with anxiety disorders, these types of trials are strikingly missing.

Any clinical wisdom?

Anxiety could be a serious clinical problem in the treatment of patients with depressive disorder(s). We have not always paid enough attention to anxiety and related issues in depressed patients. Interestingly, anxiety has not been listed among symptoms of major depression disorder (MDD) in several editions of the Diagnostic and Statistical Manual of Mental Disorders (DSM). Only and finally did DSM-5⁸ add a specifier “with anxious distress” for both MDD and persistent depressive disorder (dysthymia), although this specifier still avoids the word “anxiety” in the description of its symptomatology.

It is difficult to disentangle whether the anxiety is part of depressive disorder symptomatology or whether it is a comorbid anxiety disorder. As I noted in a previous article,⁹ psychiatric comorbidity is a confusing phenomenon. Nevertheless, anxiety and depression are highly comorbid or co-symptomatologic. In a study by Kessler et al,¹⁰ 45.7% of survey responders with lifetime MDD had ≥1 lifetime anxiety disorder. Similarly, in a STAR*D study,¹¹ in Level 1, 53.2% of patients had anxious depression.

Kessler et al¹⁰ raised an interesting question about the importance of temporally primary anxiety disorders as risk markers vs causal risk factors for the onset and persistence of subsequent MDD, including the possibility that anxiety disorders might primarily be risk markers for MDD onset and causal risk factors for MDD persistence. As is well-known, mood disorders should be treated as soon as possible after they are diagnosed, and should be treated vigorously, addressing the major symptomatology.

Continue to: These findings emphasize the need to...

These findings emphasize the need to pay more attention to anxiety in depressed patients (especially those newly diagnosed) and for forceful treatment of anxious depression. Importantly, in the STAR*D study,¹¹ remission in anxious Level 1 (treated with citalopram) depressed patients was significantly less likely and took longer to occur than in patients with nonanxious depression. In addition, ratings of adverse effects frequency, intensity, and burden, as well as the number of serious adverse events, were significantly greater in the anxious depression group. Similarly, in Level 2 (either switched to bupropion, sertraline or venlafaxine, or citalopram augmented with bupropion or buspirone), patients with anxious depression fared significantly worse in both the switching and augmentation options. One wonders if Level 1 patients treated with benzodiazepines, and Level 2 patients switched to benzodiazepines or offered augmentation with them would not have fared better, especially in view of the fact that many old and new antidepressants have significant adverse effects and are difficult to discontinue due to withdrawal symptoms such as dizziness, vertigo, and, in case of newer antidepressants, brain “zaps.” Benzodiazepines certainly have serious withdrawal symptoms, including anxiety, rebound insomnia, and withdrawal seizures, especially when discontinued abruptly and when the dose was high. Thus, as is the case for many other medications (eg, steroids, anticoagulants, and some antidepressants), benzodiazepines must be tapered carefully in order to avoid discontinuance signs and symptoms. Because benzodiazepines have been involved in nearly one-third of overdose-related deaths (either separately or in combination with opioids), and the FDA strongly warns against co-prescribing benzodiazepines and opioids, they need to be prescribed appropriately, carefully weighing their risks and benefits.¹²

Because the analysis by Benasi et al⁶ demonstrated that benzodiazepines seem comparably effective as antidepressants in anxious depression, we should be considering using benzodiazepines as monotherapy for this indication more frequently and vigorously, considering their similar efficacy, faster onset of action, and better tolerability, while also considering their risks. Clinicians use them in combinations anyway. We also need rigorous trials comparing benzodiazepines with newer antidepressants for anxious depression.

CASE Hallucinations, impaired memory
Mr. C is a 61-year-old African American man who visits the outpatient clinic for management of antipsychotic therapy for psychosis and depression. His most recent inpatient psychiatric hospitalization for auditory and visual hallucinations, paranoia, and agitation was more than 10 years ago. He has been taking chlorpromazine, 100 mg/d, for 11 years. Mr. C reports that he has had no psychotic symptoms in the past 3 years; he continues taking chlorpromazine, he says, because it helps him sleep.

How would you proceed with Mr. C’s care?
   a) continue chlorpromazine because he has been symptom free
   b) consider tapering and discontinuing chlorpromazine
   c) obtain a more detailed history from Mr. C and perform additional tests

HISTORY Validation of diagnosis
Mr. C reports that, at age 48, he started hearing babies crying and started seeing dead infants crawling out of the incinerator at the hospital where he worked. He denies any psychiatric symptoms before that time. He stopped working 10 years ago because of his psychiatric symptoms and decline in cognition.

Subsequently, Mr. C had 3 inpatient psychiatric hospitalizations for auditory hallucinations; chlorpromazine, 100 mg/d, was prescribed for psychosis. Later efforts to discontinue chlorpromazine resulted in relapse of psychotic symptoms. Mr. C has no family history of psychiatric illness.

Mr. C’s medical history is significant for aortic regurgitation, congestive cardiac failure, hypertension, and left-sided sensorineural hearing loss. He has a history of cocaine abuse from age 21 to 45, but denies using any other substances, including alcohol and nicotine.

Urine toxicology and routine blood tests are within normal limits. The QTc is slightly prolonged over the past 2 years, recording 512, 520, and 505 milliseconds on serial electrocardiograms.

Mr. C is able to perform simple abstractions. He has a goal-directed thought process, devoid of any preoccupation, paranoia, and perceptual abnormalities. Cognitive screening reveals significant impairment of memory, registration, calculation, attention, and visuospatial skills.

Careful review of Mr. C’s history and medical records reveals a diagnosis of syphilis at age 48 after unprotected sexual intercourse. He recalls that he had a solitary genital lesion, which resolved over a few weeks. He then developed a slightly itchy, non-tender macular rash over his upper back, which he did not report to a physician. After a few months, he developed unsteady gait, blurry vision, and weakness of limbs, and had to crawl to the hospital. There, he was given a diagnosis of neurosyphilis. He also developed left-sided hearing loss during that time.

Mr. C was treated with aqueous penicillin G benzathine, 4 million units IV for 2 weeks. No follow-up cerebrospinal fluid (CSF) examination was documented after antibiotic treatment. He developed auditory and visual hallucinations and paranoia a few months after completing penicillin treatment. During the following year, he had 3 inpatient psychiatric hospitalizations for psychosis, agitation, and depressed mood.

How would you treat a patient with a history of neurosyphilis who presents with psychosis years after diagnosis?
   a) repeat antibiotic treatment and stop the antipsychotic
   b) repeat antibiotic treatment and continue the antipsychotic
   c) attempt to discontinue the antipsychotic
   d) continue the antipsychotic

The authors’ observations
Mr. C’s psychotic symptoms seem to be temporally related to his diagnosis of neurosyphilis at age 48. He and his family members deny that Mr. C had any history of psychosis or depression before the neurosyphilis diagnosis. All inpatient psychiatric hospitalizations were within 1 year of the neurosyphilis diagnosis.

Mr. C has been on a low dosage of chlorpromazine, which has significant antihistaminic action. Chlorpromazine also is known to cause QTc prolongation, especially in patients with heart disease.

TREATMENT Medication change
A serum rapid plasma reagin test is non-reactive, but Treponema pallidum particle agglutination is positive. MRI shows moderate atrophy suggestive of diffuse small-vessel disease.

Mr. C’s psychotic symptoms are considered to be sequelae of neurosyphilis, based on (1) the presence of positive antibody tests, (2) residual neurologic deficits, (3) other suggestive sequelae (aortic regurgitation, sensorineural deafness), and (4) age-inappropriate gradual cognitive decline in the absence of other psychiatric history.

Because we are concerned about the prolonged QTc, chlorpromazine is discontinued. Haloperidol, 5 mg at bedtime, is started. The neurology team does not recommend antibiotic treatment because symptoms have been stable for years. Mr. C refuses a lumbar puncture.

Mr. C returns to the outpatient clinic monthly. He is psychiatrically stable without any worsening of psychosis. Cognitive impairment remains stable over the next 6 months. Haloperidol is tapered to 2 mg at bedtime 6 months after initial evaluation. Mr. C remains psychiatrically stable on subsequent follow-up visits.

The authors’ observations
Mr. C’s psychotic symptoms persisted after standard antibiotic treatment of neurosyphilis and lapsed when he stopped taking antipsychotic medication 10 years after the initial treatment of neurosyphilis. He carried a diagnosis of schizophrenia for many years, even though his psychotic symptoms were atypical for the presentation of schizophrenia.

It is important to understand the natural course of syphilis, its implication on psychiatric symptom production, and long-term psychiatric prognosis.

Syphilis is a sexually transmitted infectious disease caused by T pallidum, a spirochete, that has varied clinical presentations. Osler called syphilis the “great imitator” for its array of system involvement, ranging from asymptomatic infection and afferent pupillary defect to depression, psychosis, and dementia. With wide use of penicillin, the rate of neurosyphilis declined steadily during the mid 1990s. By 1997, the overall rate reached its lowest point in the United States; in 1999 the Centers for Disease Control and Prevention released a national plan to eliminate syphilis.¹ By 2004, however, prevalence had increased to 4.7/100,000. It is thought that this increase is mainly associated with substance use (especially crack cocaine) and HIV co-infection. Most cases were distributed in economically depressed geographical areas.

Psychiatric patients are at higher risk of acquiring the infection because of substance use, lack of education on safer sex practices, and impulsive behavior.

Stages of syphilis
Syphilis does not follow a step-wise progression. One-third of cases progress to the tertiary stage, even many years after initial infection, without adequate treatment.²

Almost 10% syphilis cases present with neurologic symptoms,³ and neurologic involvement can occur at any stage of disease progression. The most common symptoms of syphilis are presented in Table 1.

A range of psychiatric symptoms have been reported among patients with syphilis, including anhedonia, suicidality, mania, grandiosity, persecutory delusions, auditory and visual hallucinations, paranoia, and cognitive impairment. The incidence of psychiatric symptoms is not clearly described in literature.

Diagnosis and treatment
Neurosyphilis, at any disease stage, should be suspected if a patient:

• exhibits suggestive symptoms
• does not respond to antibiotic treatment
• has late latent syphilis
• is immunocompromised.

Lumbar puncture and examination of CSF is the most useful diagnostic test. Dark field microscopy to reveal T pallidum is definitive, but only is applicable during the primary stage. The role of dark field microscopy of the CSF sample to diagnose neurologic involvement has not been established. Tests and treatment protocol are described in Table 2.^2-5

Treatment of psychiatric symptoms of neurosyphilis
There are inconsistent and limited data about the prevalence of psychiatric symptoms in neurosyphilis. A retrospective study⁶ of 161 patients with neurosyphilis in South Africa reported that 50.9% exhibited a complex spectrum of symptoms that included delirium and dementia. Of treated patients, 17% continued to have residual symptoms during follow-up.

A review of the literature did not reveal any widely accepted guideline for screening for neurosyphilis in general psychiatry practice or a treatment protocol for psychiatric symptoms. This lack of guidance could be attributed to the rarity of the disease, cost-benefit analyses, and low specificity of antibody tests. In the literature, syphilis screening is recommended as a routine protocol when evaluating and treating dementia.⁷

In most studies, a diagnosis of neurosyphilis was confirmed by CSF examination; however, many of these studies did not report a specific follow-up CSF examination protocol. Most of these patients were treated with an antipsychotic with partial improvement in symptoms, even after standard antibiotic protocol.⁸

First- and second-generation antipsychotics and mood stabilizers have been shown to be useful in the acute treatment of psychosis and agitation.⁸ In few instances, the psychotropic medication was continued beyond several months and the patient was placed in a long-term care facility. Psychiatric symptoms persisted for many years with or without residual neurosyphilis symptoms, possibly because of permanent neuronal loss.

Clinical considerations
It often is difficult to distinguish a pre­existing psychiatric disorder made worse by neurosyphilis from a secondary psychiatric disorder caused by neurosyphilis. The 2 might coexist, or psychiatric symptoms could be wrongly attributed to schizophrenia because of a lack of careful clinical evaluation.

Often, the follow-up diagnostic protocol for neurosyphilis is not followed; as a result, the need for re-treatment remains unclear. Rarity of the disease makes it difficult to perform a prospective, randomized study to determine the duration and effect of long-term psychiatric treatment.

Close follow-up and consideration of the risk vs benefit of psychotropic medication is key. Because there are no proven guidelines for the length of treatment with antipsychotics, it is prudent to minimize their use until psychiatrically indicated. Side effects, such as (in Mr. C’s case) changes in the QTc interval, should warrant consideration of discontinuing psychotropic medication. Interdisciplinary collaboration with neurology and infectious disease will improve the overall outcome of a complex clinical presentation.

CASE Hallucinations, impaired memory
Mr. C is a 61-year-old African American man who visits the outpatient clinic for management of antipsychotic therapy for psychosis and depression. His most recent inpatient psychiatric hospitalization for auditory and visual hallucinations, paranoia, and agitation was more than 10 years ago. He has been taking chlorpromazine, 100 mg/d, for 11 years. Mr. C reports that he has had no psychotic symptoms in the past 3 years; he continues taking chlorpromazine, he says, because it helps him sleep.

How would you proceed with Mr. C’s care?
   a) continue chlorpromazine because he has been symptom free
   b) consider tapering and discontinuing chlorpromazine
   c) obtain a more detailed history from Mr. C and perform additional tests

HISTORY Validation of diagnosis
Mr. C reports that, at age 48, he started hearing babies crying and started seeing dead infants crawling out of the incinerator at the hospital where he worked. He denies any psychiatric symptoms before that time. He stopped working 10 years ago because of his psychiatric symptoms and decline in cognition.

Subsequently, Mr. C had 3 inpatient psychiatric hospitalizations for auditory hallucinations; chlorpromazine, 100 mg/d, was prescribed for psychosis. Later efforts to discontinue chlorpromazine resulted in relapse of psychotic symptoms. Mr. C has no family history of psychiatric illness.

Mr. C’s medical history is significant for aortic regurgitation, congestive cardiac failure, hypertension, and left-sided sensorineural hearing loss. He has a history of cocaine abuse from age 21 to 45, but denies using any other substances, including alcohol and nicotine.

Urine toxicology and routine blood tests are within normal limits. The QTc is slightly prolonged over the past 2 years, recording 512, 520, and 505 milliseconds on serial electrocardiograms.

Mr. C is able to perform simple abstractions. He has a goal-directed thought process, devoid of any preoccupation, paranoia, and perceptual abnormalities. Cognitive screening reveals significant impairment of memory, registration, calculation, attention, and visuospatial skills.

Careful review of Mr. C’s history and medical records reveals a diagnosis of syphilis at age 48 after unprotected sexual intercourse. He recalls that he had a solitary genital lesion, which resolved over a few weeks. He then developed a slightly itchy, non-tender macular rash over his upper back, which he did not report to a physician. After a few months, he developed unsteady gait, blurry vision, and weakness of limbs, and had to crawl to the hospital. There, he was given a diagnosis of neurosyphilis. He also developed left-sided hearing loss during that time.

Mr. C was treated with aqueous penicillin G benzathine, 4 million units IV for 2 weeks. No follow-up cerebrospinal fluid (CSF) examination was documented after antibiotic treatment. He developed auditory and visual hallucinations and paranoia a few months after completing penicillin treatment. During the following year, he had 3 inpatient psychiatric hospitalizations for psychosis, agitation, and depressed mood.

How would you treat a patient with a history of neurosyphilis who presents with psychosis years after diagnosis?
   a) repeat antibiotic treatment and stop the antipsychotic
   b) repeat antibiotic treatment and continue the antipsychotic
   c) attempt to discontinue the antipsychotic
   d) continue the antipsychotic

The authors’ observations
Mr. C’s psychotic symptoms seem to be temporally related to his diagnosis of neurosyphilis at age 48. He and his family members deny that Mr. C had any history of psychosis or depression before the neurosyphilis diagnosis. All inpatient psychiatric hospitalizations were within 1 year of the neurosyphilis diagnosis.

Mr. C has been on a low dosage of chlorpromazine, which has significant antihistaminic action. Chlorpromazine also is known to cause QTc prolongation, especially in patients with heart disease.

TREATMENT Medication change
A serum rapid plasma reagin test is non-reactive, but Treponema pallidum particle agglutination is positive. MRI shows moderate atrophy suggestive of diffuse small-vessel disease.

Mr. C’s psychotic symptoms are considered to be sequelae of neurosyphilis, based on (1) the presence of positive antibody tests, (2) residual neurologic deficits, (3) other suggestive sequelae (aortic regurgitation, sensorineural deafness), and (4) age-inappropriate gradual cognitive decline in the absence of other psychiatric history.

Because we are concerned about the prolonged QTc, chlorpromazine is discontinued. Haloperidol, 5 mg at bedtime, is started. The neurology team does not recommend antibiotic treatment because symptoms have been stable for years. Mr. C refuses a lumbar puncture.

Mr. C returns to the outpatient clinic monthly. He is psychiatrically stable without any worsening of psychosis. Cognitive impairment remains stable over the next 6 months. Haloperidol is tapered to 2 mg at bedtime 6 months after initial evaluation. Mr. C remains psychiatrically stable on subsequent follow-up visits.

The authors’ observations
Mr. C’s psychotic symptoms persisted after standard antibiotic treatment of neurosyphilis and lapsed when he stopped taking antipsychotic medication 10 years after the initial treatment of neurosyphilis. He carried a diagnosis of schizophrenia for many years, even though his psychotic symptoms were atypical for the presentation of schizophrenia.

It is important to understand the natural course of syphilis, its implication on psychiatric symptom production, and long-term psychiatric prognosis.

Syphilis is a sexually transmitted infectious disease caused by T pallidum, a spirochete, that has varied clinical presentations. Osler called syphilis the “great imitator” for its array of system involvement, ranging from asymptomatic infection and afferent pupillary defect to depression, psychosis, and dementia. With wide use of penicillin, the rate of neurosyphilis declined steadily during the mid 1990s. By 1997, the overall rate reached its lowest point in the United States; in 1999 the Centers for Disease Control and Prevention released a national plan to eliminate syphilis.¹ By 2004, however, prevalence had increased to 4.7/100,000. It is thought that this increase is mainly associated with substance use (especially crack cocaine) and HIV co-infection. Most cases were distributed in economically depressed geographical areas.

Psychiatric patients are at higher risk of acquiring the infection because of substance use, lack of education on safer sex practices, and impulsive behavior.

Stages of syphilis
Syphilis does not follow a step-wise progression. One-third of cases progress to the tertiary stage, even many years after initial infection, without adequate treatment.²

Almost 10% syphilis cases present with neurologic symptoms,³ and neurologic involvement can occur at any stage of disease progression. The most common symptoms of syphilis are presented in Table 1.

A range of psychiatric symptoms have been reported among patients with syphilis, including anhedonia, suicidality, mania, grandiosity, persecutory delusions, auditory and visual hallucinations, paranoia, and cognitive impairment. The incidence of psychiatric symptoms is not clearly described in literature.

Diagnosis and treatment
Neurosyphilis, at any disease stage, should be suspected if a patient:

• exhibits suggestive symptoms
• does not respond to antibiotic treatment
• has late latent syphilis
• is immunocompromised.

Lumbar puncture and examination of CSF is the most useful diagnostic test. Dark field microscopy to reveal T pallidum is definitive, but only is applicable during the primary stage. The role of dark field microscopy of the CSF sample to diagnose neurologic involvement has not been established. Tests and treatment protocol are described in Table 2.^2-5

Treatment of psychiatric symptoms of neurosyphilis
There are inconsistent and limited data about the prevalence of psychiatric symptoms in neurosyphilis. A retrospective study⁶ of 161 patients with neurosyphilis in South Africa reported that 50.9% exhibited a complex spectrum of symptoms that included delirium and dementia. Of treated patients, 17% continued to have residual symptoms during follow-up.

A review of the literature did not reveal any widely accepted guideline for screening for neurosyphilis in general psychiatry practice or a treatment protocol for psychiatric symptoms. This lack of guidance could be attributed to the rarity of the disease, cost-benefit analyses, and low specificity of antibody tests. In the literature, syphilis screening is recommended as a routine protocol when evaluating and treating dementia.⁷

In most studies, a diagnosis of neurosyphilis was confirmed by CSF examination; however, many of these studies did not report a specific follow-up CSF examination protocol. Most of these patients were treated with an antipsychotic with partial improvement in symptoms, even after standard antibiotic protocol.⁸

First- and second-generation antipsychotics and mood stabilizers have been shown to be useful in the acute treatment of psychosis and agitation.⁸ In few instances, the psychotropic medication was continued beyond several months and the patient was placed in a long-term care facility. Psychiatric symptoms persisted for many years with or without residual neurosyphilis symptoms, possibly because of permanent neuronal loss.

Clinical considerations
It often is difficult to distinguish a pre­existing psychiatric disorder made worse by neurosyphilis from a secondary psychiatric disorder caused by neurosyphilis. The 2 might coexist, or psychiatric symptoms could be wrongly attributed to schizophrenia because of a lack of careful clinical evaluation.

Often, the follow-up diagnostic protocol for neurosyphilis is not followed; as a result, the need for re-treatment remains unclear. Rarity of the disease makes it difficult to perform a prospective, randomized study to determine the duration and effect of long-term psychiatric treatment.

Close follow-up and consideration of the risk vs benefit of psychotropic medication is key. Because there are no proven guidelines for the length of treatment with antipsychotics, it is prudent to minimize their use until psychiatrically indicated. Side effects, such as (in Mr. C’s case) changes in the QTc interval, should warrant consideration of discontinuing psychotropic medication. Interdisciplinary collaboration with neurology and infectious disease will improve the overall outcome of a complex clinical presentation.

CASE Hallucinations, impaired memory
Mr. C is a 61-year-old African American man who visits the outpatient clinic for management of antipsychotic therapy for psychosis and depression. His most recent inpatient psychiatric hospitalization for auditory and visual hallucinations, paranoia, and agitation was more than 10 years ago. He has been taking chlorpromazine, 100 mg/d, for 11 years. Mr. C reports that he has had no psychotic symptoms in the past 3 years; he continues taking chlorpromazine, he says, because it helps him sleep.

How would you proceed with Mr. C’s care?
   a) continue chlorpromazine because he has been symptom free
   b) consider tapering and discontinuing chlorpromazine
   c) obtain a more detailed history from Mr. C and perform additional tests

HISTORY Validation of diagnosis
Mr. C reports that, at age 48, he started hearing babies crying and started seeing dead infants crawling out of the incinerator at the hospital where he worked. He denies any psychiatric symptoms before that time. He stopped working 10 years ago because of his psychiatric symptoms and decline in cognition.

Subsequently, Mr. C had 3 inpatient psychiatric hospitalizations for auditory hallucinations; chlorpromazine, 100 mg/d, was prescribed for psychosis. Later efforts to discontinue chlorpromazine resulted in relapse of psychotic symptoms. Mr. C has no family history of psychiatric illness.

Mr. C’s medical history is significant for aortic regurgitation, congestive cardiac failure, hypertension, and left-sided sensorineural hearing loss. He has a history of cocaine abuse from age 21 to 45, but denies using any other substances, including alcohol and nicotine.

Urine toxicology and routine blood tests are within normal limits. The QTc is slightly prolonged over the past 2 years, recording 512, 520, and 505 milliseconds on serial electrocardiograms.

Mr. C is able to perform simple abstractions. He has a goal-directed thought process, devoid of any preoccupation, paranoia, and perceptual abnormalities. Cognitive screening reveals significant impairment of memory, registration, calculation, attention, and visuospatial skills.

Careful review of Mr. C’s history and medical records reveals a diagnosis of syphilis at age 48 after unprotected sexual intercourse. He recalls that he had a solitary genital lesion, which resolved over a few weeks. He then developed a slightly itchy, non-tender macular rash over his upper back, which he did not report to a physician. After a few months, he developed unsteady gait, blurry vision, and weakness of limbs, and had to crawl to the hospital. There, he was given a diagnosis of neurosyphilis. He also developed left-sided hearing loss during that time.

Mr. C was treated with aqueous penicillin G benzathine, 4 million units IV for 2 weeks. No follow-up cerebrospinal fluid (CSF) examination was documented after antibiotic treatment. He developed auditory and visual hallucinations and paranoia a few months after completing penicillin treatment. During the following year, he had 3 inpatient psychiatric hospitalizations for psychosis, agitation, and depressed mood.

How would you treat a patient with a history of neurosyphilis who presents with psychosis years after diagnosis?
   a) repeat antibiotic treatment and stop the antipsychotic
   b) repeat antibiotic treatment and continue the antipsychotic
   c) attempt to discontinue the antipsychotic
   d) continue the antipsychotic

The authors’ observations
Mr. C’s psychotic symptoms seem to be temporally related to his diagnosis of neurosyphilis at age 48. He and his family members deny that Mr. C had any history of psychosis or depression before the neurosyphilis diagnosis. All inpatient psychiatric hospitalizations were within 1 year of the neurosyphilis diagnosis.

Mr. C has been on a low dosage of chlorpromazine, which has significant antihistaminic action. Chlorpromazine also is known to cause QTc prolongation, especially in patients with heart disease.

TREATMENT Medication change
A serum rapid plasma reagin test is non-reactive, but Treponema pallidum particle agglutination is positive. MRI shows moderate atrophy suggestive of diffuse small-vessel disease.

Mr. C’s psychotic symptoms are considered to be sequelae of neurosyphilis, based on (1) the presence of positive antibody tests, (2) residual neurologic deficits, (3) other suggestive sequelae (aortic regurgitation, sensorineural deafness), and (4) age-inappropriate gradual cognitive decline in the absence of other psychiatric history.

Because we are concerned about the prolonged QTc, chlorpromazine is discontinued. Haloperidol, 5 mg at bedtime, is started. The neurology team does not recommend antibiotic treatment because symptoms have been stable for years. Mr. C refuses a lumbar puncture.

Mr. C returns to the outpatient clinic monthly. He is psychiatrically stable without any worsening of psychosis. Cognitive impairment remains stable over the next 6 months. Haloperidol is tapered to 2 mg at bedtime 6 months after initial evaluation. Mr. C remains psychiatrically stable on subsequent follow-up visits.

The authors’ observations
Mr. C’s psychotic symptoms persisted after standard antibiotic treatment of neurosyphilis and lapsed when he stopped taking antipsychotic medication 10 years after the initial treatment of neurosyphilis. He carried a diagnosis of schizophrenia for many years, even though his psychotic symptoms were atypical for the presentation of schizophrenia.

It is important to understand the natural course of syphilis, its implication on psychiatric symptom production, and long-term psychiatric prognosis.

Syphilis is a sexually transmitted infectious disease caused by T pallidum, a spirochete, that has varied clinical presentations. Osler called syphilis the “great imitator” for its array of system involvement, ranging from asymptomatic infection and afferent pupillary defect to depression, psychosis, and dementia. With wide use of penicillin, the rate of neurosyphilis declined steadily during the mid 1990s. By 1997, the overall rate reached its lowest point in the United States; in 1999 the Centers for Disease Control and Prevention released a national plan to eliminate syphilis.¹ By 2004, however, prevalence had increased to 4.7/100,000. It is thought that this increase is mainly associated with substance use (especially crack cocaine) and HIV co-infection. Most cases were distributed in economically depressed geographical areas.

Psychiatric patients are at higher risk of acquiring the infection because of substance use, lack of education on safer sex practices, and impulsive behavior.

Stages of syphilis
Syphilis does not follow a step-wise progression. One-third of cases progress to the tertiary stage, even many years after initial infection, without adequate treatment.²

Almost 10% syphilis cases present with neurologic symptoms,³ and neurologic involvement can occur at any stage of disease progression. The most common symptoms of syphilis are presented in Table 1.

A range of psychiatric symptoms have been reported among patients with syphilis, including anhedonia, suicidality, mania, grandiosity, persecutory delusions, auditory and visual hallucinations, paranoia, and cognitive impairment. The incidence of psychiatric symptoms is not clearly described in literature.

Diagnosis and treatment
Neurosyphilis, at any disease stage, should be suspected if a patient:

• exhibits suggestive symptoms
• does not respond to antibiotic treatment
• has late latent syphilis
• is immunocompromised.

Lumbar puncture and examination of CSF is the most useful diagnostic test. Dark field microscopy to reveal T pallidum is definitive, but only is applicable during the primary stage. The role of dark field microscopy of the CSF sample to diagnose neurologic involvement has not been established. Tests and treatment protocol are described in Table 2.^2-5

Treatment of psychiatric symptoms of neurosyphilis
There are inconsistent and limited data about the prevalence of psychiatric symptoms in neurosyphilis. A retrospective study⁶ of 161 patients with neurosyphilis in South Africa reported that 50.9% exhibited a complex spectrum of symptoms that included delirium and dementia. Of treated patients, 17% continued to have residual symptoms during follow-up.

A review of the literature did not reveal any widely accepted guideline for screening for neurosyphilis in general psychiatry practice or a treatment protocol for psychiatric symptoms. This lack of guidance could be attributed to the rarity of the disease, cost-benefit analyses, and low specificity of antibody tests. In the literature, syphilis screening is recommended as a routine protocol when evaluating and treating dementia.⁷

In most studies, a diagnosis of neurosyphilis was confirmed by CSF examination; however, many of these studies did not report a specific follow-up CSF examination protocol. Most of these patients were treated with an antipsychotic with partial improvement in symptoms, even after standard antibiotic protocol.⁸

First- and second-generation antipsychotics and mood stabilizers have been shown to be useful in the acute treatment of psychosis and agitation.⁸ In few instances, the psychotropic medication was continued beyond several months and the patient was placed in a long-term care facility. Psychiatric symptoms persisted for many years with or without residual neurosyphilis symptoms, possibly because of permanent neuronal loss.

Clinical considerations
It often is difficult to distinguish a pre­existing psychiatric disorder made worse by neurosyphilis from a secondary psychiatric disorder caused by neurosyphilis. The 2 might coexist, or psychiatric symptoms could be wrongly attributed to schizophrenia because of a lack of careful clinical evaluation.

Often, the follow-up diagnostic protocol for neurosyphilis is not followed; as a result, the need for re-treatment remains unclear. Rarity of the disease makes it difficult to perform a prospective, randomized study to determine the duration and effect of long-term psychiatric treatment.

Close follow-up and consideration of the risk vs benefit of psychotropic medication is key. Because there are no proven guidelines for the length of treatment with antipsychotics, it is prudent to minimize their use until psychiatrically indicated. Side effects, such as (in Mr. C’s case) changes in the QTc interval, should warrant consideration of discontinuing psychotropic medication. Interdisciplinary collaboration with neurology and infectious disease will improve the overall outcome of a complex clinical presentation.

Flibanserin, FDA-approved in August 2015, is the first medication approved to treat acquired, generalized hypoactive sexual desire disorder (HSDD) in premenopausal women (Table 1). In clinical trials,^1-4 the drug has shown modest efficacy in improving symptoms of low sexual desire (number of satisfying sexual events [SSEs], sexual desire, and overall sexual function). Flibanserin is not indicated to enhance sexual performance, for HSDD in postmenopausal women, or in men.
 

Clinical implications
Flibanserin could help premenopausal women who have distressing low sexual desire, which must be acquired and generalized:

• “Acquired low sexual desire” means that a patient had an adequate sexual desire that decreased or ceased for an unknown reason.
• “Generalized low sexual desire” means that lack of sexual desire occurs all the time and in all situations, not only with a certain partner or in some situations.

Women taking flibanserin could experience gradually increased sexual desire, increase in SSEs, and decrease of sexual distress. Flibanserin is indicated for long-term use; however, it should be discontinued after 8 weeks if the patient does not report any improvement in symptoms.

The number needed to treat with flibanserin likely would be rather large, but it is not available because of complex outcome measures in clinical trials. Flibanserin was not approved at 2 previous FDA committee hearings—mainly because of safety issues but also because of concerns about efficacy. For example, during the 2013 FDA hearing, the results presented showed statistically significant, but numerically small, treatment differences at 24 weeks compared with placebo. In an FDA responder analysis of the Phase-III trials, after accounting for the placebo effect, approximately 8% to 13% women were at least “much improved” on at least 1 of the primary outcomes.⁵

Flibanserin is not indicated for women whose sexual desire is due to (1) coexisting medical or psychiatric condition, (2) effects of medication or substance abuse, or (3) a relationship problem. It is unknown whether supplemental treatment would help these patients; however, it seems reasonable that combining flibanserin with psychosocial treatment, such as sex therapy or individual therapy, could be beneficial because it may be difficult to disentangle sexual dysfunction and relationship issues—2 problems that often are interwoven.

How it works
Flibanserin is a serotonin 1A receptor agonist and serotonin 2A receptor antagonist. In vitro, flibanserin demonstrated high affinity for the following 5-HT receptors:

• agonist activity at 5-HT1A
• antagonist activity at 5-HT2A, mostly in the prefrontal cortex.

Flibanserin also has moderate antagonist activities at the 5-HT2B, 5-HT2C, and dopamine D4 receptors. Flibanserin presumably acts centrally in the CNS; it has been suggested that flibanserin could rebalance neural circuitry involved in processing sexual desire by reducing serotonin activity and enhancing dopamine and epinephrine activity. The exact mechanism of how flibanserin improves sexual desire in women is unknown.

Pharmacokinetics
Flibanserin has a mean termination half-life of approximately 11 hours. It is administered once a day (50 to 100 mg) at bedtime. Steady state in healthy women was achieved after 3 days. Based on clinical observations, onset of action seems to be gradual and reaches maximum efficacy in approximately 8 weeks. Patients should discontinue the drug if no improvement is reported after 8 weeks. Flibanserin is readily absorbed from the gastrointestinal tract; however, food slows its absorption. The drug is 98% protein (mostly albumin)-bound.

Flibanserin is primarily metabolized in the liver by cytochrome P450 (CYP) 3A4 and to a lesser extent by CYP2C19. Co-administration of moderate (diltiazem, erythromycin, fluconazole, fosamprenavir, verapamil) or strong (eg, ketoconazole, clarithromycin, nefazodone, ritonavir) CYP3A4 inhibitors increases the concentration of flibanserin. This could lead to severe hypotension and syncope; therefore, co-administering flibanserin with a strong CYP3A4 inhibitor is contraindicated. Grapefruit juice is a moderate inhibitor of CYP3A4, and in a study of 26 healthy females, 240 mL of grapefruit juice increased flibanserin concentration 1.4-fold. Flibanserin is excreted though urine and feces. Flibanserin should be taken once a day at bedtime because of sedation, somnolence, and possible syncope.

Efficacy
The efficacy of flibanserin for treating HSDD was established in three 24-week, randomized, double-blind, placebo-controlled studies (Table 2). The target population in these studies was premenopausal women (mean age 36, range 19 to 55) with acquired HSDD lasting at least 6 months (mean duration, approximately 5 years). The 3 studies included 1,187 women who received flibanserin, 100 mg at bedtime, and 1,188 women who received placebo. Participants were mostly white (88.6%), and included black (9.6%) and Asian (1.5%) women. The completion rates were 69% for flibanserin and 78% for placebo. Some of the trials included arms with a lower dosage of flibanserin (25 mg and 50 mg), which are not included in this analysis.
 

As noted in the package insert, these trials each had 2 co-primary efficacy endpoints, SSEs and sexual desire:

• change from baseline to Week 24 in the number of monthly SSEs (ie, sexual intercourse, oral sex, masturbation, or genital stimulation by the partner)
• change in sexual desire from baseline to 24-week endpoint.

In Study 1 and 2, change in sexual desire from baseline to Week 24 was measured daily by using an electronic diary. Every day, patients rated their sexual desire level by answering the question, “Indicate your most intense level of sexual desire” from 0 (no desire) to 3 (strong desire). These responses were totaled over a 28-day period to yield the monthly sexual desire score, which ranged from 0 to 84. These 2 studies also used the Female Sexual Function Index (FSFI) Desire domain as a secondary endpoint.

Study 3 used the FSFI Desire domain, comprising 2 questions, as the sexual desire co-primary endpoint:

• “Over the past 4 weeks, how often did you feel sexual desire or interest?” Responses ranged from 1 (almost never or never) to 5 (almost always or always).
• “Over the past 4 weeks, how would you rate your level (degree) of sexual desire or interest?” Responses ranged from 1 (very low or none at all) to 5 (very high).

In all 3 trials, flibanserin was associated with a small, yet statistically significant, improvement in change in monthly SSEs from baseline to Week 24 compared with placebo. In Study 1 and 2, there were no statistically significant differences between flibanserin and placebo for the electronic diary sexual desire endpoint. In the third study, there was statistically significant improvement in the change in sexual desire using the FSFI Desire domain with flibanserin compared with placebo. The FSFI Desire domain findings were consistent across all 3 trials. Flibanserin was associated with a decrease in sexual distress compared with placebo in all 3 studies.

Tolerability
Flibanserin was well tolerated in the 3 clinical trials. As the FDA noted, clinical trials are conducted under widely varying conditions and therefore adverse reaction rates observed in trials of flibanserin cannot be directly compared with those reported in clinical trials of another drug and might not reflect rates observed in clinical practice.

The discontinuation rate due to adverse reactions was 13% among patients treated with flibanserin, 100 mg at bedtime, and 6% among those taking placebo. The most common side effects were somnolence, dizziness, fatigue, nausea, insomnia, and dry mouth, which appear dose-dependent. Onset of most of these adverse events was within 14 days after the start of treatment.

Although hypotension and syncope rarely were seen with flibanserin alone in clinical trials, these adverse events occurred more frequently in the morning and when taken with alcohol and with some drugs (moderate or strong CYP3A4 inhibitors), and in patients with hepatic impairment. Therefore, women who drink alcohol or take a moderate or strong inhibitor of CYP3A4—both of which are contraindicated—and those with hepatic impairment should not take flibanserin.

Flibanserin should be taken at bedtime, because the risk of hypotension and syncope is higher when flibanserin is taken in the morning and because of associated sedation and somnolence.

Unique clinical issues
Flibanserin is the first FDA-approved medication for treating HSDD. It is important to note that the drug originally was developed as an antidepressant, but failed to show efficacy. Researchers noted that the drug was more effective than placebo when patients were asked, “How strong is your sexual desire?” The focus of development then shifted to a potential treatment of HSDD.

Flibanserin was not approved at 2 previous FDA hearings, mainly because of safety concerns. For the second hearing, the manufacturer, Boehringer Ingelheim, which sold the rights to the drug to Sprout Pharmaceuticals in 2011,⁶ did not present any new efficacy data, but provided additional safety data, such as research suggesting the absence of next-day driving impairment and data related to alcohol use (the study confirming hypotension associated with alcohol abuse used a small sample, and only 2 of 25 participants were women).

Contraindications
Flibanserin is contraindicated in patients using alcohol because of an increased risk of hypotension and syncope. A patient’s alcohol use should be evaluated before administering flibanserin, and patients should be counseled about the importance of abstaining from alcohol.

Similarly, concomitant use of flibanserin with a moderate or strong inhibitor of CYP3A4 increases the concentration of flibanserin and raises the risk of hypotension and syncope. Therefore, the use of a moderate or strong inhibitor of CYP3A4 in patients taking flibanserin is contraindicated. Similarly, patients with liver impairment should not take this drug.

Strong CYP2C19 inhibitors (proton-pump inhibitors, selective serotonin reuptake inhibitors, benzodiazepines, antifungals) could increase flibanserin exposure, which may increase risk of hypotension, syncope, and CNS depression. Discuss these risks with your patients; doing so is particularly important when treating women of Chinese heritage, and some other Asian women, because 20% of these populations are genotypic CYP2C19 poor metabolizers.

Because of the increased risk of hypotension and syncope with alcohol use, flibanserin is available only through a restricted program under a Risk Evaluation and Mitigation Strategy (REMS) called the Addyi REMS Program. Flibanserin can be prescribed or dispensed only by physicians and pharmacists who watch this program’s online slide presentation and passed a comprehension test.^a

Pregnant women should not take flibanserin because the effect on the fetus is unknown. Also, because the interaction with some oral contraceptives is unknown, patients should be cautioned about unwanted pregnancy. Women who are breastfeeding also should avoid using flibanserin because it is not known whether the drug is excreted in breast milk.

Women taking flibanserin also should avoid grapefruit juice, which increases flibanserin levels, and avoid using herbal products, resveratrol, and some over-the-counter drugs such as cimetidine. Women who have a depressive disorder also should avoid using flibanserin because their low sexual desire is more likely due to depression, which is not a therapeutic target for the drug.

Dosing
Flibanserin is provided in 100-mg film-coated tablets. It should be taken once a day at bedtime; titration is unnecessary. Length of treatment has not been determined, but it is recommended that patients stop flibanserin if they do not experience any benefit after 8 weeks. Although there is no guidance in the prescribing information, the medication probably could be stopped without tapering because withdrawal effects have not been observed.

Bottom Line
Flibanserin is FDA-approved for treating generalized, acquired hypoactive sexual desire disorder in premenopausal women. In clinical trials, the drug increased the number of satisfying sexual events and sexual desire, as measured by a diary and rating scales. Alcohol use and use of any moderate or strong inhibitor of cytochrome P450 3A4 are contraindicated in patients taking flibanserin because of an increased risk of hypotension and syncope.

Flibanserin, FDA-approved in August 2015, is the first medication approved to treat acquired, generalized hypoactive sexual desire disorder (HSDD) in premenopausal women (Table 1). In clinical trials,^1-4 the drug has shown modest efficacy in improving symptoms of low sexual desire (number of satisfying sexual events [SSEs], sexual desire, and overall sexual function). Flibanserin is not indicated to enhance sexual performance, for HSDD in postmenopausal women, or in men.
 

Clinical implications
Flibanserin could help premenopausal women who have distressing low sexual desire, which must be acquired and generalized:

• “Acquired low sexual desire” means that a patient had an adequate sexual desire that decreased or ceased for an unknown reason.
• “Generalized low sexual desire” means that lack of sexual desire occurs all the time and in all situations, not only with a certain partner or in some situations.

Women taking flibanserin could experience gradually increased sexual desire, increase in SSEs, and decrease of sexual distress. Flibanserin is indicated for long-term use; however, it should be discontinued after 8 weeks if the patient does not report any improvement in symptoms.

The number needed to treat with flibanserin likely would be rather large, but it is not available because of complex outcome measures in clinical trials. Flibanserin was not approved at 2 previous FDA committee hearings—mainly because of safety issues but also because of concerns about efficacy. For example, during the 2013 FDA hearing, the results presented showed statistically significant, but numerically small, treatment differences at 24 weeks compared with placebo. In an FDA responder analysis of the Phase-III trials, after accounting for the placebo effect, approximately 8% to 13% women were at least “much improved” on at least 1 of the primary outcomes.⁵

Flibanserin is not indicated for women whose sexual desire is due to (1) coexisting medical or psychiatric condition, (2) effects of medication or substance abuse, or (3) a relationship problem. It is unknown whether supplemental treatment would help these patients; however, it seems reasonable that combining flibanserin with psychosocial treatment, such as sex therapy or individual therapy, could be beneficial because it may be difficult to disentangle sexual dysfunction and relationship issues—2 problems that often are interwoven.

How it works
Flibanserin is a serotonin 1A receptor agonist and serotonin 2A receptor antagonist. In vitro, flibanserin demonstrated high affinity for the following 5-HT receptors:

• agonist activity at 5-HT1A
• antagonist activity at 5-HT2A, mostly in the prefrontal cortex.

Flibanserin also has moderate antagonist activities at the 5-HT2B, 5-HT2C, and dopamine D4 receptors. Flibanserin presumably acts centrally in the CNS; it has been suggested that flibanserin could rebalance neural circuitry involved in processing sexual desire by reducing serotonin activity and enhancing dopamine and epinephrine activity. The exact mechanism of how flibanserin improves sexual desire in women is unknown.

Pharmacokinetics
Flibanserin has a mean termination half-life of approximately 11 hours. It is administered once a day (50 to 100 mg) at bedtime. Steady state in healthy women was achieved after 3 days. Based on clinical observations, onset of action seems to be gradual and reaches maximum efficacy in approximately 8 weeks. Patients should discontinue the drug if no improvement is reported after 8 weeks. Flibanserin is readily absorbed from the gastrointestinal tract; however, food slows its absorption. The drug is 98% protein (mostly albumin)-bound.

Flibanserin is primarily metabolized in the liver by cytochrome P450 (CYP) 3A4 and to a lesser extent by CYP2C19. Co-administration of moderate (diltiazem, erythromycin, fluconazole, fosamprenavir, verapamil) or strong (eg, ketoconazole, clarithromycin, nefazodone, ritonavir) CYP3A4 inhibitors increases the concentration of flibanserin. This could lead to severe hypotension and syncope; therefore, co-administering flibanserin with a strong CYP3A4 inhibitor is contraindicated. Grapefruit juice is a moderate inhibitor of CYP3A4, and in a study of 26 healthy females, 240 mL of grapefruit juice increased flibanserin concentration 1.4-fold. Flibanserin is excreted though urine and feces. Flibanserin should be taken once a day at bedtime because of sedation, somnolence, and possible syncope.

Efficacy
The efficacy of flibanserin for treating HSDD was established in three 24-week, randomized, double-blind, placebo-controlled studies (Table 2). The target population in these studies was premenopausal women (mean age 36, range 19 to 55) with acquired HSDD lasting at least 6 months (mean duration, approximately 5 years). The 3 studies included 1,187 women who received flibanserin, 100 mg at bedtime, and 1,188 women who received placebo. Participants were mostly white (88.6%), and included black (9.6%) and Asian (1.5%) women. The completion rates were 69% for flibanserin and 78% for placebo. Some of the trials included arms with a lower dosage of flibanserin (25 mg and 50 mg), which are not included in this analysis.
 

As noted in the package insert, these trials each had 2 co-primary efficacy endpoints, SSEs and sexual desire:

• change from baseline to Week 24 in the number of monthly SSEs (ie, sexual intercourse, oral sex, masturbation, or genital stimulation by the partner)
• change in sexual desire from baseline to 24-week endpoint.

In Study 1 and 2, change in sexual desire from baseline to Week 24 was measured daily by using an electronic diary. Every day, patients rated their sexual desire level by answering the question, “Indicate your most intense level of sexual desire” from 0 (no desire) to 3 (strong desire). These responses were totaled over a 28-day period to yield the monthly sexual desire score, which ranged from 0 to 84. These 2 studies also used the Female Sexual Function Index (FSFI) Desire domain as a secondary endpoint.

Study 3 used the FSFI Desire domain, comprising 2 questions, as the sexual desire co-primary endpoint:

• “Over the past 4 weeks, how often did you feel sexual desire or interest?” Responses ranged from 1 (almost never or never) to 5 (almost always or always).
• “Over the past 4 weeks, how would you rate your level (degree) of sexual desire or interest?” Responses ranged from 1 (very low or none at all) to 5 (very high).

In all 3 trials, flibanserin was associated with a small, yet statistically significant, improvement in change in monthly SSEs from baseline to Week 24 compared with placebo. In Study 1 and 2, there were no statistically significant differences between flibanserin and placebo for the electronic diary sexual desire endpoint. In the third study, there was statistically significant improvement in the change in sexual desire using the FSFI Desire domain with flibanserin compared with placebo. The FSFI Desire domain findings were consistent across all 3 trials. Flibanserin was associated with a decrease in sexual distress compared with placebo in all 3 studies.

Tolerability
Flibanserin was well tolerated in the 3 clinical trials. As the FDA noted, clinical trials are conducted under widely varying conditions and therefore adverse reaction rates observed in trials of flibanserin cannot be directly compared with those reported in clinical trials of another drug and might not reflect rates observed in clinical practice.

The discontinuation rate due to adverse reactions was 13% among patients treated with flibanserin, 100 mg at bedtime, and 6% among those taking placebo. The most common side effects were somnolence, dizziness, fatigue, nausea, insomnia, and dry mouth, which appear dose-dependent. Onset of most of these adverse events was within 14 days after the start of treatment.

Although hypotension and syncope rarely were seen with flibanserin alone in clinical trials, these adverse events occurred more frequently in the morning and when taken with alcohol and with some drugs (moderate or strong CYP3A4 inhibitors), and in patients with hepatic impairment. Therefore, women who drink alcohol or take a moderate or strong inhibitor of CYP3A4—both of which are contraindicated—and those with hepatic impairment should not take flibanserin.

Flibanserin should be taken at bedtime, because the risk of hypotension and syncope is higher when flibanserin is taken in the morning and because of associated sedation and somnolence.

Unique clinical issues
Flibanserin is the first FDA-approved medication for treating HSDD. It is important to note that the drug originally was developed as an antidepressant, but failed to show efficacy. Researchers noted that the drug was more effective than placebo when patients were asked, “How strong is your sexual desire?” The focus of development then shifted to a potential treatment of HSDD.

Flibanserin was not approved at 2 previous FDA hearings, mainly because of safety concerns. For the second hearing, the manufacturer, Boehringer Ingelheim, which sold the rights to the drug to Sprout Pharmaceuticals in 2011,⁶ did not present any new efficacy data, but provided additional safety data, such as research suggesting the absence of next-day driving impairment and data related to alcohol use (the study confirming hypotension associated with alcohol abuse used a small sample, and only 2 of 25 participants were women).

Contraindications
Flibanserin is contraindicated in patients using alcohol because of an increased risk of hypotension and syncope. A patient’s alcohol use should be evaluated before administering flibanserin, and patients should be counseled about the importance of abstaining from alcohol.

Similarly, concomitant use of flibanserin with a moderate or strong inhibitor of CYP3A4 increases the concentration of flibanserin and raises the risk of hypotension and syncope. Therefore, the use of a moderate or strong inhibitor of CYP3A4 in patients taking flibanserin is contraindicated. Similarly, patients with liver impairment should not take this drug.

Strong CYP2C19 inhibitors (proton-pump inhibitors, selective serotonin reuptake inhibitors, benzodiazepines, antifungals) could increase flibanserin exposure, which may increase risk of hypotension, syncope, and CNS depression. Discuss these risks with your patients; doing so is particularly important when treating women of Chinese heritage, and some other Asian women, because 20% of these populations are genotypic CYP2C19 poor metabolizers.

Because of the increased risk of hypotension and syncope with alcohol use, flibanserin is available only through a restricted program under a Risk Evaluation and Mitigation Strategy (REMS) called the Addyi REMS Program. Flibanserin can be prescribed or dispensed only by physicians and pharmacists who watch this program’s online slide presentation and passed a comprehension test.^a

Pregnant women should not take flibanserin because the effect on the fetus is unknown. Also, because the interaction with some oral contraceptives is unknown, patients should be cautioned about unwanted pregnancy. Women who are breastfeeding also should avoid using flibanserin because it is not known whether the drug is excreted in breast milk.

Women taking flibanserin also should avoid grapefruit juice, which increases flibanserin levels, and avoid using herbal products, resveratrol, and some over-the-counter drugs such as cimetidine. Women who have a depressive disorder also should avoid using flibanserin because their low sexual desire is more likely due to depression, which is not a therapeutic target for the drug.

Dosing
Flibanserin is provided in 100-mg film-coated tablets. It should be taken once a day at bedtime; titration is unnecessary. Length of treatment has not been determined, but it is recommended that patients stop flibanserin if they do not experience any benefit after 8 weeks. Although there is no guidance in the prescribing information, the medication probably could be stopped without tapering because withdrawal effects have not been observed.

Bottom Line
Flibanserin is FDA-approved for treating generalized, acquired hypoactive sexual desire disorder in premenopausal women. In clinical trials, the drug increased the number of satisfying sexual events and sexual desire, as measured by a diary and rating scales. Alcohol use and use of any moderate or strong inhibitor of cytochrome P450 3A4 are contraindicated in patients taking flibanserin because of an increased risk of hypotension and syncope.

Flibanserin, FDA-approved in August 2015, is the first medication approved to treat acquired, generalized hypoactive sexual desire disorder (HSDD) in premenopausal women (Table 1). In clinical trials,^1-4 the drug has shown modest efficacy in improving symptoms of low sexual desire (number of satisfying sexual events [SSEs], sexual desire, and overall sexual function). Flibanserin is not indicated to enhance sexual performance, for HSDD in postmenopausal women, or in men.
 

Clinical implications
Flibanserin could help premenopausal women who have distressing low sexual desire, which must be acquired and generalized:

• “Acquired low sexual desire” means that a patient had an adequate sexual desire that decreased or ceased for an unknown reason.
• “Generalized low sexual desire” means that lack of sexual desire occurs all the time and in all situations, not only with a certain partner or in some situations.

Women taking flibanserin could experience gradually increased sexual desire, increase in SSEs, and decrease of sexual distress. Flibanserin is indicated for long-term use; however, it should be discontinued after 8 weeks if the patient does not report any improvement in symptoms.

The number needed to treat with flibanserin likely would be rather large, but it is not available because of complex outcome measures in clinical trials. Flibanserin was not approved at 2 previous FDA committee hearings—mainly because of safety issues but also because of concerns about efficacy. For example, during the 2013 FDA hearing, the results presented showed statistically significant, but numerically small, treatment differences at 24 weeks compared with placebo. In an FDA responder analysis of the Phase-III trials, after accounting for the placebo effect, approximately 8% to 13% women were at least “much improved” on at least 1 of the primary outcomes.⁵

Flibanserin is not indicated for women whose sexual desire is due to (1) coexisting medical or psychiatric condition, (2) effects of medication or substance abuse, or (3) a relationship problem. It is unknown whether supplemental treatment would help these patients; however, it seems reasonable that combining flibanserin with psychosocial treatment, such as sex therapy or individual therapy, could be beneficial because it may be difficult to disentangle sexual dysfunction and relationship issues—2 problems that often are interwoven.

How it works
Flibanserin is a serotonin 1A receptor agonist and serotonin 2A receptor antagonist. In vitro, flibanserin demonstrated high affinity for the following 5-HT receptors:

• agonist activity at 5-HT1A
• antagonist activity at 5-HT2A, mostly in the prefrontal cortex.

Flibanserin also has moderate antagonist activities at the 5-HT2B, 5-HT2C, and dopamine D4 receptors. Flibanserin presumably acts centrally in the CNS; it has been suggested that flibanserin could rebalance neural circuitry involved in processing sexual desire by reducing serotonin activity and enhancing dopamine and epinephrine activity. The exact mechanism of how flibanserin improves sexual desire in women is unknown.

Pharmacokinetics
Flibanserin has a mean termination half-life of approximately 11 hours. It is administered once a day (50 to 100 mg) at bedtime. Steady state in healthy women was achieved after 3 days. Based on clinical observations, onset of action seems to be gradual and reaches maximum efficacy in approximately 8 weeks. Patients should discontinue the drug if no improvement is reported after 8 weeks. Flibanserin is readily absorbed from the gastrointestinal tract; however, food slows its absorption. The drug is 98% protein (mostly albumin)-bound.

Flibanserin is primarily metabolized in the liver by cytochrome P450 (CYP) 3A4 and to a lesser extent by CYP2C19. Co-administration of moderate (diltiazem, erythromycin, fluconazole, fosamprenavir, verapamil) or strong (eg, ketoconazole, clarithromycin, nefazodone, ritonavir) CYP3A4 inhibitors increases the concentration of flibanserin. This could lead to severe hypotension and syncope; therefore, co-administering flibanserin with a strong CYP3A4 inhibitor is contraindicated. Grapefruit juice is a moderate inhibitor of CYP3A4, and in a study of 26 healthy females, 240 mL of grapefruit juice increased flibanserin concentration 1.4-fold. Flibanserin is excreted though urine and feces. Flibanserin should be taken once a day at bedtime because of sedation, somnolence, and possible syncope.

Efficacy
The efficacy of flibanserin for treating HSDD was established in three 24-week, randomized, double-blind, placebo-controlled studies (Table 2). The target population in these studies was premenopausal women (mean age 36, range 19 to 55) with acquired HSDD lasting at least 6 months (mean duration, approximately 5 years). The 3 studies included 1,187 women who received flibanserin, 100 mg at bedtime, and 1,188 women who received placebo. Participants were mostly white (88.6%), and included black (9.6%) and Asian (1.5%) women. The completion rates were 69% for flibanserin and 78% for placebo. Some of the trials included arms with a lower dosage of flibanserin (25 mg and 50 mg), which are not included in this analysis.
 

As noted in the package insert, these trials each had 2 co-primary efficacy endpoints, SSEs and sexual desire:

• change from baseline to Week 24 in the number of monthly SSEs (ie, sexual intercourse, oral sex, masturbation, or genital stimulation by the partner)
• change in sexual desire from baseline to 24-week endpoint.

In Study 1 and 2, change in sexual desire from baseline to Week 24 was measured daily by using an electronic diary. Every day, patients rated their sexual desire level by answering the question, “Indicate your most intense level of sexual desire” from 0 (no desire) to 3 (strong desire). These responses were totaled over a 28-day period to yield the monthly sexual desire score, which ranged from 0 to 84. These 2 studies also used the Female Sexual Function Index (FSFI) Desire domain as a secondary endpoint.

Study 3 used the FSFI Desire domain, comprising 2 questions, as the sexual desire co-primary endpoint:

• “Over the past 4 weeks, how often did you feel sexual desire or interest?” Responses ranged from 1 (almost never or never) to 5 (almost always or always).
• “Over the past 4 weeks, how would you rate your level (degree) of sexual desire or interest?” Responses ranged from 1 (very low or none at all) to 5 (very high).

In all 3 trials, flibanserin was associated with a small, yet statistically significant, improvement in change in monthly SSEs from baseline to Week 24 compared with placebo. In Study 1 and 2, there were no statistically significant differences between flibanserin and placebo for the electronic diary sexual desire endpoint. In the third study, there was statistically significant improvement in the change in sexual desire using the FSFI Desire domain with flibanserin compared with placebo. The FSFI Desire domain findings were consistent across all 3 trials. Flibanserin was associated with a decrease in sexual distress compared with placebo in all 3 studies.

Tolerability
Flibanserin was well tolerated in the 3 clinical trials. As the FDA noted, clinical trials are conducted under widely varying conditions and therefore adverse reaction rates observed in trials of flibanserin cannot be directly compared with those reported in clinical trials of another drug and might not reflect rates observed in clinical practice.

The discontinuation rate due to adverse reactions was 13% among patients treated with flibanserin, 100 mg at bedtime, and 6% among those taking placebo. The most common side effects were somnolence, dizziness, fatigue, nausea, insomnia, and dry mouth, which appear dose-dependent. Onset of most of these adverse events was within 14 days after the start of treatment.

Although hypotension and syncope rarely were seen with flibanserin alone in clinical trials, these adverse events occurred more frequently in the morning and when taken with alcohol and with some drugs (moderate or strong CYP3A4 inhibitors), and in patients with hepatic impairment. Therefore, women who drink alcohol or take a moderate or strong inhibitor of CYP3A4—both of which are contraindicated—and those with hepatic impairment should not take flibanserin.

Flibanserin should be taken at bedtime, because the risk of hypotension and syncope is higher when flibanserin is taken in the morning and because of associated sedation and somnolence.

Unique clinical issues
Flibanserin is the first FDA-approved medication for treating HSDD. It is important to note that the drug originally was developed as an antidepressant, but failed to show efficacy. Researchers noted that the drug was more effective than placebo when patients were asked, “How strong is your sexual desire?” The focus of development then shifted to a potential treatment of HSDD.

Flibanserin was not approved at 2 previous FDA hearings, mainly because of safety concerns. For the second hearing, the manufacturer, Boehringer Ingelheim, which sold the rights to the drug to Sprout Pharmaceuticals in 2011,⁶ did not present any new efficacy data, but provided additional safety data, such as research suggesting the absence of next-day driving impairment and data related to alcohol use (the study confirming hypotension associated with alcohol abuse used a small sample, and only 2 of 25 participants were women).

Contraindications
Flibanserin is contraindicated in patients using alcohol because of an increased risk of hypotension and syncope. A patient’s alcohol use should be evaluated before administering flibanserin, and patients should be counseled about the importance of abstaining from alcohol.

Similarly, concomitant use of flibanserin with a moderate or strong inhibitor of CYP3A4 increases the concentration of flibanserin and raises the risk of hypotension and syncope. Therefore, the use of a moderate or strong inhibitor of CYP3A4 in patients taking flibanserin is contraindicated. Similarly, patients with liver impairment should not take this drug.

Strong CYP2C19 inhibitors (proton-pump inhibitors, selective serotonin reuptake inhibitors, benzodiazepines, antifungals) could increase flibanserin exposure, which may increase risk of hypotension, syncope, and CNS depression. Discuss these risks with your patients; doing so is particularly important when treating women of Chinese heritage, and some other Asian women, because 20% of these populations are genotypic CYP2C19 poor metabolizers.

Because of the increased risk of hypotension and syncope with alcohol use, flibanserin is available only through a restricted program under a Risk Evaluation and Mitigation Strategy (REMS) called the Addyi REMS Program. Flibanserin can be prescribed or dispensed only by physicians and pharmacists who watch this program’s online slide presentation and passed a comprehension test.^a

Pregnant women should not take flibanserin because the effect on the fetus is unknown. Also, because the interaction with some oral contraceptives is unknown, patients should be cautioned about unwanted pregnancy. Women who are breastfeeding also should avoid using flibanserin because it is not known whether the drug is excreted in breast milk.

Women taking flibanserin also should avoid grapefruit juice, which increases flibanserin levels, and avoid using herbal products, resveratrol, and some over-the-counter drugs such as cimetidine. Women who have a depressive disorder also should avoid using flibanserin because their low sexual desire is more likely due to depression, which is not a therapeutic target for the drug.

Dosing
Flibanserin is provided in 100-mg film-coated tablets. It should be taken once a day at bedtime; titration is unnecessary. Length of treatment has not been determined, but it is recommended that patients stop flibanserin if they do not experience any benefit after 8 weeks. Although there is no guidance in the prescribing information, the medication probably could be stopped without tapering because withdrawal effects have not been observed.

Bottom Line
Flibanserin is FDA-approved for treating generalized, acquired hypoactive sexual desire disorder in premenopausal women. In clinical trials, the drug increased the number of satisfying sexual events and sexual desire, as measured by a diary and rating scales. Alcohol use and use of any moderate or strong inhibitor of cytochrome P450 3A4 are contraindicated in patients taking flibanserin because of an increased risk of hypotension and syncope.