Evidence-Based Reviews

FIRST OF 2 PARTS

A thorough evaluation can identify modifiable factors and guide treatment

Sleep problems are common among children and adolescents,¹ with prevalence rates of 25% to 40%.^2-4 Young children most commonly exhibit what is referred to as bedtime problems and night wakenings, whereas children in middle childhood (age 4 to 12) through adolescence (age 13 to 17) report insomnia. For many children, these problems persist.³ Insufficient sleep in children and adolescents worsens inattention, daytime fatigue, and cognitive and behavioral deficits.⁵ Assessment and treatment of sleep problems in children and adolescents is critical because poor sleep among youth increases the risk for depression, self-harm, and suicide,^6,7 increases family stress, and decreases parental well-being.¹

This 2-part article describes the assessment, diagnosis, and treatment of sleep problems among children and adolescents. In part 1, we focus on:

• sleep architecture (circadian rhythms, stages of sleep)
• sleep in healthy youth (age 6 to 17) and those with attention-deficit/hyperactivity disorder (ADHD), depressive disorders, and anxiety
• how to assess sleep, and the differential diagnosis of behavioral sleep problems in pediatric patients.

In Part 2, we will cover psychotherapeutic and psychopharmacologic interventions for youth with insomnia, and describe an effective approach to consultation with pediatric sleep medicine specialists.

How much sleep do children and adolescents need?

Throughout their development, children spend 40% to 50% of their time asleep. Sleep schedules are based on circadian rhythms, which are physical, mental, and behavioral changes that follow an approximately 24-hour cycle. Human circadian rhythm varies between 24 and 25 hours and is vital in determining our sleep patterns. Exposure to sunlight drives our circadian rhythm, sending signals to our bodies to “turn on” melatonin production at night (ie, 9 pm melatonin secretion starts) and “turn off” melatonin secretion in the morning (ie, 7:30 am) in adults. Exposure to sunlight also drives circadian rhythms for children, and melatonin secretion production occurs earlier in the evening for children. Families are encouraged to expose children to bright light in the morning by taking an early walk, eating breakfast in a sunny room, or having adolescents open up the window shades rather than keeping their eyes shielded from the sun.² How sleepy or alert a child could be is determined by the duration and quality of a child’s previous sleep, and how long they have been awake (“sleep drive”).² The 3 states of sleep architecture are wake; non-rapid eye movement sleep; and rapid eye movement sleep (“dreaming” sleep) (Box²).

Box

Sleep needs also change predictably throughout the lifespan. The National Sleep Foundation guidelines for sleep duration provide clinicians and parents with a range of recommended sleep for each stage of development. Infants require 14 to 17 hours of sleep, whereas adolescents need 8 to 10 hours by age 14 to 17.⁸ The key for clinicians is to determine if the child is within the recommended range, and how they are functioning on the number of hours of sleep they report. This allows for variation in how much sleep an individual child might need while acknowledging that some children within a specific age group might need more or less sleep than other children of the same age.

Sleep in healthy youth: Middle childhood

School-age children (age 6 to 12) typically need 9 to 10 hours of sleep over a 24-hour period.² This developmental period is especially important for children to develop healthy sleep habits; however, developmentally appropriate cognitive and social/emotional factors might interfere with the quality and quantity of sleep. Middle childhood is a time when children can understand the dangers of the outside world (ie, violence, health problems) and resulting anxiety can disrupt sleep. Parents usually are less involved in bedtime as children approach adolescence, which leads to later bedtimes. At this stage, many children begin to take on more serious roles in their academics and extracurricular activities, peer relationships become more important, and use of electronics (eg, television, video games, internet, and handheld devices) increases—all of which compete with sleep.⁹ Frequent sleep issues during middle childhood include:

• irregular sleep-wake schedules
• later bedtimes
• decreased nighttime sleep
• increased caffeine intake
• reduced parental presence at bedtime
• daytime sleepiness.³

In school-age children, regular napping, falling asleep during short car rides, and daytime fatigue at school or home are cause for concern. When these symptoms are present, an evaluation is warranted.

Sleep in healthy youth: Adolescence

The National Sleep Foundation recommends adolescents obtain 8 to 10 hours of sleep per night; for some adolescents, as much as 11 hours of sleep per night might be appropriate.⁸ However, this contrasts with findings from the National Sleep Foundation’s Sleep in America Poll, which revealed that 75% of 12th graders report <8 hours of sleep nightly.¹⁰ Many adolescents experience delayed sleep phase syndrome or delayed sleep-wake phase disorder, which involves a persistent phase shift of >2 hours in the sleep-wake schedule that conflicts with the adolescent’s school, work, or lifestyle demands.¹¹ Such circadian rhythm disorders typically result from a poor match between the sleep-wake schedule and the demands of the adolescent’s life, or a failure to synchronize their internal clock with a 24-hour circadian clock.¹² Children typically become tired after sunset, but puberty is associated with reduced slow-wave sleep and changes in circadian rhythms. As a result, a 3-hour delay (delayed phase preference) is common among adolescents. At approximately age 20, people start to become tired after sunset and awaken earlier in the morning—a pattern driven by sunlight and the timing of melatonin release that will remain stable until the sixth decade of life.

Continue to: Effects of chronic sleep deprivation...

Most older studies of sleep loss examined the impact of total sleep loss (sleep deprivation) rather than the effect of partial sleep loss or sleep restriction, a more commonly experienced phenomenon. More recent research shows that a cumulative sleep deficit could cause the body to override voluntary wakefulness and a sleep-deprived individual can experience brief “microsleeps” where they are unaware and lose attention/wakefulness for several seconds.² This can be deadly if a sleep-deprived adolescent experiences microsleeps while driving.¹³

There is a well-studied correlation between chronic sleep deprivation and increased body mass index in children.¹⁴ This might be caused by reduction in physical activity as well as alterations in the “hunger hormones”—ghrelin and leptin—that have been observed with sleep deprivation.^15-17 Other studies have noted decreased glucose tolerance, reduced insulin sensitivity, and catecholamine and cortisol secretion abnormalities, which place children at higher risk for metabolic syndrome and hypertension.^13,18 Sleep deprivation also is associated with mood and anxiety disorders and is an independent risk factor for substance use and suicidal ideation among adolescents.¹⁹ Sleep deprivation increases impairments in impulse control, concentration, and attention, which could be especially problematic in school-age children.

How sleep is assessed

The sleep history is the first step in evaluating a child or adolescent for a sleep disorder. The sleep history includes exploring the chief complaint, sleep patterns and schedules, bedtime routines, and nocturnal and daytime behaviors (Table).

Chief complaint

Behavioral sleep specialists will assess the primary problem with everyone involved in the child’s bedtime.²⁰ This might include parents (custodial and noncustodial), grandparents, or stepparents as well as the child/adolescent. This important step can reveal a sleep disorder or an inappropriately early bedtime relative to the child’s development. During this assessment, ask detailed questions about how long the sleep problem has persisted, the frequency of sleep problems, and any precipitating stressors. Parents and caregivers can review strategies they have tried, and for how long and to what extent interventions were implemented consistently to result in change.

Sleep patterns and schedules

Review the child/adolescent’s typical sleep patterns and behaviors. Ask parents and caregivers, as well as the patient, about general sleep schedules for the past few weeks or a typical 2-week time period.² A behavioral assessment of sleep should include asking families about how the child/adolescent sleeps during the week and over the weekend, and if school-year sleep differs from summer or holiday sleep schedules. These questions can illuminate how long a sleep problem has been occurring and what sleep habits might be contributing to the problem. Bedtime

Determine if there is a set bedtime or if the child goes to bed when they wish. It is important to ascertain if the bedtime is age-appropriate, if weekday and weekend bedtimes differ, and to what extent extracurricular activities or school demands impact bedtime. Assess the consistency of the bedtime, the nature of bedtime routines (eg, is the child engaging in stimulating activities before bed), where the bedtime routine occurs (eg, sibling’s room, parents’ room, child’s room), and what role (if any) electronic devices play.²

Nocturnal behaviors

Assessment should include a series of questions and age-specific questionnaires to focus on what behaviors occur at night, including awakenings. Parents should be asked how frequent night awakenings occur, how long arousals last, and how the child signals for the parent (eg, calling out, climbing into parents’ bed).² Additionally, ask how parents respond and what is required to help the child fall back asleep (eg, rocking, soothing, feeding). The presence of nightmares, night terrors, parasomnias, and sleep-related breathing disorders also must be assessed.²⁰

Daytime behaviors

A sleep history should include assessment of daytime functioning, including daytime sleepiness, fatigue, morning waking, and functioning during school, extracurriculars, and homework. For children and teens, falling asleep in the car, while in school, or during passive activities (meals, conversation) suggests insufficient sleep, sleep disruption, or excessive daytime sleepiness.²

Continue to: Sleep disruption in youth with psychiatric disorders...

Sleep disruption in youth with psychiatric disorders

Disordered sleep is common across psychiatric disorders. The National Comorbidity Survey Adolescent Supplement—a nationally representative cross-sectional survey of adolescents (N = 10,123)—found that a later weeknight bedtime, shorter weeknight sleep duration, and greater weekend bedtime delay increased the risk of developing a mood, anxiety, or substance use (including nicotine) disorder, and suicidality. These risk factors also were associated with lower “perceived mental and physical health.”²¹ Clinicians should routinely obtain a sleep history in children and adolescents with these disorders. Consider using the sleep screening tool BEARS:

• Bedtime issues
• Excessive daytime sleepiness
• Awakenings
• Regularity and duration of sleep
• Snoring.

ADHD

Up to one-half of children and adolescents with ADHD experience sleep problems,^22,23 including delayed sleep onset, bedtime resistance, daytime fatigue, and feeling groggy in the morning beyond what is typical (>20 minutes). Pharmacotherapy for ADHD contributes to sleep disturbances^24,25 while sleep deprivation exacerbates inattention and hyperactivity. In youth with ADHD, restless leg syndrome, periodic limb movement disorder, and sleep-disordered breathing disorder are more common than in the general population.

Depressive disorders

Up to three-quarters of depressed children and 90% of depressed adolescents report sleep disturbances, including initial, middle, and terminal insomnia as well as hypersomnia.²⁶ Disrupted sleep in pediatric patients with major depressive disorder could be moderated by the patient’s age, with depressive symptoms more common among adolescents (age 12 to 17) than among younger children (age 6 to 11).²⁷ Successful treatment of depression fails to relieve dyssomnia in 10% of children. Sleep problems that persist after successfully treating a depressive episode could increase the risk of another depressive episode.²⁸

Anxiety disorders

Sleep problems are common among children and adolescents with anxiety disorders.²⁹ Longitudinal data from >900 children found that symptoms of sleep disturbance in early childhood were correlated with experiencing an anxiety disorder 20 years later.³⁰ Fears related to the dark or monsters under the bed that are developmentally appropriate for younger children may interfere with sleep. However, in anxious children, fears might also be related to separation, sleeping alone, worry about the loss of a loved one, concerns about personal safety, fear of frightening dreams, or concerns about academics and social relationships. Anxious individuals ruminate about their worries, and this might be especially true for children at bedtime, when there are limited distractions from ruminative fears.³¹ Bedtime resistance, parental involvement in bedtime rituals, and cultural factors related to sleep also could play a role for children with anxiety symptoms and sleep problems.

Having an anxiety disorder is significantly associated with an increased risk of insomnia; however, 73% of the time anxiety symptoms precede an insomnia diagnosis.²⁹ Sleep problems and anxiety symptoms might have a reciprocal influence on one another; tiredness that results from sleep problems could exacerbate anxiety, which further worsens sleep problems.

A large body of research on sleep and anxiety reveals that abuse or exposure to trauma significantly affects sleep.³¹ Common sleep problems for children with posttraumatic stress disorder include difficulty falling asleep, maintaining sleep, and parasomnias of bedwetting and nightmares.^31,32 Compared with depressed and non-abused children, those with history of abuse have prolonged sleep latency, decreased sleep efficiency, and higher levels of activity during the night.³³ In addition to the relationship between anxiety disorders and sleep disorders, many of the selective serotonin reuptake inhibitors—which are the first-line pharmacotherapy for pediatric anxiety disorders^34,35—could affect sleep in anxious youth.³⁶

A bridge to treatment

A thorough assessment can help identify modifiable factors and guide treatment selections. In Part 2 of this article, we will describe healthy sleep practices, cognitive-behavioral therapy for insomnia, when pharmacotherapy might be indicated, and the evidence supporting several medications commonly used to treat pediatric insomnia. We also will discuss factors to consider when seeking consultation with a pediatric behavioral sleep specialist.

FIRST OF 2 PARTS

A thorough evaluation can identify modifiable factors and guide treatment

Sleep problems are common among children and adolescents,¹ with prevalence rates of 25% to 40%.^2-4 Young children most commonly exhibit what is referred to as bedtime problems and night wakenings, whereas children in middle childhood (age 4 to 12) through adolescence (age 13 to 17) report insomnia. For many children, these problems persist.³ Insufficient sleep in children and adolescents worsens inattention, daytime fatigue, and cognitive and behavioral deficits.⁵ Assessment and treatment of sleep problems in children and adolescents is critical because poor sleep among youth increases the risk for depression, self-harm, and suicide,^6,7 increases family stress, and decreases parental well-being.¹

This 2-part article describes the assessment, diagnosis, and treatment of sleep problems among children and adolescents. In part 1, we focus on:

• sleep architecture (circadian rhythms, stages of sleep)
• sleep in healthy youth (age 6 to 17) and those with attention-deficit/hyperactivity disorder (ADHD), depressive disorders, and anxiety
• how to assess sleep, and the differential diagnosis of behavioral sleep problems in pediatric patients.

In Part 2, we will cover psychotherapeutic and psychopharmacologic interventions for youth with insomnia, and describe an effective approach to consultation with pediatric sleep medicine specialists.

How much sleep do children and adolescents need?

Throughout their development, children spend 40% to 50% of their time asleep. Sleep schedules are based on circadian rhythms, which are physical, mental, and behavioral changes that follow an approximately 24-hour cycle. Human circadian rhythm varies between 24 and 25 hours and is vital in determining our sleep patterns. Exposure to sunlight drives our circadian rhythm, sending signals to our bodies to “turn on” melatonin production at night (ie, 9 pm melatonin secretion starts) and “turn off” melatonin secretion in the morning (ie, 7:30 am) in adults. Exposure to sunlight also drives circadian rhythms for children, and melatonin secretion production occurs earlier in the evening for children. Families are encouraged to expose children to bright light in the morning by taking an early walk, eating breakfast in a sunny room, or having adolescents open up the window shades rather than keeping their eyes shielded from the sun.² How sleepy or alert a child could be is determined by the duration and quality of a child’s previous sleep, and how long they have been awake (“sleep drive”).² The 3 states of sleep architecture are wake; non-rapid eye movement sleep; and rapid eye movement sleep (“dreaming” sleep) (Box²).

Box

Sleep needs also change predictably throughout the lifespan. The National Sleep Foundation guidelines for sleep duration provide clinicians and parents with a range of recommended sleep for each stage of development. Infants require 14 to 17 hours of sleep, whereas adolescents need 8 to 10 hours by age 14 to 17.⁸ The key for clinicians is to determine if the child is within the recommended range, and how they are functioning on the number of hours of sleep they report. This allows for variation in how much sleep an individual child might need while acknowledging that some children within a specific age group might need more or less sleep than other children of the same age.

Sleep in healthy youth: Middle childhood

School-age children (age 6 to 12) typically need 9 to 10 hours of sleep over a 24-hour period.² This developmental period is especially important for children to develop healthy sleep habits; however, developmentally appropriate cognitive and social/emotional factors might interfere with the quality and quantity of sleep. Middle childhood is a time when children can understand the dangers of the outside world (ie, violence, health problems) and resulting anxiety can disrupt sleep. Parents usually are less involved in bedtime as children approach adolescence, which leads to later bedtimes. At this stage, many children begin to take on more serious roles in their academics and extracurricular activities, peer relationships become more important, and use of electronics (eg, television, video games, internet, and handheld devices) increases—all of which compete with sleep.⁹ Frequent sleep issues during middle childhood include:

• irregular sleep-wake schedules
• later bedtimes
• decreased nighttime sleep
• increased caffeine intake
• reduced parental presence at bedtime
• daytime sleepiness.³

In school-age children, regular napping, falling asleep during short car rides, and daytime fatigue at school or home are cause for concern. When these symptoms are present, an evaluation is warranted.

Sleep in healthy youth: Adolescence

The National Sleep Foundation recommends adolescents obtain 8 to 10 hours of sleep per night; for some adolescents, as much as 11 hours of sleep per night might be appropriate.⁸ However, this contrasts with findings from the National Sleep Foundation’s Sleep in America Poll, which revealed that 75% of 12th graders report <8 hours of sleep nightly.¹⁰ Many adolescents experience delayed sleep phase syndrome or delayed sleep-wake phase disorder, which involves a persistent phase shift of >2 hours in the sleep-wake schedule that conflicts with the adolescent’s school, work, or lifestyle demands.¹¹ Such circadian rhythm disorders typically result from a poor match between the sleep-wake schedule and the demands of the adolescent’s life, or a failure to synchronize their internal clock with a 24-hour circadian clock.¹² Children typically become tired after sunset, but puberty is associated with reduced slow-wave sleep and changes in circadian rhythms. As a result, a 3-hour delay (delayed phase preference) is common among adolescents. At approximately age 20, people start to become tired after sunset and awaken earlier in the morning—a pattern driven by sunlight and the timing of melatonin release that will remain stable until the sixth decade of life.

Continue to: Effects of chronic sleep deprivation...

Most older studies of sleep loss examined the impact of total sleep loss (sleep deprivation) rather than the effect of partial sleep loss or sleep restriction, a more commonly experienced phenomenon. More recent research shows that a cumulative sleep deficit could cause the body to override voluntary wakefulness and a sleep-deprived individual can experience brief “microsleeps” where they are unaware and lose attention/wakefulness for several seconds.² This can be deadly if a sleep-deprived adolescent experiences microsleeps while driving.¹³

There is a well-studied correlation between chronic sleep deprivation and increased body mass index in children.¹⁴ This might be caused by reduction in physical activity as well as alterations in the “hunger hormones”—ghrelin and leptin—that have been observed with sleep deprivation.^15-17 Other studies have noted decreased glucose tolerance, reduced insulin sensitivity, and catecholamine and cortisol secretion abnormalities, which place children at higher risk for metabolic syndrome and hypertension.^13,18 Sleep deprivation also is associated with mood and anxiety disorders and is an independent risk factor for substance use and suicidal ideation among adolescents.¹⁹ Sleep deprivation increases impairments in impulse control, concentration, and attention, which could be especially problematic in school-age children.

How sleep is assessed

The sleep history is the first step in evaluating a child or adolescent for a sleep disorder. The sleep history includes exploring the chief complaint, sleep patterns and schedules, bedtime routines, and nocturnal and daytime behaviors (Table).

Chief complaint

Behavioral sleep specialists will assess the primary problem with everyone involved in the child’s bedtime.²⁰ This might include parents (custodial and noncustodial), grandparents, or stepparents as well as the child/adolescent. This important step can reveal a sleep disorder or an inappropriately early bedtime relative to the child’s development. During this assessment, ask detailed questions about how long the sleep problem has persisted, the frequency of sleep problems, and any precipitating stressors. Parents and caregivers can review strategies they have tried, and for how long and to what extent interventions were implemented consistently to result in change.

Sleep patterns and schedules

Review the child/adolescent’s typical sleep patterns and behaviors. Ask parents and caregivers, as well as the patient, about general sleep schedules for the past few weeks or a typical 2-week time period.² A behavioral assessment of sleep should include asking families about how the child/adolescent sleeps during the week and over the weekend, and if school-year sleep differs from summer or holiday sleep schedules. These questions can illuminate how long a sleep problem has been occurring and what sleep habits might be contributing to the problem. Bedtime

Determine if there is a set bedtime or if the child goes to bed when they wish. It is important to ascertain if the bedtime is age-appropriate, if weekday and weekend bedtimes differ, and to what extent extracurricular activities or school demands impact bedtime. Assess the consistency of the bedtime, the nature of bedtime routines (eg, is the child engaging in stimulating activities before bed), where the bedtime routine occurs (eg, sibling’s room, parents’ room, child’s room), and what role (if any) electronic devices play.²

Nocturnal behaviors

Assessment should include a series of questions and age-specific questionnaires to focus on what behaviors occur at night, including awakenings. Parents should be asked how frequent night awakenings occur, how long arousals last, and how the child signals for the parent (eg, calling out, climbing into parents’ bed).² Additionally, ask how parents respond and what is required to help the child fall back asleep (eg, rocking, soothing, feeding). The presence of nightmares, night terrors, parasomnias, and sleep-related breathing disorders also must be assessed.²⁰

Daytime behaviors

A sleep history should include assessment of daytime functioning, including daytime sleepiness, fatigue, morning waking, and functioning during school, extracurriculars, and homework. For children and teens, falling asleep in the car, while in school, or during passive activities (meals, conversation) suggests insufficient sleep, sleep disruption, or excessive daytime sleepiness.²

Continue to: Sleep disruption in youth with psychiatric disorders...

Sleep disruption in youth with psychiatric disorders

Disordered sleep is common across psychiatric disorders. The National Comorbidity Survey Adolescent Supplement—a nationally representative cross-sectional survey of adolescents (N = 10,123)—found that a later weeknight bedtime, shorter weeknight sleep duration, and greater weekend bedtime delay increased the risk of developing a mood, anxiety, or substance use (including nicotine) disorder, and suicidality. These risk factors also were associated with lower “perceived mental and physical health.”²¹ Clinicians should routinely obtain a sleep history in children and adolescents with these disorders. Consider using the sleep screening tool BEARS:

• Bedtime issues
• Excessive daytime sleepiness
• Awakenings
• Regularity and duration of sleep
• Snoring.

ADHD

Up to one-half of children and adolescents with ADHD experience sleep problems,^22,23 including delayed sleep onset, bedtime resistance, daytime fatigue, and feeling groggy in the morning beyond what is typical (>20 minutes). Pharmacotherapy for ADHD contributes to sleep disturbances^24,25 while sleep deprivation exacerbates inattention and hyperactivity. In youth with ADHD, restless leg syndrome, periodic limb movement disorder, and sleep-disordered breathing disorder are more common than in the general population.

Depressive disorders

Up to three-quarters of depressed children and 90% of depressed adolescents report sleep disturbances, including initial, middle, and terminal insomnia as well as hypersomnia.²⁶ Disrupted sleep in pediatric patients with major depressive disorder could be moderated by the patient’s age, with depressive symptoms more common among adolescents (age 12 to 17) than among younger children (age 6 to 11).²⁷ Successful treatment of depression fails to relieve dyssomnia in 10% of children. Sleep problems that persist after successfully treating a depressive episode could increase the risk of another depressive episode.²⁸

Anxiety disorders

Sleep problems are common among children and adolescents with anxiety disorders.²⁹ Longitudinal data from >900 children found that symptoms of sleep disturbance in early childhood were correlated with experiencing an anxiety disorder 20 years later.³⁰ Fears related to the dark or monsters under the bed that are developmentally appropriate for younger children may interfere with sleep. However, in anxious children, fears might also be related to separation, sleeping alone, worry about the loss of a loved one, concerns about personal safety, fear of frightening dreams, or concerns about academics and social relationships. Anxious individuals ruminate about their worries, and this might be especially true for children at bedtime, when there are limited distractions from ruminative fears.³¹ Bedtime resistance, parental involvement in bedtime rituals, and cultural factors related to sleep also could play a role for children with anxiety symptoms and sleep problems.

Having an anxiety disorder is significantly associated with an increased risk of insomnia; however, 73% of the time anxiety symptoms precede an insomnia diagnosis.²⁹ Sleep problems and anxiety symptoms might have a reciprocal influence on one another; tiredness that results from sleep problems could exacerbate anxiety, which further worsens sleep problems.

A large body of research on sleep and anxiety reveals that abuse or exposure to trauma significantly affects sleep.³¹ Common sleep problems for children with posttraumatic stress disorder include difficulty falling asleep, maintaining sleep, and parasomnias of bedwetting and nightmares.^31,32 Compared with depressed and non-abused children, those with history of abuse have prolonged sleep latency, decreased sleep efficiency, and higher levels of activity during the night.³³ In addition to the relationship between anxiety disorders and sleep disorders, many of the selective serotonin reuptake inhibitors—which are the first-line pharmacotherapy for pediatric anxiety disorders^34,35—could affect sleep in anxious youth.³⁶

A bridge to treatment

A thorough assessment can help identify modifiable factors and guide treatment selections. In Part 2 of this article, we will describe healthy sleep practices, cognitive-behavioral therapy for insomnia, when pharmacotherapy might be indicated, and the evidence supporting several medications commonly used to treat pediatric insomnia. We also will discuss factors to consider when seeking consultation with a pediatric behavioral sleep specialist.

FIRST OF 2 PARTS

A thorough evaluation can identify modifiable factors and guide treatment

Sleep problems are common among children and adolescents,¹ with prevalence rates of 25% to 40%.^2-4 Young children most commonly exhibit what is referred to as bedtime problems and night wakenings, whereas children in middle childhood (age 4 to 12) through adolescence (age 13 to 17) report insomnia. For many children, these problems persist.³ Insufficient sleep in children and adolescents worsens inattention, daytime fatigue, and cognitive and behavioral deficits.⁵ Assessment and treatment of sleep problems in children and adolescents is critical because poor sleep among youth increases the risk for depression, self-harm, and suicide,^6,7 increases family stress, and decreases parental well-being.¹

This 2-part article describes the assessment, diagnosis, and treatment of sleep problems among children and adolescents. In part 1, we focus on:

• sleep architecture (circadian rhythms, stages of sleep)
• sleep in healthy youth (age 6 to 17) and those with attention-deficit/hyperactivity disorder (ADHD), depressive disorders, and anxiety
• how to assess sleep, and the differential diagnosis of behavioral sleep problems in pediatric patients.

In Part 2, we will cover psychotherapeutic and psychopharmacologic interventions for youth with insomnia, and describe an effective approach to consultation with pediatric sleep medicine specialists.

How much sleep do children and adolescents need?

Throughout their development, children spend 40% to 50% of their time asleep. Sleep schedules are based on circadian rhythms, which are physical, mental, and behavioral changes that follow an approximately 24-hour cycle. Human circadian rhythm varies between 24 and 25 hours and is vital in determining our sleep patterns. Exposure to sunlight drives our circadian rhythm, sending signals to our bodies to “turn on” melatonin production at night (ie, 9 pm melatonin secretion starts) and “turn off” melatonin secretion in the morning (ie, 7:30 am) in adults. Exposure to sunlight also drives circadian rhythms for children, and melatonin secretion production occurs earlier in the evening for children. Families are encouraged to expose children to bright light in the morning by taking an early walk, eating breakfast in a sunny room, or having adolescents open up the window shades rather than keeping their eyes shielded from the sun.² How sleepy or alert a child could be is determined by the duration and quality of a child’s previous sleep, and how long they have been awake (“sleep drive”).² The 3 states of sleep architecture are wake; non-rapid eye movement sleep; and rapid eye movement sleep (“dreaming” sleep) (Box²).

Box

Sleep needs also change predictably throughout the lifespan. The National Sleep Foundation guidelines for sleep duration provide clinicians and parents with a range of recommended sleep for each stage of development. Infants require 14 to 17 hours of sleep, whereas adolescents need 8 to 10 hours by age 14 to 17.⁸ The key for clinicians is to determine if the child is within the recommended range, and how they are functioning on the number of hours of sleep they report. This allows for variation in how much sleep an individual child might need while acknowledging that some children within a specific age group might need more or less sleep than other children of the same age.

Sleep in healthy youth: Middle childhood

School-age children (age 6 to 12) typically need 9 to 10 hours of sleep over a 24-hour period.² This developmental period is especially important for children to develop healthy sleep habits; however, developmentally appropriate cognitive and social/emotional factors might interfere with the quality and quantity of sleep. Middle childhood is a time when children can understand the dangers of the outside world (ie, violence, health problems) and resulting anxiety can disrupt sleep. Parents usually are less involved in bedtime as children approach adolescence, which leads to later bedtimes. At this stage, many children begin to take on more serious roles in their academics and extracurricular activities, peer relationships become more important, and use of electronics (eg, television, video games, internet, and handheld devices) increases—all of which compete with sleep.⁹ Frequent sleep issues during middle childhood include:

• irregular sleep-wake schedules
• later bedtimes
• decreased nighttime sleep
• increased caffeine intake
• reduced parental presence at bedtime
• daytime sleepiness.³

In school-age children, regular napping, falling asleep during short car rides, and daytime fatigue at school or home are cause for concern. When these symptoms are present, an evaluation is warranted.

Sleep in healthy youth: Adolescence

The National Sleep Foundation recommends adolescents obtain 8 to 10 hours of sleep per night; for some adolescents, as much as 11 hours of sleep per night might be appropriate.⁸ However, this contrasts with findings from the National Sleep Foundation’s Sleep in America Poll, which revealed that 75% of 12th graders report <8 hours of sleep nightly.¹⁰ Many adolescents experience delayed sleep phase syndrome or delayed sleep-wake phase disorder, which involves a persistent phase shift of >2 hours in the sleep-wake schedule that conflicts with the adolescent’s school, work, or lifestyle demands.¹¹ Such circadian rhythm disorders typically result from a poor match between the sleep-wake schedule and the demands of the adolescent’s life, or a failure to synchronize their internal clock with a 24-hour circadian clock.¹² Children typically become tired after sunset, but puberty is associated with reduced slow-wave sleep and changes in circadian rhythms. As a result, a 3-hour delay (delayed phase preference) is common among adolescents. At approximately age 20, people start to become tired after sunset and awaken earlier in the morning—a pattern driven by sunlight and the timing of melatonin release that will remain stable until the sixth decade of life.

Continue to: Effects of chronic sleep deprivation...

Most older studies of sleep loss examined the impact of total sleep loss (sleep deprivation) rather than the effect of partial sleep loss or sleep restriction, a more commonly experienced phenomenon. More recent research shows that a cumulative sleep deficit could cause the body to override voluntary wakefulness and a sleep-deprived individual can experience brief “microsleeps” where they are unaware and lose attention/wakefulness for several seconds.² This can be deadly if a sleep-deprived adolescent experiences microsleeps while driving.¹³

There is a well-studied correlation between chronic sleep deprivation and increased body mass index in children.¹⁴ This might be caused by reduction in physical activity as well as alterations in the “hunger hormones”—ghrelin and leptin—that have been observed with sleep deprivation.^15-17 Other studies have noted decreased glucose tolerance, reduced insulin sensitivity, and catecholamine and cortisol secretion abnormalities, which place children at higher risk for metabolic syndrome and hypertension.^13,18 Sleep deprivation also is associated with mood and anxiety disorders and is an independent risk factor for substance use and suicidal ideation among adolescents.¹⁹ Sleep deprivation increases impairments in impulse control, concentration, and attention, which could be especially problematic in school-age children.

How sleep is assessed

The sleep history is the first step in evaluating a child or adolescent for a sleep disorder. The sleep history includes exploring the chief complaint, sleep patterns and schedules, bedtime routines, and nocturnal and daytime behaviors (Table).

Chief complaint

Behavioral sleep specialists will assess the primary problem with everyone involved in the child’s bedtime.²⁰ This might include parents (custodial and noncustodial), grandparents, or stepparents as well as the child/adolescent. This important step can reveal a sleep disorder or an inappropriately early bedtime relative to the child’s development. During this assessment, ask detailed questions about how long the sleep problem has persisted, the frequency of sleep problems, and any precipitating stressors. Parents and caregivers can review strategies they have tried, and for how long and to what extent interventions were implemented consistently to result in change.

Sleep patterns and schedules

Review the child/adolescent’s typical sleep patterns and behaviors. Ask parents and caregivers, as well as the patient, about general sleep schedules for the past few weeks or a typical 2-week time period.² A behavioral assessment of sleep should include asking families about how the child/adolescent sleeps during the week and over the weekend, and if school-year sleep differs from summer or holiday sleep schedules. These questions can illuminate how long a sleep problem has been occurring and what sleep habits might be contributing to the problem. Bedtime

Determine if there is a set bedtime or if the child goes to bed when they wish. It is important to ascertain if the bedtime is age-appropriate, if weekday and weekend bedtimes differ, and to what extent extracurricular activities or school demands impact bedtime. Assess the consistency of the bedtime, the nature of bedtime routines (eg, is the child engaging in stimulating activities before bed), where the bedtime routine occurs (eg, sibling’s room, parents’ room, child’s room), and what role (if any) electronic devices play.²

Nocturnal behaviors

Assessment should include a series of questions and age-specific questionnaires to focus on what behaviors occur at night, including awakenings. Parents should be asked how frequent night awakenings occur, how long arousals last, and how the child signals for the parent (eg, calling out, climbing into parents’ bed).² Additionally, ask how parents respond and what is required to help the child fall back asleep (eg, rocking, soothing, feeding). The presence of nightmares, night terrors, parasomnias, and sleep-related breathing disorders also must be assessed.²⁰

Daytime behaviors

A sleep history should include assessment of daytime functioning, including daytime sleepiness, fatigue, morning waking, and functioning during school, extracurriculars, and homework. For children and teens, falling asleep in the car, while in school, or during passive activities (meals, conversation) suggests insufficient sleep, sleep disruption, or excessive daytime sleepiness.²

Continue to: Sleep disruption in youth with psychiatric disorders...

Sleep disruption in youth with psychiatric disorders

Disordered sleep is common across psychiatric disorders. The National Comorbidity Survey Adolescent Supplement—a nationally representative cross-sectional survey of adolescents (N = 10,123)—found that a later weeknight bedtime, shorter weeknight sleep duration, and greater weekend bedtime delay increased the risk of developing a mood, anxiety, or substance use (including nicotine) disorder, and suicidality. These risk factors also were associated with lower “perceived mental and physical health.”²¹ Clinicians should routinely obtain a sleep history in children and adolescents with these disorders. Consider using the sleep screening tool BEARS:

• Bedtime issues
• Excessive daytime sleepiness
• Awakenings
• Regularity and duration of sleep
• Snoring.

ADHD

Up to one-half of children and adolescents with ADHD experience sleep problems,^22,23 including delayed sleep onset, bedtime resistance, daytime fatigue, and feeling groggy in the morning beyond what is typical (>20 minutes). Pharmacotherapy for ADHD contributes to sleep disturbances^24,25 while sleep deprivation exacerbates inattention and hyperactivity. In youth with ADHD, restless leg syndrome, periodic limb movement disorder, and sleep-disordered breathing disorder are more common than in the general population.

Depressive disorders

Up to three-quarters of depressed children and 90% of depressed adolescents report sleep disturbances, including initial, middle, and terminal insomnia as well as hypersomnia.²⁶ Disrupted sleep in pediatric patients with major depressive disorder could be moderated by the patient’s age, with depressive symptoms more common among adolescents (age 12 to 17) than among younger children (age 6 to 11).²⁷ Successful treatment of depression fails to relieve dyssomnia in 10% of children. Sleep problems that persist after successfully treating a depressive episode could increase the risk of another depressive episode.²⁸

Anxiety disorders

Sleep problems are common among children and adolescents with anxiety disorders.²⁹ Longitudinal data from >900 children found that symptoms of sleep disturbance in early childhood were correlated with experiencing an anxiety disorder 20 years later.³⁰ Fears related to the dark or monsters under the bed that are developmentally appropriate for younger children may interfere with sleep. However, in anxious children, fears might also be related to separation, sleeping alone, worry about the loss of a loved one, concerns about personal safety, fear of frightening dreams, or concerns about academics and social relationships. Anxious individuals ruminate about their worries, and this might be especially true for children at bedtime, when there are limited distractions from ruminative fears.³¹ Bedtime resistance, parental involvement in bedtime rituals, and cultural factors related to sleep also could play a role for children with anxiety symptoms and sleep problems.

Having an anxiety disorder is significantly associated with an increased risk of insomnia; however, 73% of the time anxiety symptoms precede an insomnia diagnosis.²⁹ Sleep problems and anxiety symptoms might have a reciprocal influence on one another; tiredness that results from sleep problems could exacerbate anxiety, which further worsens sleep problems.

A large body of research on sleep and anxiety reveals that abuse or exposure to trauma significantly affects sleep.³¹ Common sleep problems for children with posttraumatic stress disorder include difficulty falling asleep, maintaining sleep, and parasomnias of bedwetting and nightmares.^31,32 Compared with depressed and non-abused children, those with history of abuse have prolonged sleep latency, decreased sleep efficiency, and higher levels of activity during the night.³³ In addition to the relationship between anxiety disorders and sleep disorders, many of the selective serotonin reuptake inhibitors—which are the first-line pharmacotherapy for pediatric anxiety disorders^34,35—could affect sleep in anxious youth.³⁶

A bridge to treatment

A thorough assessment can help identify modifiable factors and guide treatment selections. In Part 2 of this article, we will describe healthy sleep practices, cognitive-behavioral therapy for insomnia, when pharmacotherapy might be indicated, and the evidence supporting several medications commonly used to treat pediatric insomnia. We also will discuss factors to consider when seeking consultation with a pediatric behavioral sleep specialist.

In clinicians and patients alike, lithium triggers reactions ranging from apprehension and fear about adverse effects and toxicity to confusion over lithium’s usefulness compared with other mood stabilizers that do not require blood monitoring. Research from the 1950s to the 1970s demonstrated that lithium is effective for prophylaxis of mood episodes in patients with bipolar disorder and could reduce the frequency of hospitalization in patients who are depressed.¹ For years, lithium was commonly prescribed to treat bipolar disorder, but in recent years its use has fallen out of favor due to concerns about its risks, and the availability of newer medications. This article reviews lithium’s origins (Box^1-4), pharmacology, risks, and benefits, and makes a case for why it should remain a first-line therapy for bipolar disorder.

Box

How lithium works

Lithium has effects on neurotransmitters implicated in mania, such as glutamate, dopamine, and gamma-aminobutyric acid.⁵ Quiroz et al⁶ provide a detailed description of lithium’s effects, which can be summarized as modulating neuronal signaling pathways, including B-cell lymphoma 2 (BCL2), cAMP-response element binding protein (CREB), and glycogen synthase kinase-3 (GSK-3). Through these signaling cascades, lithium can curtail progression of neuronal apoptosis caused by the biochemical stress commonly seen in bipolar disorder pathogenesis.⁶  

A wide range of potential adverse effects

Lithium can cause adverse effects in several organ systems. Clinicians must be aware of these effects before prescribing lithium or continuing long-term use. The most commonly documented adverse effects and symptoms of toxicity are:

• tremor
• renal dysfunction, including renal insufficiency and polyuria or polydipsia
• hypothyroidism
• hyperparathyroidism (with subsequent hypercalcemia)
• weight gain
• gastrointestinal (GI) symptoms.

These symptoms tend to occur when lithium serum levels are outside the reference range of 0.6 to 1.2 mEq/L, typically once blood levels reach ≥1.5 mEq/L.⁷ However, thyroid and renal abnormalities can occur at levels below this value, and might be related to cumulative lithium exposure.⁷ Adverse effects usually are precipitated by inadequate water intake or inadvertently taking an extra dose. Symptoms of lithium toxicity can be mild, moderate (GI complaints, tremor, weakness, fatigue), or severe (agitation, seizures, autonomic dysregulation, confusion, coma, death).

Lithium adverse effects and toxicity are infrequent. An analysis of 17 years of data in Sweden showed the incidence of moderate to severe lithium intoxication (serum level ≥1.5 mEq/L) was .01 patients per year.⁸ A recently published US analysis found the prevalence rate of lithium toxicity was 2.2%.⁹ Results from both groups show that drug interactions were an important cause of increased lithium levels, and specifically that initiating a medication that could interact with lithium was associated with 30-fold higher risk of needing acute care for lithium toxicity.⁹ Possible drug interactions include nonsteroidal anti-inflammatory drugs, diuretics, and renin-angiotensin-aldosterone system inhibitors.⁹ Because lithium is eliminated exclusively by the kidneys, impaired or altered renal function can increase the risk of lithium retention, leading to intoxication. Other risk factors include older age, alteration of water-salt homeostasis (fever, diarrhea, vomiting), higher number of treated chronic diseases as measured by Chronic Disease Score (range: 0 to 35; higher scores denotes higher number of treated chronic diseases and increased hospitalization risk), and higher total daily lithium dosage.⁹

Presentation of lithium intoxication often is mild or nonspecific, and physicians should have a low threshold for checking lithium blood levels.⁸ Lithium intoxication can be safely managed with volume expansion, forced diuresis, and hemodialysis.

Continue to: Lithium use during pregnancy...

Lithium use during pregnancy

When considering lithium for a woman who is pregnant, it is important to weigh the potential teratogenic risks against the benefit of successful management of the mood disorder. Ebstein’s anomaly (abnormal tricuspid valve leaflets) is the most well-known teratogenic risk associated with lithium, with an estimated absolute risk of 1 in 1,000 in patients treated with lithium compared with 1 in 20,000 in controls.^10,11 The risk of congenital anomalies is increased in infants exposed to lithium in utero (4% to 12% vs 2% to 4% in controls)¹²; exposure during the first trimester of pregnancy is associated with increased risk. Lithium levels must be adjusted during pregnancy. Pregnant patients are at higher risk of relapse to mania because renal lithium clearance increases by 30% to 50% during pregnancy, and normalizes shortly after delivery.¹³

Lithium exposure during pregnancy has been linked to increased risk of miscarriage and preterm delivery; however, more research is needed to define the true risk of noncardiac teratogenicity associated with lithium.¹¹ Because there is a lack of definitive data regarding teratogenicity, and because of lithium’s well-documented effectiveness in mood disorders, lithium should be considered a first-line therapy for pregnant patients with bipolar disorder.¹⁰

Prescribing trends

Despite data showing the efficacy and benefits of lithium, there has been a paradoxical decrease in lithium prescribing. This is the result of multiple factors, including fear of adverse effects and lithium toxicity and a shift toward newer medications, such as anticonvulsants and antipsychotics, for treatment and prophylaxis of mania.

A 2011 study examined prescribing trends for bipolar disorder in the United Kingdom.¹⁴ Overall, it found increased usage of valproate, carbamazepine, and lamotrigine from 1995 to 2009. During that time, lithium prescribing mostly remained steady at approximately 30%, whereas valproate use increased from 0% to 22.7%. Overall, antipsychotic and valproate prescribing increased relative to lithium.¹⁴ A literature review¹⁵ analyzed 6 studies of lithium prescribing trends from 1950 to 2010. Four of these studies (2 in the United States, 1 in Canada, and 1 in German-Swiss-Austrian hospitals) found lithium use was declining. The increased use found in Italy and Spain was attributed to multiple factors, including a broader definition of bipolar disorders and the unavailability of valproate in Spain, lithium’s low cost, and mental health reforms in both countries that resulted in overall increased psychotropic prescribing. Decreased lithium use was attributed to increased use of valproate and second-generation antipsychotics, lack of clinician training in lithium therapy, and aggressive marketing of brand-name medications.¹⁵
 

Reduced suicides, possible protection against dementia

A 2013 meta-analysis of 48 randomized controlled trials (RCTs) that included a total of 6,674 patients with mood disorders indicated that compared with placebo, lithium was more effective in reducing suicides and deaths from any cause.¹⁶

Large retrospective studies have demonstrated that compared with valproate, lithium has superior anti-suicide properties.¹⁷ Researchers found that risk of suicide attempt or completion was 1.5 to 3 times higher during periods of valproate treatment compared with lithium.¹⁸ Both short- and long-term lithium use was associated with decreased non-suicide mortality compared with valproate.¹⁹ In Denmark, compared with valproate, lithium was associated with fewer psychiatric hospital admissions.¹⁹ One RCT, the BALANCE trial, showed that lithium (alone or in combination with valproate) is more likely to prevent relapse in persons with bipolar I disorder than valproate monotherapy.²⁰

Recent research in Denmark suggests that long-term doses of naturally occurring lithium in drinking water might confer some level of protection against dementia.²¹ Researchers examined the Danish National Patient Register to determine where participants lived and their local water supply. Drinking water lithium levels were assessed, and the mean lithium level for each municipality was calculated. This case-control study selected patients with dementia and 10 age- and sex-matched controls.²¹

Researchers found that the incidence rate ratio of Alzheimer disease, vascular dementia, and dementia overall was significantly lower among individuals whose drinking water contained lithium, 15.1 to 27.0 µg/L, compared with those whose water had lithium levels 2.0 to 5.0 µg/L.²¹ Although this study does not prove causality, it opens the door for continued research on lithium as a neuroprotective agent involved in pathways beyond mood stabilization.

Why should you prescribe lithium?

Lithium, which is available in several formulations (Table), should continue to be first-line pharmacotherapy for treating acute mood episodes, prophylaxis, and suicide prevention in bipolar disorder. Although there are many effective medications for treating bipolar disorder—such as second-generation antipsychotics that are available as a long-acting injectable formulation or can be combined with a mood stabilizer—lithium is a thoroughly researched medication with a long history of effectiveness for managing bipolar disorder. As is the case with all psychotropic medications, lithium has adverse effects and necessary precautions, but these are outweighed by its neuroprotective benefits and efficacy. Research has demonstrated that lithium outperforms medications that have largely replaced it, specifically valproate.

Related Resources

• Ali ZA, El-Mallakh RS. Lithium and kidney disease: Understand the risks. Current Psychiatry. 2021;20(6):34- 38,50. doi:10.12788/cp.0130
• Malhi GS, Gessler D, Outhred T. The use of lithium for the treatment of bipolar disorder: recommendations from clinical practice guidelines. J Affect Disord. 2017;217: 266-280. doi:10.1016/j.jad.2017.03.052

Drug Brand Names

Carbamazepine • Tegretol

Lamotrigine • Lamictal

Lithium • Eskalith, Lithobid

Valproate • Depacon, Depakote, Depakene

Bottom Line

Lithium is a well-researched first-line pharmacotherapy for bipolar disorder, with efficacy equivalent to—or superior to—newer pharmacotherapies such as valproate and second-generation antipsychotics. When prescribing lithium, carefully monitor patients for symptoms of adverse effects or toxicity. Despite teratogenic risks, lithium can be considered for pregnant patients with bipolar disorder.

In clinicians and patients alike, lithium triggers reactions ranging from apprehension and fear about adverse effects and toxicity to confusion over lithium’s usefulness compared with other mood stabilizers that do not require blood monitoring. Research from the 1950s to the 1970s demonstrated that lithium is effective for prophylaxis of mood episodes in patients with bipolar disorder and could reduce the frequency of hospitalization in patients who are depressed.¹ For years, lithium was commonly prescribed to treat bipolar disorder, but in recent years its use has fallen out of favor due to concerns about its risks, and the availability of newer medications. This article reviews lithium’s origins (Box^1-4), pharmacology, risks, and benefits, and makes a case for why it should remain a first-line therapy for bipolar disorder.

Box

How lithium works

Lithium has effects on neurotransmitters implicated in mania, such as glutamate, dopamine, and gamma-aminobutyric acid.⁵ Quiroz et al⁶ provide a detailed description of lithium’s effects, which can be summarized as modulating neuronal signaling pathways, including B-cell lymphoma 2 (BCL2), cAMP-response element binding protein (CREB), and glycogen synthase kinase-3 (GSK-3). Through these signaling cascades, lithium can curtail progression of neuronal apoptosis caused by the biochemical stress commonly seen in bipolar disorder pathogenesis.⁶  

A wide range of potential adverse effects

Lithium can cause adverse effects in several organ systems. Clinicians must be aware of these effects before prescribing lithium or continuing long-term use. The most commonly documented adverse effects and symptoms of toxicity are:

• tremor
• renal dysfunction, including renal insufficiency and polyuria or polydipsia
• hypothyroidism
• hyperparathyroidism (with subsequent hypercalcemia)
• weight gain
• gastrointestinal (GI) symptoms.

These symptoms tend to occur when lithium serum levels are outside the reference range of 0.6 to 1.2 mEq/L, typically once blood levels reach ≥1.5 mEq/L.⁷ However, thyroid and renal abnormalities can occur at levels below this value, and might be related to cumulative lithium exposure.⁷ Adverse effects usually are precipitated by inadequate water intake or inadvertently taking an extra dose. Symptoms of lithium toxicity can be mild, moderate (GI complaints, tremor, weakness, fatigue), or severe (agitation, seizures, autonomic dysregulation, confusion, coma, death).

Lithium adverse effects and toxicity are infrequent. An analysis of 17 years of data in Sweden showed the incidence of moderate to severe lithium intoxication (serum level ≥1.5 mEq/L) was .01 patients per year.⁸ A recently published US analysis found the prevalence rate of lithium toxicity was 2.2%.⁹ Results from both groups show that drug interactions were an important cause of increased lithium levels, and specifically that initiating a medication that could interact with lithium was associated with 30-fold higher risk of needing acute care for lithium toxicity.⁹ Possible drug interactions include nonsteroidal anti-inflammatory drugs, diuretics, and renin-angiotensin-aldosterone system inhibitors.⁹ Because lithium is eliminated exclusively by the kidneys, impaired or altered renal function can increase the risk of lithium retention, leading to intoxication. Other risk factors include older age, alteration of water-salt homeostasis (fever, diarrhea, vomiting), higher number of treated chronic diseases as measured by Chronic Disease Score (range: 0 to 35; higher scores denotes higher number of treated chronic diseases and increased hospitalization risk), and higher total daily lithium dosage.⁹

Presentation of lithium intoxication often is mild or nonspecific, and physicians should have a low threshold for checking lithium blood levels.⁸ Lithium intoxication can be safely managed with volume expansion, forced diuresis, and hemodialysis.

Continue to: Lithium use during pregnancy...

Lithium use during pregnancy

When considering lithium for a woman who is pregnant, it is important to weigh the potential teratogenic risks against the benefit of successful management of the mood disorder. Ebstein’s anomaly (abnormal tricuspid valve leaflets) is the most well-known teratogenic risk associated with lithium, with an estimated absolute risk of 1 in 1,000 in patients treated with lithium compared with 1 in 20,000 in controls.^10,11 The risk of congenital anomalies is increased in infants exposed to lithium in utero (4% to 12% vs 2% to 4% in controls)¹²; exposure during the first trimester of pregnancy is associated with increased risk. Lithium levels must be adjusted during pregnancy. Pregnant patients are at higher risk of relapse to mania because renal lithium clearance increases by 30% to 50% during pregnancy, and normalizes shortly after delivery.¹³

Lithium exposure during pregnancy has been linked to increased risk of miscarriage and preterm delivery; however, more research is needed to define the true risk of noncardiac teratogenicity associated with lithium.¹¹ Because there is a lack of definitive data regarding teratogenicity, and because of lithium’s well-documented effectiveness in mood disorders, lithium should be considered a first-line therapy for pregnant patients with bipolar disorder.¹⁰

Prescribing trends

Despite data showing the efficacy and benefits of lithium, there has been a paradoxical decrease in lithium prescribing. This is the result of multiple factors, including fear of adverse effects and lithium toxicity and a shift toward newer medications, such as anticonvulsants and antipsychotics, for treatment and prophylaxis of mania.

A 2011 study examined prescribing trends for bipolar disorder in the United Kingdom.¹⁴ Overall, it found increased usage of valproate, carbamazepine, and lamotrigine from 1995 to 2009. During that time, lithium prescribing mostly remained steady at approximately 30%, whereas valproate use increased from 0% to 22.7%. Overall, antipsychotic and valproate prescribing increased relative to lithium.¹⁴ A literature review¹⁵ analyzed 6 studies of lithium prescribing trends from 1950 to 2010. Four of these studies (2 in the United States, 1 in Canada, and 1 in German-Swiss-Austrian hospitals) found lithium use was declining. The increased use found in Italy and Spain was attributed to multiple factors, including a broader definition of bipolar disorders and the unavailability of valproate in Spain, lithium’s low cost, and mental health reforms in both countries that resulted in overall increased psychotropic prescribing. Decreased lithium use was attributed to increased use of valproate and second-generation antipsychotics, lack of clinician training in lithium therapy, and aggressive marketing of brand-name medications.¹⁵
 

Reduced suicides, possible protection against dementia

A 2013 meta-analysis of 48 randomized controlled trials (RCTs) that included a total of 6,674 patients with mood disorders indicated that compared with placebo, lithium was more effective in reducing suicides and deaths from any cause.¹⁶

Large retrospective studies have demonstrated that compared with valproate, lithium has superior anti-suicide properties.¹⁷ Researchers found that risk of suicide attempt or completion was 1.5 to 3 times higher during periods of valproate treatment compared with lithium.¹⁸ Both short- and long-term lithium use was associated with decreased non-suicide mortality compared with valproate.¹⁹ In Denmark, compared with valproate, lithium was associated with fewer psychiatric hospital admissions.¹⁹ One RCT, the BALANCE trial, showed that lithium (alone or in combination with valproate) is more likely to prevent relapse in persons with bipolar I disorder than valproate monotherapy.²⁰

Recent research in Denmark suggests that long-term doses of naturally occurring lithium in drinking water might confer some level of protection against dementia.²¹ Researchers examined the Danish National Patient Register to determine where participants lived and their local water supply. Drinking water lithium levels were assessed, and the mean lithium level for each municipality was calculated. This case-control study selected patients with dementia and 10 age- and sex-matched controls.²¹

Researchers found that the incidence rate ratio of Alzheimer disease, vascular dementia, and dementia overall was significantly lower among individuals whose drinking water contained lithium, 15.1 to 27.0 µg/L, compared with those whose water had lithium levels 2.0 to 5.0 µg/L.²¹ Although this study does not prove causality, it opens the door for continued research on lithium as a neuroprotective agent involved in pathways beyond mood stabilization.

Why should you prescribe lithium?

Lithium, which is available in several formulations (Table), should continue to be first-line pharmacotherapy for treating acute mood episodes, prophylaxis, and suicide prevention in bipolar disorder. Although there are many effective medications for treating bipolar disorder—such as second-generation antipsychotics that are available as a long-acting injectable formulation or can be combined with a mood stabilizer—lithium is a thoroughly researched medication with a long history of effectiveness for managing bipolar disorder. As is the case with all psychotropic medications, lithium has adverse effects and necessary precautions, but these are outweighed by its neuroprotective benefits and efficacy. Research has demonstrated that lithium outperforms medications that have largely replaced it, specifically valproate.

Related Resources

• Ali ZA, El-Mallakh RS. Lithium and kidney disease: Understand the risks. Current Psychiatry. 2021;20(6):34- 38,50. doi:10.12788/cp.0130
• Malhi GS, Gessler D, Outhred T. The use of lithium for the treatment of bipolar disorder: recommendations from clinical practice guidelines. J Affect Disord. 2017;217: 266-280. doi:10.1016/j.jad.2017.03.052

Drug Brand Names

Carbamazepine • Tegretol

Lamotrigine • Lamictal

Lithium • Eskalith, Lithobid

Valproate • Depacon, Depakote, Depakene

Bottom Line

Lithium is a well-researched first-line pharmacotherapy for bipolar disorder, with efficacy equivalent to—or superior to—newer pharmacotherapies such as valproate and second-generation antipsychotics. When prescribing lithium, carefully monitor patients for symptoms of adverse effects or toxicity. Despite teratogenic risks, lithium can be considered for pregnant patients with bipolar disorder.

In clinicians and patients alike, lithium triggers reactions ranging from apprehension and fear about adverse effects and toxicity to confusion over lithium’s usefulness compared with other mood stabilizers that do not require blood monitoring. Research from the 1950s to the 1970s demonstrated that lithium is effective for prophylaxis of mood episodes in patients with bipolar disorder and could reduce the frequency of hospitalization in patients who are depressed.¹ For years, lithium was commonly prescribed to treat bipolar disorder, but in recent years its use has fallen out of favor due to concerns about its risks, and the availability of newer medications. This article reviews lithium’s origins (Box^1-4), pharmacology, risks, and benefits, and makes a case for why it should remain a first-line therapy for bipolar disorder.

Box

How lithium works

Lithium has effects on neurotransmitters implicated in mania, such as glutamate, dopamine, and gamma-aminobutyric acid.⁵ Quiroz et al⁶ provide a detailed description of lithium’s effects, which can be summarized as modulating neuronal signaling pathways, including B-cell lymphoma 2 (BCL2), cAMP-response element binding protein (CREB), and glycogen synthase kinase-3 (GSK-3). Through these signaling cascades, lithium can curtail progression of neuronal apoptosis caused by the biochemical stress commonly seen in bipolar disorder pathogenesis.⁶  

A wide range of potential adverse effects

Lithium can cause adverse effects in several organ systems. Clinicians must be aware of these effects before prescribing lithium or continuing long-term use. The most commonly documented adverse effects and symptoms of toxicity are:

• tremor
• renal dysfunction, including renal insufficiency and polyuria or polydipsia
• hypothyroidism
• hyperparathyroidism (with subsequent hypercalcemia)
• weight gain
• gastrointestinal (GI) symptoms.

These symptoms tend to occur when lithium serum levels are outside the reference range of 0.6 to 1.2 mEq/L, typically once blood levels reach ≥1.5 mEq/L.⁷ However, thyroid and renal abnormalities can occur at levels below this value, and might be related to cumulative lithium exposure.⁷ Adverse effects usually are precipitated by inadequate water intake or inadvertently taking an extra dose. Symptoms of lithium toxicity can be mild, moderate (GI complaints, tremor, weakness, fatigue), or severe (agitation, seizures, autonomic dysregulation, confusion, coma, death).

Lithium adverse effects and toxicity are infrequent. An analysis of 17 years of data in Sweden showed the incidence of moderate to severe lithium intoxication (serum level ≥1.5 mEq/L) was .01 patients per year.⁸ A recently published US analysis found the prevalence rate of lithium toxicity was 2.2%.⁹ Results from both groups show that drug interactions were an important cause of increased lithium levels, and specifically that initiating a medication that could interact with lithium was associated with 30-fold higher risk of needing acute care for lithium toxicity.⁹ Possible drug interactions include nonsteroidal anti-inflammatory drugs, diuretics, and renin-angiotensin-aldosterone system inhibitors.⁹ Because lithium is eliminated exclusively by the kidneys, impaired or altered renal function can increase the risk of lithium retention, leading to intoxication. Other risk factors include older age, alteration of water-salt homeostasis (fever, diarrhea, vomiting), higher number of treated chronic diseases as measured by Chronic Disease Score (range: 0 to 35; higher scores denotes higher number of treated chronic diseases and increased hospitalization risk), and higher total daily lithium dosage.⁹

Presentation of lithium intoxication often is mild or nonspecific, and physicians should have a low threshold for checking lithium blood levels.⁸ Lithium intoxication can be safely managed with volume expansion, forced diuresis, and hemodialysis.

Continue to: Lithium use during pregnancy...

Lithium use during pregnancy

When considering lithium for a woman who is pregnant, it is important to weigh the potential teratogenic risks against the benefit of successful management of the mood disorder. Ebstein’s anomaly (abnormal tricuspid valve leaflets) is the most well-known teratogenic risk associated with lithium, with an estimated absolute risk of 1 in 1,000 in patients treated with lithium compared with 1 in 20,000 in controls.^10,11 The risk of congenital anomalies is increased in infants exposed to lithium in utero (4% to 12% vs 2% to 4% in controls)¹²; exposure during the first trimester of pregnancy is associated with increased risk. Lithium levels must be adjusted during pregnancy. Pregnant patients are at higher risk of relapse to mania because renal lithium clearance increases by 30% to 50% during pregnancy, and normalizes shortly after delivery.¹³

Lithium exposure during pregnancy has been linked to increased risk of miscarriage and preterm delivery; however, more research is needed to define the true risk of noncardiac teratogenicity associated with lithium.¹¹ Because there is a lack of definitive data regarding teratogenicity, and because of lithium’s well-documented effectiveness in mood disorders, lithium should be considered a first-line therapy for pregnant patients with bipolar disorder.¹⁰

Prescribing trends

Despite data showing the efficacy and benefits of lithium, there has been a paradoxical decrease in lithium prescribing. This is the result of multiple factors, including fear of adverse effects and lithium toxicity and a shift toward newer medications, such as anticonvulsants and antipsychotics, for treatment and prophylaxis of mania.

A 2011 study examined prescribing trends for bipolar disorder in the United Kingdom.¹⁴ Overall, it found increased usage of valproate, carbamazepine, and lamotrigine from 1995 to 2009. During that time, lithium prescribing mostly remained steady at approximately 30%, whereas valproate use increased from 0% to 22.7%. Overall, antipsychotic and valproate prescribing increased relative to lithium.¹⁴ A literature review¹⁵ analyzed 6 studies of lithium prescribing trends from 1950 to 2010. Four of these studies (2 in the United States, 1 in Canada, and 1 in German-Swiss-Austrian hospitals) found lithium use was declining. The increased use found in Italy and Spain was attributed to multiple factors, including a broader definition of bipolar disorders and the unavailability of valproate in Spain, lithium’s low cost, and mental health reforms in both countries that resulted in overall increased psychotropic prescribing. Decreased lithium use was attributed to increased use of valproate and second-generation antipsychotics, lack of clinician training in lithium therapy, and aggressive marketing of brand-name medications.¹⁵
 

Reduced suicides, possible protection against dementia

A 2013 meta-analysis of 48 randomized controlled trials (RCTs) that included a total of 6,674 patients with mood disorders indicated that compared with placebo, lithium was more effective in reducing suicides and deaths from any cause.¹⁶

Large retrospective studies have demonstrated that compared with valproate, lithium has superior anti-suicide properties.¹⁷ Researchers found that risk of suicide attempt or completion was 1.5 to 3 times higher during periods of valproate treatment compared with lithium.¹⁸ Both short- and long-term lithium use was associated with decreased non-suicide mortality compared with valproate.¹⁹ In Denmark, compared with valproate, lithium was associated with fewer psychiatric hospital admissions.¹⁹ One RCT, the BALANCE trial, showed that lithium (alone or in combination with valproate) is more likely to prevent relapse in persons with bipolar I disorder than valproate monotherapy.²⁰

Recent research in Denmark suggests that long-term doses of naturally occurring lithium in drinking water might confer some level of protection against dementia.²¹ Researchers examined the Danish National Patient Register to determine where participants lived and their local water supply. Drinking water lithium levels were assessed, and the mean lithium level for each municipality was calculated. This case-control study selected patients with dementia and 10 age- and sex-matched controls.²¹

Researchers found that the incidence rate ratio of Alzheimer disease, vascular dementia, and dementia overall was significantly lower among individuals whose drinking water contained lithium, 15.1 to 27.0 µg/L, compared with those whose water had lithium levels 2.0 to 5.0 µg/L.²¹ Although this study does not prove causality, it opens the door for continued research on lithium as a neuroprotective agent involved in pathways beyond mood stabilization.

Why should you prescribe lithium?

Lithium, which is available in several formulations (Table), should continue to be first-line pharmacotherapy for treating acute mood episodes, prophylaxis, and suicide prevention in bipolar disorder. Although there are many effective medications for treating bipolar disorder—such as second-generation antipsychotics that are available as a long-acting injectable formulation or can be combined with a mood stabilizer—lithium is a thoroughly researched medication with a long history of effectiveness for managing bipolar disorder. As is the case with all psychotropic medications, lithium has adverse effects and necessary precautions, but these are outweighed by its neuroprotective benefits and efficacy. Research has demonstrated that lithium outperforms medications that have largely replaced it, specifically valproate.

Related Resources

• Ali ZA, El-Mallakh RS. Lithium and kidney disease: Understand the risks. Current Psychiatry. 2021;20(6):34- 38,50. doi:10.12788/cp.0130
• Malhi GS, Gessler D, Outhred T. The use of lithium for the treatment of bipolar disorder: recommendations from clinical practice guidelines. J Affect Disord. 2017;217: 266-280. doi:10.1016/j.jad.2017.03.052

Drug Brand Names

Carbamazepine • Tegretol

Lamotrigine • Lamictal

Lithium • Eskalith, Lithobid

Valproate • Depacon, Depakote, Depakene

Bottom Line

Lithium is a well-researched first-line pharmacotherapy for bipolar disorder, with efficacy equivalent to—or superior to—newer pharmacotherapies such as valproate and second-generation antipsychotics. When prescribing lithium, carefully monitor patients for symptoms of adverse effects or toxicity. Despite teratogenic risks, lithium can be considered for pregnant patients with bipolar disorder.

Sara R. Abell, MD, and Rif S. El-Mallakh, MD

Individuals with anxiety will experience frequent or chronic excessive worry, nervousness, a sense of unease, a feeling of being unfocused, and distress, which result in functional impairment.¹ Frequently, anxiety is accompanied by restlessness or muscle tension. Generalized anxiety disorder is one of the most common psychiatric diagnoses in the United States and has a prevalence of 2% to 6% globally.² Although research has been conducted regarding anxiety’s pathogenesis, to date a firm consensus on its etiology has not been reached.³ It is likely multifactorial, with environmental and biologic components.

One area of focus has been neurotransmitters and the possible role they play in the pathogenesis of anxiety. Specifically, the monoamine neurotransmitters have been implicated in the clinical manifestations of anxiety. Among the amines, normal roles include stimulating the autonomic nervous system and regulating numerous cognitive phenomena, such as volition and emotion. Many psychiatric medications modify aminergic transmission, and many current anxiety medications target amine neurotransmitters. Medications that target histamine, serotonin, norepinephrine, and dopamine all play a role in treating anxiety.

In this article, we focus on serotonin (5-hydroxytryptamine, 5-HT) as a mediator of anxiety and on excessive synaptic 5-HT as the cause of anxiety. We discuss how 5-HT–mediated anxiety can be identified and offer some solutions for its treatment.

The amine neurotransmitters

There are 6 amine neurotransmitters in the CNS. These are derived from tyrosine (dopamine [DA], norepinephrine [NE], and epinephrine), histidine (histamine), and tryptophan (serotonin [5-HT] and melatonin). In addition to their physiologic actions, amines have been implicated in both acute and chronic anxiety. Excessive DA stimulation has been linked with fear^4,5; NE elevations are central to hypervigilance and hyperarousal of posttraumatic stress disorder⁶; and histamine may mediate emotional memories involved in fear and anxiety.⁷ Understanding the normal function of 5-HT will aid in understanding its potential problematic role (Box,^8-18page 38).

How serotonin-mediated anxiety presents

“Anxiety” is a collection of signs and symptoms that likely represent multiple processes and have the common characteristic of being subjectively unpleasant, with a subjective wish for the feeling to end. The expression of anxiety disorders is quite diverse and ranges from brief episodes such as panic attacks (which may be mediated, in part, by epinephrine/NE¹⁹) to lifelong stereotypic obsessions and compulsions (which may be mediated, in part, by DA and modified by 5-HT^20,21). Biochemical separation of the anxiety disorders is key to achieving tailored treatment.⁶ Towards this end, it is important to investigate the phenomenon of serotonin-mediated anxiety.

Because clinicians are familiar with reductions of anxiety as selective serotonin reuptake inhibitors (SSRIs) increase 5-HT levels in the synapse, it is difficult to conceptualize serotonin-mediated anxiety. However, many of the effects at postsynaptic 5-HT receptors may be biphasic.^15-18 Serotonin-mediated anxiety appears to occur when levels of 5-HT (or stimulation of 5-HT receptors) are particularly high. This is most frequently seen in patients who genetically have high synaptic 5-HT (by virtue of the short form of the 5-HT transporter),²² whose synaptic 5-HT is further increased by treatment with an SSRI,²³ and who are experiencing a stressor that yet further increases their synaptic 5-HT.²⁴ However, it may occur in some individuals with only 2 of these 3 conditions.Clinically, individuals with serotonin-mediated anxiety will usually appear calm. The anxiety they are experiencing is not exhibited in any way in the motor system (ie, they do not appear restless, do not pace, muscle tone is not increased, etc.). However, they will generally complain of an internal agitation, a sense of a negative internal energy. Frequently, they will use descriptions such as “I feel I could jump out of my skin.” As previously mentioned, this is usually in the setting of some environmental stress, in addition to either a pharmacologic (SSRI) or genetic (short form of the 5-HT transporter) reason for increasing synaptic 5-HT, or both.

Almost always, interventions that block multiple postsynaptic 5-HT receptors or discontinuation of the SSRI (if applicable) will alleviate the anxiety, quickly or more slowly, respectively. Sublingual asenapine, which at low doses can block 5-HT2C (Ki = 0.03 nM), 5-HT2A (Ki = 0.07 nM), 5-HT7 (Ki = 0.11 nM), 5-HT2B (Ki = 0.18 nM), and 5-HT6 (Ki = 0.25 nM),^25,26 and which will produce peak plasma levels within 10 minutes,²⁷ usually is quite effective.

Box

Continue to: A key difference: No motor system involvement...

A key difference: No motor system involvement

What distinguishes 5-HT from the other amine transmitters as a mediator of anxiety is the lack of involvement of the motor system. Multiple studies in rats illustrate that exogenously augmenting 5-HT has no effect on levels of locomotor activity. Dopamine depletion is well-characterized in the motor dysfunction of Parkinson’s disease, and DA excess can cause repetitive, stereotyped movements, such as seen in tardive dyskinesia or Huntington’s disease.⁸ In humans, serotonin-mediated anxiety is usually without a motoric component; patients appear calm but complain of extreme anxiety or agitation. Agitation has been reported after initiation of an SSRI,²⁹ and is more likely to occur in patients with the short form of the 5-HT transporter.³⁰ Motoric activation has been reported in some of these studies, but does not seem to cluster with the complaint of agitation.²⁹ The reduced number of available transporters means a chronic steady-state elevation of serotonin, because less serotonin is being removed from the synapse after it is released. This is one of the reasons patients with the short form of the 5-HT transporter may be more susceptible to serotonin-mediated anxiety.

What you need to keep in mind

Pharmacologic treatment of anxiety begins with an SSRI, a serotonin-norepinephrine reuptake inhibitor (SNRI), or buspirone. Second-line treatments include hydroxyzine, gabapentin, pregabalin, and quetiapine.^3,31 However, clinicians need to be aware that a fraction of their patients will report anxiety that will not have any external manifestations, but will be experienced as an unpleasant internal energy. These patients may report an increase in their anxiety levels when started on an SSRI or SNRI.^29,30 This anxiety is most likely mediated by increases of synaptic 5-HT. This occurs because many serotonergic receptors may have a biphasic response, so that too much stimulation is experienced as excessive internal energy.^16-18 In such patients, blockade of key 5-HT receptors may reduce that internal agitation. The advantage of recognizing serotonin-mediated anxiety is that one can specifically tailor treatment to address the patient’s specific physiology.

It is important to note that the anxiolytic effect of asenapine is specific to patients with serotonin-mediated anxiety. Unlike quetiapine, which is effective as augmentation therapy in generalized anxiety disorder,³¹ asenapine does not appear to reduce anxiety in patients with schizophrenia³² or borderline personality disorder³³ when administered for other reasons. However, it may reduce anxiety in patients with the short form of the 5-HT transporter.^30,34

Bottom Line

Serotonin-mediated anxiety occurs when levels of synaptic serotonin (5-HT) are high. Patients with serotonin-mediated anxiety appear calm but will report experiencing an unpleasant internal energy. Interventions that block multiple postsynaptic 5-HT receptors or discontinuation of a selective serotonin reuptake inhibitor (if applicable) will alleviate the anxiety.

Related Resource

• Bhatt NV. Anxiety disorders. https://emedicine.medscape. com/article/286227-overview

Drug Brand Names

Asenapine • Saphris, Secuado

Gabapentin • Neurontin

Hydroxyzine • Vistaril

Pregabalin • Lyrica

Quetiapine • Seroquel

Trazodone • Oleptro

Sara R. Abell, MD, and Rif S. El-Mallakh, MD

Individuals with anxiety will experience frequent or chronic excessive worry, nervousness, a sense of unease, a feeling of being unfocused, and distress, which result in functional impairment.¹ Frequently, anxiety is accompanied by restlessness or muscle tension. Generalized anxiety disorder is one of the most common psychiatric diagnoses in the United States and has a prevalence of 2% to 6% globally.² Although research has been conducted regarding anxiety’s pathogenesis, to date a firm consensus on its etiology has not been reached.³ It is likely multifactorial, with environmental and biologic components.

One area of focus has been neurotransmitters and the possible role they play in the pathogenesis of anxiety. Specifically, the monoamine neurotransmitters have been implicated in the clinical manifestations of anxiety. Among the amines, normal roles include stimulating the autonomic nervous system and regulating numerous cognitive phenomena, such as volition and emotion. Many psychiatric medications modify aminergic transmission, and many current anxiety medications target amine neurotransmitters. Medications that target histamine, serotonin, norepinephrine, and dopamine all play a role in treating anxiety.

In this article, we focus on serotonin (5-hydroxytryptamine, 5-HT) as a mediator of anxiety and on excessive synaptic 5-HT as the cause of anxiety. We discuss how 5-HT–mediated anxiety can be identified and offer some solutions for its treatment.

The amine neurotransmitters

There are 6 amine neurotransmitters in the CNS. These are derived from tyrosine (dopamine [DA], norepinephrine [NE], and epinephrine), histidine (histamine), and tryptophan (serotonin [5-HT] and melatonin). In addition to their physiologic actions, amines have been implicated in both acute and chronic anxiety. Excessive DA stimulation has been linked with fear^4,5; NE elevations are central to hypervigilance and hyperarousal of posttraumatic stress disorder⁶; and histamine may mediate emotional memories involved in fear and anxiety.⁷ Understanding the normal function of 5-HT will aid in understanding its potential problematic role (Box,^8-18page 38).

How serotonin-mediated anxiety presents

“Anxiety” is a collection of signs and symptoms that likely represent multiple processes and have the common characteristic of being subjectively unpleasant, with a subjective wish for the feeling to end. The expression of anxiety disorders is quite diverse and ranges from brief episodes such as panic attacks (which may be mediated, in part, by epinephrine/NE¹⁹) to lifelong stereotypic obsessions and compulsions (which may be mediated, in part, by DA and modified by 5-HT^20,21). Biochemical separation of the anxiety disorders is key to achieving tailored treatment.⁶ Towards this end, it is important to investigate the phenomenon of serotonin-mediated anxiety.

Because clinicians are familiar with reductions of anxiety as selective serotonin reuptake inhibitors (SSRIs) increase 5-HT levels in the synapse, it is difficult to conceptualize serotonin-mediated anxiety. However, many of the effects at postsynaptic 5-HT receptors may be biphasic.^15-18 Serotonin-mediated anxiety appears to occur when levels of 5-HT (or stimulation of 5-HT receptors) are particularly high. This is most frequently seen in patients who genetically have high synaptic 5-HT (by virtue of the short form of the 5-HT transporter),²² whose synaptic 5-HT is further increased by treatment with an SSRI,²³ and who are experiencing a stressor that yet further increases their synaptic 5-HT.²⁴ However, it may occur in some individuals with only 2 of these 3 conditions.Clinically, individuals with serotonin-mediated anxiety will usually appear calm. The anxiety they are experiencing is not exhibited in any way in the motor system (ie, they do not appear restless, do not pace, muscle tone is not increased, etc.). However, they will generally complain of an internal agitation, a sense of a negative internal energy. Frequently, they will use descriptions such as “I feel I could jump out of my skin.” As previously mentioned, this is usually in the setting of some environmental stress, in addition to either a pharmacologic (SSRI) or genetic (short form of the 5-HT transporter) reason for increasing synaptic 5-HT, or both.

Almost always, interventions that block multiple postsynaptic 5-HT receptors or discontinuation of the SSRI (if applicable) will alleviate the anxiety, quickly or more slowly, respectively. Sublingual asenapine, which at low doses can block 5-HT2C (Ki = 0.03 nM), 5-HT2A (Ki = 0.07 nM), 5-HT7 (Ki = 0.11 nM), 5-HT2B (Ki = 0.18 nM), and 5-HT6 (Ki = 0.25 nM),^25,26 and which will produce peak plasma levels within 10 minutes,²⁷ usually is quite effective.

Box

Continue to: A key difference: No motor system involvement...

A key difference: No motor system involvement

What distinguishes 5-HT from the other amine transmitters as a mediator of anxiety is the lack of involvement of the motor system. Multiple studies in rats illustrate that exogenously augmenting 5-HT has no effect on levels of locomotor activity. Dopamine depletion is well-characterized in the motor dysfunction of Parkinson’s disease, and DA excess can cause repetitive, stereotyped movements, such as seen in tardive dyskinesia or Huntington’s disease.⁸ In humans, serotonin-mediated anxiety is usually without a motoric component; patients appear calm but complain of extreme anxiety or agitation. Agitation has been reported after initiation of an SSRI,²⁹ and is more likely to occur in patients with the short form of the 5-HT transporter.³⁰ Motoric activation has been reported in some of these studies, but does not seem to cluster with the complaint of agitation.²⁹ The reduced number of available transporters means a chronic steady-state elevation of serotonin, because less serotonin is being removed from the synapse after it is released. This is one of the reasons patients with the short form of the 5-HT transporter may be more susceptible to serotonin-mediated anxiety.

What you need to keep in mind

Pharmacologic treatment of anxiety begins with an SSRI, a serotonin-norepinephrine reuptake inhibitor (SNRI), or buspirone. Second-line treatments include hydroxyzine, gabapentin, pregabalin, and quetiapine.^3,31 However, clinicians need to be aware that a fraction of their patients will report anxiety that will not have any external manifestations, but will be experienced as an unpleasant internal energy. These patients may report an increase in their anxiety levels when started on an SSRI or SNRI.^29,30 This anxiety is most likely mediated by increases of synaptic 5-HT. This occurs because many serotonergic receptors may have a biphasic response, so that too much stimulation is experienced as excessive internal energy.^16-18 In such patients, blockade of key 5-HT receptors may reduce that internal agitation. The advantage of recognizing serotonin-mediated anxiety is that one can specifically tailor treatment to address the patient’s specific physiology.

It is important to note that the anxiolytic effect of asenapine is specific to patients with serotonin-mediated anxiety. Unlike quetiapine, which is effective as augmentation therapy in generalized anxiety disorder,³¹ asenapine does not appear to reduce anxiety in patients with schizophrenia³² or borderline personality disorder³³ when administered for other reasons. However, it may reduce anxiety in patients with the short form of the 5-HT transporter.^30,34

Bottom Line

Serotonin-mediated anxiety occurs when levels of synaptic serotonin (5-HT) are high. Patients with serotonin-mediated anxiety appear calm but will report experiencing an unpleasant internal energy. Interventions that block multiple postsynaptic 5-HT receptors or discontinuation of a selective serotonin reuptake inhibitor (if applicable) will alleviate the anxiety.

Related Resource

• Bhatt NV. Anxiety disorders. https://emedicine.medscape. com/article/286227-overview

Drug Brand Names

Asenapine • Saphris, Secuado

Gabapentin • Neurontin

Hydroxyzine • Vistaril

Pregabalin • Lyrica

Quetiapine • Seroquel

Trazodone • Oleptro

Sara R. Abell, MD, and Rif S. El-Mallakh, MD

Individuals with anxiety will experience frequent or chronic excessive worry, nervousness, a sense of unease, a feeling of being unfocused, and distress, which result in functional impairment.¹ Frequently, anxiety is accompanied by restlessness or muscle tension. Generalized anxiety disorder is one of the most common psychiatric diagnoses in the United States and has a prevalence of 2% to 6% globally.² Although research has been conducted regarding anxiety’s pathogenesis, to date a firm consensus on its etiology has not been reached.³ It is likely multifactorial, with environmental and biologic components.

One area of focus has been neurotransmitters and the possible role they play in the pathogenesis of anxiety. Specifically, the monoamine neurotransmitters have been implicated in the clinical manifestations of anxiety. Among the amines, normal roles include stimulating the autonomic nervous system and regulating numerous cognitive phenomena, such as volition and emotion. Many psychiatric medications modify aminergic transmission, and many current anxiety medications target amine neurotransmitters. Medications that target histamine, serotonin, norepinephrine, and dopamine all play a role in treating anxiety.

In this article, we focus on serotonin (5-hydroxytryptamine, 5-HT) as a mediator of anxiety and on excessive synaptic 5-HT as the cause of anxiety. We discuss how 5-HT–mediated anxiety can be identified and offer some solutions for its treatment.

The amine neurotransmitters

There are 6 amine neurotransmitters in the CNS. These are derived from tyrosine (dopamine [DA], norepinephrine [NE], and epinephrine), histidine (histamine), and tryptophan (serotonin [5-HT] and melatonin). In addition to their physiologic actions, amines have been implicated in both acute and chronic anxiety. Excessive DA stimulation has been linked with fear^4,5; NE elevations are central to hypervigilance and hyperarousal of posttraumatic stress disorder⁶; and histamine may mediate emotional memories involved in fear and anxiety.⁷ Understanding the normal function of 5-HT will aid in understanding its potential problematic role (Box,^8-18page 38).

How serotonin-mediated anxiety presents

“Anxiety” is a collection of signs and symptoms that likely represent multiple processes and have the common characteristic of being subjectively unpleasant, with a subjective wish for the feeling to end. The expression of anxiety disorders is quite diverse and ranges from brief episodes such as panic attacks (which may be mediated, in part, by epinephrine/NE¹⁹) to lifelong stereotypic obsessions and compulsions (which may be mediated, in part, by DA and modified by 5-HT^20,21). Biochemical separation of the anxiety disorders is key to achieving tailored treatment.⁶ Towards this end, it is important to investigate the phenomenon of serotonin-mediated anxiety.

Because clinicians are familiar with reductions of anxiety as selective serotonin reuptake inhibitors (SSRIs) increase 5-HT levels in the synapse, it is difficult to conceptualize serotonin-mediated anxiety. However, many of the effects at postsynaptic 5-HT receptors may be biphasic.^15-18 Serotonin-mediated anxiety appears to occur when levels of 5-HT (or stimulation of 5-HT receptors) are particularly high. This is most frequently seen in patients who genetically have high synaptic 5-HT (by virtue of the short form of the 5-HT transporter),²² whose synaptic 5-HT is further increased by treatment with an SSRI,²³ and who are experiencing a stressor that yet further increases their synaptic 5-HT.²⁴ However, it may occur in some individuals with only 2 of these 3 conditions.Clinically, individuals with serotonin-mediated anxiety will usually appear calm. The anxiety they are experiencing is not exhibited in any way in the motor system (ie, they do not appear restless, do not pace, muscle tone is not increased, etc.). However, they will generally complain of an internal agitation, a sense of a negative internal energy. Frequently, they will use descriptions such as “I feel I could jump out of my skin.” As previously mentioned, this is usually in the setting of some environmental stress, in addition to either a pharmacologic (SSRI) or genetic (short form of the 5-HT transporter) reason for increasing synaptic 5-HT, or both.

Almost always, interventions that block multiple postsynaptic 5-HT receptors or discontinuation of the SSRI (if applicable) will alleviate the anxiety, quickly or more slowly, respectively. Sublingual asenapine, which at low doses can block 5-HT2C (Ki = 0.03 nM), 5-HT2A (Ki = 0.07 nM), 5-HT7 (Ki = 0.11 nM), 5-HT2B (Ki = 0.18 nM), and 5-HT6 (Ki = 0.25 nM),^25,26 and which will produce peak plasma levels within 10 minutes,²⁷ usually is quite effective.

Box

Continue to: A key difference: No motor system involvement...

A key difference: No motor system involvement

What distinguishes 5-HT from the other amine transmitters as a mediator of anxiety is the lack of involvement of the motor system. Multiple studies in rats illustrate that exogenously augmenting 5-HT has no effect on levels of locomotor activity. Dopamine depletion is well-characterized in the motor dysfunction of Parkinson’s disease, and DA excess can cause repetitive, stereotyped movements, such as seen in tardive dyskinesia or Huntington’s disease.⁸ In humans, serotonin-mediated anxiety is usually without a motoric component; patients appear calm but complain of extreme anxiety or agitation. Agitation has been reported after initiation of an SSRI,²⁹ and is more likely to occur in patients with the short form of the 5-HT transporter.³⁰ Motoric activation has been reported in some of these studies, but does not seem to cluster with the complaint of agitation.²⁹ The reduced number of available transporters means a chronic steady-state elevation of serotonin, because less serotonin is being removed from the synapse after it is released. This is one of the reasons patients with the short form of the 5-HT transporter may be more susceptible to serotonin-mediated anxiety.

What you need to keep in mind

Pharmacologic treatment of anxiety begins with an SSRI, a serotonin-norepinephrine reuptake inhibitor (SNRI), or buspirone. Second-line treatments include hydroxyzine, gabapentin, pregabalin, and quetiapine.^3,31 However, clinicians need to be aware that a fraction of their patients will report anxiety that will not have any external manifestations, but will be experienced as an unpleasant internal energy. These patients may report an increase in their anxiety levels when started on an SSRI or SNRI.^29,30 This anxiety is most likely mediated by increases of synaptic 5-HT. This occurs because many serotonergic receptors may have a biphasic response, so that too much stimulation is experienced as excessive internal energy.^16-18 In such patients, blockade of key 5-HT receptors may reduce that internal agitation. The advantage of recognizing serotonin-mediated anxiety is that one can specifically tailor treatment to address the patient’s specific physiology.

It is important to note that the anxiolytic effect of asenapine is specific to patients with serotonin-mediated anxiety. Unlike quetiapine, which is effective as augmentation therapy in generalized anxiety disorder,³¹ asenapine does not appear to reduce anxiety in patients with schizophrenia³² or borderline personality disorder³³ when administered for other reasons. However, it may reduce anxiety in patients with the short form of the 5-HT transporter.^30,34

Bottom Line

Serotonin-mediated anxiety occurs when levels of synaptic serotonin (5-HT) are high. Patients with serotonin-mediated anxiety appear calm but will report experiencing an unpleasant internal energy. Interventions that block multiple postsynaptic 5-HT receptors or discontinuation of a selective serotonin reuptake inhibitor (if applicable) will alleviate the anxiety.

Related Resource

• Bhatt NV. Anxiety disorders. https://emedicine.medscape. com/article/286227-overview

Drug Brand Names

Asenapine • Saphris, Secuado

Gabapentin • Neurontin

Hydroxyzine • Vistaril

Pregabalin • Lyrica

Quetiapine • Seroquel

Trazodone • Oleptro

FIRST OF 2 PARTS

Borderline personality disorder (BPD) is marked by an ongoing pattern of mood instability, cognitive distortions, problems with self-image, and impulsive behavior, often resulting in problems in relationships. BPD is associated with serious impairment in psychosocial functioning.¹ Patients with BPD tend to use more mental health services than patients with other personality disorders or those with major depressive disorder (MDD).² However, there has been little consensus on the best treatment(s) for this serious and debilitating disorder, and some clinicians view BPD as difficult to treat.

Current treatments for BPD include psychological and pharmacological interventions. Neuromodulation techniques, such as repetitive transcranial magnetic stimulation, may also positively affect BPD symptomatology. In recent years, there have been some promising findings in the treatment of BPD. In this 2-part article, we focus on current (within the last 5 years) findings from randomized controlled trials (RCTs) of BPD treatments. Here in Part 1, we focus on 6 studies that evaluated biological interventions (Table,^3-8). In Part 2, we will focus on RCTs that investigated psychological interventions.

1. Lisoni J, Miotto P, Barlati S, et al. Change in core symptoms of borderline personality disorder by tDCS: a pilot study. Psychiatry Res. 2020;291:113261. doi: 10.1016/j.psychres.2020.113261

Impulsivity has been described as the core feature of BPD that best explains its behavioral, cognitive, and clinical manifestations. Studies have repeatedly demonstrated the role of the prefrontal cortex in modulating impulsivity. Dysfunction of the dorsolateral prefrontal cortex (DLPFC) has been implicated in BPD. DLPFC transcranial direct current stimulation (tDCS) is a well-tolerated, noninvasive neurostimulation technique that can be used to alter cortical brain activity. Lisoni et al³ examined whether a bilateral right anodal/left cathodal tDCS montage could modulate the psychopathology of BPD.

Continue to: Study design...

• In a double-blind, sham-controlled trial, adults who met DSM-IV-TR criteria for BPD were randomized to 3 weeks (15 sessions) of right anodal/left cathodal DLPFC tCDS (n = 15) or sham tDCS (n = 15). This study included patients with comorbid psychiatric disorders, including substance use disorders. Discontinuation or alteration of existing medications was not allowed.
• The presence, severity, and change over time of BPD core symptoms was assessed at baseline and after 3 weeks using several clinical scales, self-questionnaires, and neuropsychological tests, including the Barratt Impulsiveness Scale-11 (BIS-11), Buss-Perry Aggression Questionnaire (BP-AQ), Difficulties in Emotion Regulation Scale (DERS), Hamilton Depression Rating Scale (HAM-D), Beck Depression Inventory (BDI), Hamilton Anxiety Rating Scale (HAM-A), Irritability-Depression Anxiety Scale (IDA), Visual Analog Scales (VAS), and Iowa Gambling Task.

Outcomes

• Participants in the active tDCS group experienced significant reductions in impulsivity, aggression, and craving as measured by the BIS-11, BP-AQ, and VAS.
• Compared to the sham group, the active tDCS group had greater reductions in HAM-D and BDI scores.
• HAM-A and IDA scores were improved in both groups, although the active tDCS group showed greater reductions in IDA scores compared with the sham group.
• As measured by DERS, active tDCS did not improve affective dysregulation more than sham tDCS.

Conclusions/limitations

• Bilateral tDCS targeting the right DLPFC with anodal stimulation is a safe, well-tolerated technique that may modulate core dimensions of BPD, including impulsivity, aggression, and craving.
• Excitatory anodal stimulation of the right DLFPC coupled with inhibitory cathodal stimulation on the left DLPFC may be an effective montage for targeting impulsivity in patients with BPD.
• Study limitations include a small sample size, use of targeted questionnaires only, inclusion of patients with BPD who also had certain comorbid psychiatric disorders, lack of analysis of the contributions of medications, lack of functional neuroimaging, and lack of a follow-up phase.

2. Molavi P, Aziziaram S, Basharpoor S, et al. Repeated transcranial direct current stimulation of dorsolateral-prefrontal cortex improves executive functions, cognitive reappraisal emotion regulation, and control over emotional processing in borderline personality disorder: a randomized, sham-controlled, parallel-group study. J Affect Disord. 2020;274:93-102. doi: 10.1016/j.jad.2020.05.007

Emotional dysregulation is considered a core feature of BPD psychopathology and is closely associated with executive dysfunction and cognitive control. Manifestations of executive dysfunction include aggressiveness, impulsive decision-making, disinhibition, and self-destructive behaviors. Neuroimaging of patients with BPD has shown enhanced activity in the insula, posterior cingulate cortex, and amygdala, with reduced activity in the medial PFC, subgenual anterior cingulate cortex, and DLPFC. Molavi et al⁴ postulated that increasing DLPFC activation with left anodal tDCS would result in improved executive functioning and emotion dysregulation in patients with BPD.

Study design

• In this single-blind, sham-controlled, parallel-group study, adults who met DSM-5 criteria for BPD were randomized to receive 10 consecutive daily sessions of left anodal/right cathodal DLPFC tDCS (n = 16) or sham tDCS (n = 16).
• The effect of tDCS on executive dysfunction, emotion dysregulation, and emotional processing was measured using the Executive Skills Questionnaire for Adults (ESQ), Emotion Regulation Questionnaire (ERQ), and Emotional Processing Scale (EPS). Measurements occurred at baseline and after 10 sessions of active or sham tDCS.

Outcomes

• Participants who received active tDCS experienced significant improvements in ESQ overall score and most of the executive function domains measured by the ESQ.
• Those in the active tDCS group also experienced significant improvement in emotion regulation as measured by the cognitive reappraisal subscale (but not the expressive suppression subscale) of the ERQ after the intervention.
• Overall emotional processing as measured by the EPS was significantly improved in the active tDCS group following the intervention.

Conclusions/limitations

• Repeated bilateral left anodal/right cathodal tDCS stimulation of the DLPFC significantly improved executive functioning and aspects of emotion regulation and emotional processing in patients with BPD. This improvement was presumed to be the result of increased activity of left DLPFC.
• Study limitations include a single-blind design, lack of follow-up to assess durability and stability of response over time, reliance on self-report measures, lack of functional neuroimaging, and limited focality of tDCS.

3. Crawford MJ, Sanatinia R, Barrett B, et al; LABILE study team. The clinical effectiveness and cost-effectiveness of lamotrigine in borderline personality disorder: a randomized placebo-controlled trial. Am J Psychiatry. 2018;175(8):756-764. doi: 10.1176/appi.ajp.2018.17091006

One of the hallmark symptoms of BPD is mood dysregulation. Current treatment guidelines recommend the use of mood stabilizers for BPD despite limited quality evidence of effectiveness and a lack of FDA-approved medications with this indication. In this RCT, Crawford et al⁵ examined whether lamotrigine is a clinically effective and cost-effective treatment for people with BPD.

Continue to: Study design...

• In this 2-arm, parallel-group, double-blind, placebo-controlled trial, 276 adults who met DSM-IV criteria for BPD were randomized to receive lamotrigine (up to 400 mg/d) or placebo for 52 weeks.
• The primary outcome was the score on the Zanarini Rating Scale for Borderline Personality Disorder (ZAN-BPD) at 52 weeks. Secondary outcomes included depressive symptoms, deliberate self-harm, social functioning, health-related quality of life, resource use and costs, treatment adverse effects, and adverse events. These were assessed using the BDI; Acts of Deliberate Self-Harm Inventory; Social Functioning Questionnaire; Alcohol, Smoking, and Substance Involvement Screening Test; and the EQ-5D-3L.

Outcomes

• Mean ZAN-BPD score decreased at 12 weeks in both groups, after which time the score remained stable.
• There was no difference in ZAN-BPD scores at 52 weeks between treatment arms. No difference was found in any secondary outcome measures.
• Difference in costs between groups was not significant.

Conclusions/limitations

• There was no evidence that lamotrigine led to clinical improvements in BPD symptomatology, social functioning, health-related quality of life, or substance use.
• Lamotrigine is neither clinically effective nor a cost-effective use of resources in the treatment of BPD.
• Limitations include a low level of adherence.

4. Domes G, Ower N, von Dawans B, et al. Effects of intranasal oxytocin administration on empathy and approach motivation in women with borderline personality disorder: a randomized controlled trial. Transl Psychiatry. 2019;9(1):328. doi: 10.1038/s41398-019-0658-4

A core feature of BPD is impairment in empathy; adequate empathy is required for intact social functioning. Oxytocin is a neuropeptide that helps regulate complex social cognition and behavior. Prior research has found that oxytocin administration enhances emotion regulation and empathy. Women with BPD have been observed to have lower levels of oxytocin. Domes et al⁶ conducted an RCT to see if oxytocin could have a beneficial effect on social approach and social cognition in women with BPD.

Study design

• In a double-blind, placebo-controlled, between-subject trial, 61 women who met DSM-IV criteria for BPD and 68 matched healthy controls were randomized to receive intranasal oxytocin, 24 IU, or placebo 45 minutes before completing an empathy task.
• An extended version of the Multifaceted Empathy Test was used to assess empathy and approach motivation.

Outcomes

• For cognitive empathy, patients with BPD exhibited significantly lower overall performance compared to controls. There was no effect of oxytocin on this performance in either group.
• Patients with BPD had significantly lower affective empathy compared with controls. After oxytocin administration, patients with BPD had significantly higher affective empathy than those with BPD who received placebo, reaching the level of healthy controls who received placebo.
• For positive stimuli, patients with BPD showed lower affective empathy than controls. Oxytocin treatment increased affective empathy in both groups.
• For negative stimuli, oxytocin increased affective empathy more in patients with BPD than in controls.
• Patients with BPD demonstrated less approach motivation than controls. Oxytocin increased approach motivation more in patients with BPD than in controls. For approach motivation toward positive stimuli, oxytocin had a significant effect on patients with BPD.

Continue to: Conclusions/limitations...

Conclusions/limitations

• Patients with BPD showed reduced cognitive and affective empathy and less approach behavior motivation than healthy controls.
• Patients with BPD who received oxytocin attained a level of affective empathy and approach motivation similar to that of healthy controls who received placebo. For positive stimuli, both groups exhibited comparable improvements from oxytocin. For negative stimuli, patients with BPD patients showed significant improvement with oxytocin, whereas healthy controls received no such benefit.
• Limitations include the use of self-report scales, lack of a control group, and inclusion of patients using psychotherapeutic medications. The study lacks generalizability because only women were included; the effect of exogenous oxytocin on men may differ.

5. Bozzatello P, Rocca P, Uscinska M, et al. Efficacy and tolerability of asenapine compared with olanzapine in borderline personality disorder: an open-label randomized controlled trial. CNS Drugs. 2017;31(9):809-819. doi: 10.1007/s40263-017-0458-4

The last decade has seen a noticeable shift in clinical practice from the use of antidepressants to mood stabilizers and second-generation antipsychotics (SGAs) in the treatment of BPD. Studies have demonstrated therapeutic effects of antipsychotic drugs across a wide range of BPD symptoms. Among SGAs, olanzapine is the most extensively studied across case reports, open-label studies, and RCTs of patients with BPD. In an RCT, Bozzatello et al⁷ compared the efficacy and tolerability of asenapine to olanzapine.

Study design

• In this open-label RCT, adults who met DSM-5 criteria for BPD were assigned to receive asenapine (n = 25) or olanzapine (n = 26) for 12 weeks.
• Study measurements included the Clinical Global Impression Scale, Severity item, HAM-D, HAM-A, Social and Occupational Functioning Assessment Scale, Borderline Personality Disorder Severity Index (BPDSI), BIS-11, Modified Overt Aggression Scale, and Dosage Record Treatment Emergent Symptom Scale.

Outcomes

• Asenapine and olanzapine had similar effects on BPD-related psychopathology, anxiety, and social and occupational functioning.
• Neither medication significantly decreased depressive or aggressive symptoms.
• Asenapine was superior to olanzapine in reducing the affective instability score of the BPDSI.
• Akathisia and restlessness/anxiety were more common with asenapine, and somnolence and fatigue were more common with olanzapine.

Conclusions/limitations

• The overall efficacy of asenapine was not different from olanzapine, and both medications were well-tolerated.
• Neither medication led to an improvement in depression or aggression, but asenapine was superior to olanzapine in reducing the severity of affective instability.
• Limitations include an open-label design, lack of placebo group, small sample size, high drop-out rate, exclusion of participants with co-occurring MDD and substance abuse/dependence, lack of data on prior pharmacotherapies and psychotherapies, and lack of power to detect a difference on the dissociation/paranoid ideation item of BPDSI.

6. Kulkarni J, Thomas N, Hudaib AR, et al. Effect of the glutamate NMDA receptor antagonist memantine as adjunctive treatment in borderline personality disorder: an exploratory, randomised, double-blind, placebo-controlled trial. CNS Drugs. 2018;32(2):179-187. doi: 10.1007/s40263-018-0506-8

It has been hypothesized that glutamate dysregulation and excitotoxicity are crucial to the development of the cognitive disturbances that underlie BPD. As such, glutamate modulators such as memantine hold promise for the treatment of BPD. In this RCT, Kulkarni et al⁸ examined the efficacy and tolerability of memantine compared with treatment as usual in patients with BPD.

Continue to: Study design...

• In an 8-week, double-blind, placebo-controlled trial, adults diagnosed with BPD according to the Diagnostic Interview for Borderline Patients were randomized to receive memantine (n = 17) or placebo (n = 16) in addition to treatment as usual. Treatment as usual included the use of antidepressants, mood stabilizers, and antipsychotics as well as psychotherapy and other psychosocial interventions.
• Patients were initiated on placebo or memantine, 10 mg/d. Memantine was increased to 20 mg/d after 7 days.
• ZAN-BPD score was the primary outcome and was measured at baseline and 2, 4, 6, and 8 weeks. An adverse effects questionnaire was administered every 2 weeks to assess tolerability.

Outcomes

• During the first 2 weeks of treatment, there were no significant improvements in ZAN-BPD score in the memantine group compared with the placebo group.
• Beginning with Week 2, compared with the placebo group, the memantine group experienced a significant reduction in total symptoms as measured by ZAN-BPD.
• There were no statistically significant differences in adverse events between groups.

Conclusions/limitations

• Memantine appears to be a well-tolerated treatment option for patients with BPD and merits further study.
• Limitations include a small sample size, and an inability to reach plateau of ZAN-BPD total score in either group. Also, there is considerable individual variability in memantine steady-state plasma concentrations, but plasma levels were not measured in this study.

Bottom Line

Findings from small randomized controlled trials suggest that transcranial direct current stimulation, oxytocin, asenapine, olanzapine, and memantine may have beneficial effects on some core symptoms of borderline personality disorder. These findings need to be replicated in larger studies.

FIRST OF 2 PARTS

Borderline personality disorder (BPD) is marked by an ongoing pattern of mood instability, cognitive distortions, problems with self-image, and impulsive behavior, often resulting in problems in relationships. BPD is associated with serious impairment in psychosocial functioning.¹ Patients with BPD tend to use more mental health services than patients with other personality disorders or those with major depressive disorder (MDD).² However, there has been little consensus on the best treatment(s) for this serious and debilitating disorder, and some clinicians view BPD as difficult to treat.

Current treatments for BPD include psychological and pharmacological interventions. Neuromodulation techniques, such as repetitive transcranial magnetic stimulation, may also positively affect BPD symptomatology. In recent years, there have been some promising findings in the treatment of BPD. In this 2-part article, we focus on current (within the last 5 years) findings from randomized controlled trials (RCTs) of BPD treatments. Here in Part 1, we focus on 6 studies that evaluated biological interventions (Table,^3-8). In Part 2, we will focus on RCTs that investigated psychological interventions.

1. Lisoni J, Miotto P, Barlati S, et al. Change in core symptoms of borderline personality disorder by tDCS: a pilot study. Psychiatry Res. 2020;291:113261. doi: 10.1016/j.psychres.2020.113261

Impulsivity has been described as the core feature of BPD that best explains its behavioral, cognitive, and clinical manifestations. Studies have repeatedly demonstrated the role of the prefrontal cortex in modulating impulsivity. Dysfunction of the dorsolateral prefrontal cortex (DLPFC) has been implicated in BPD. DLPFC transcranial direct current stimulation (tDCS) is a well-tolerated, noninvasive neurostimulation technique that can be used to alter cortical brain activity. Lisoni et al³ examined whether a bilateral right anodal/left cathodal tDCS montage could modulate the psychopathology of BPD.

Continue to: Study design...

• In a double-blind, sham-controlled trial, adults who met DSM-IV-TR criteria for BPD were randomized to 3 weeks (15 sessions) of right anodal/left cathodal DLPFC tCDS (n = 15) or sham tDCS (n = 15). This study included patients with comorbid psychiatric disorders, including substance use disorders. Discontinuation or alteration of existing medications was not allowed.
• The presence, severity, and change over time of BPD core symptoms was assessed at baseline and after 3 weeks using several clinical scales, self-questionnaires, and neuropsychological tests, including the Barratt Impulsiveness Scale-11 (BIS-11), Buss-Perry Aggression Questionnaire (BP-AQ), Difficulties in Emotion Regulation Scale (DERS), Hamilton Depression Rating Scale (HAM-D), Beck Depression Inventory (BDI), Hamilton Anxiety Rating Scale (HAM-A), Irritability-Depression Anxiety Scale (IDA), Visual Analog Scales (VAS), and Iowa Gambling Task.

Outcomes

• Participants in the active tDCS group experienced significant reductions in impulsivity, aggression, and craving as measured by the BIS-11, BP-AQ, and VAS.
• Compared to the sham group, the active tDCS group had greater reductions in HAM-D and BDI scores.
• HAM-A and IDA scores were improved in both groups, although the active tDCS group showed greater reductions in IDA scores compared with the sham group.
• As measured by DERS, active tDCS did not improve affective dysregulation more than sham tDCS.

Conclusions/limitations

• Bilateral tDCS targeting the right DLPFC with anodal stimulation is a safe, well-tolerated technique that may modulate core dimensions of BPD, including impulsivity, aggression, and craving.
• Excitatory anodal stimulation of the right DLFPC coupled with inhibitory cathodal stimulation on the left DLPFC may be an effective montage for targeting impulsivity in patients with BPD.
• Study limitations include a small sample size, use of targeted questionnaires only, inclusion of patients with BPD who also had certain comorbid psychiatric disorders, lack of analysis of the contributions of medications, lack of functional neuroimaging, and lack of a follow-up phase.

2. Molavi P, Aziziaram S, Basharpoor S, et al. Repeated transcranial direct current stimulation of dorsolateral-prefrontal cortex improves executive functions, cognitive reappraisal emotion regulation, and control over emotional processing in borderline personality disorder: a randomized, sham-controlled, parallel-group study. J Affect Disord. 2020;274:93-102. doi: 10.1016/j.jad.2020.05.007

Emotional dysregulation is considered a core feature of BPD psychopathology and is closely associated with executive dysfunction and cognitive control. Manifestations of executive dysfunction include aggressiveness, impulsive decision-making, disinhibition, and self-destructive behaviors. Neuroimaging of patients with BPD has shown enhanced activity in the insula, posterior cingulate cortex, and amygdala, with reduced activity in the medial PFC, subgenual anterior cingulate cortex, and DLPFC. Molavi et al⁴ postulated that increasing DLPFC activation with left anodal tDCS would result in improved executive functioning and emotion dysregulation in patients with BPD.

Study design

• In this single-blind, sham-controlled, parallel-group study, adults who met DSM-5 criteria for BPD were randomized to receive 10 consecutive daily sessions of left anodal/right cathodal DLPFC tDCS (n = 16) or sham tDCS (n = 16).
• The effect of tDCS on executive dysfunction, emotion dysregulation, and emotional processing was measured using the Executive Skills Questionnaire for Adults (ESQ), Emotion Regulation Questionnaire (ERQ), and Emotional Processing Scale (EPS). Measurements occurred at baseline and after 10 sessions of active or sham tDCS.

Outcomes

• Participants who received active tDCS experienced significant improvements in ESQ overall score and most of the executive function domains measured by the ESQ.
• Those in the active tDCS group also experienced significant improvement in emotion regulation as measured by the cognitive reappraisal subscale (but not the expressive suppression subscale) of the ERQ after the intervention.
• Overall emotional processing as measured by the EPS was significantly improved in the active tDCS group following the intervention.

Conclusions/limitations

• Repeated bilateral left anodal/right cathodal tDCS stimulation of the DLPFC significantly improved executive functioning and aspects of emotion regulation and emotional processing in patients with BPD. This improvement was presumed to be the result of increased activity of left DLPFC.
• Study limitations include a single-blind design, lack of follow-up to assess durability and stability of response over time, reliance on self-report measures, lack of functional neuroimaging, and limited focality of tDCS.

3. Crawford MJ, Sanatinia R, Barrett B, et al; LABILE study team. The clinical effectiveness and cost-effectiveness of lamotrigine in borderline personality disorder: a randomized placebo-controlled trial. Am J Psychiatry. 2018;175(8):756-764. doi: 10.1176/appi.ajp.2018.17091006

One of the hallmark symptoms of BPD is mood dysregulation. Current treatment guidelines recommend the use of mood stabilizers for BPD despite limited quality evidence of effectiveness and a lack of FDA-approved medications with this indication. In this RCT, Crawford et al⁵ examined whether lamotrigine is a clinically effective and cost-effective treatment for people with BPD.

Continue to: Study design...

• In this 2-arm, parallel-group, double-blind, placebo-controlled trial, 276 adults who met DSM-IV criteria for BPD were randomized to receive lamotrigine (up to 400 mg/d) or placebo for 52 weeks.
• The primary outcome was the score on the Zanarini Rating Scale for Borderline Personality Disorder (ZAN-BPD) at 52 weeks. Secondary outcomes included depressive symptoms, deliberate self-harm, social functioning, health-related quality of life, resource use and costs, treatment adverse effects, and adverse events. These were assessed using the BDI; Acts of Deliberate Self-Harm Inventory; Social Functioning Questionnaire; Alcohol, Smoking, and Substance Involvement Screening Test; and the EQ-5D-3L.

Outcomes

• Mean ZAN-BPD score decreased at 12 weeks in both groups, after which time the score remained stable.
• There was no difference in ZAN-BPD scores at 52 weeks between treatment arms. No difference was found in any secondary outcome measures.
• Difference in costs between groups was not significant.

Conclusions/limitations

• There was no evidence that lamotrigine led to clinical improvements in BPD symptomatology, social functioning, health-related quality of life, or substance use.
• Lamotrigine is neither clinically effective nor a cost-effective use of resources in the treatment of BPD.
• Limitations include a low level of adherence.

4. Domes G, Ower N, von Dawans B, et al. Effects of intranasal oxytocin administration on empathy and approach motivation in women with borderline personality disorder: a randomized controlled trial. Transl Psychiatry. 2019;9(1):328. doi: 10.1038/s41398-019-0658-4

A core feature of BPD is impairment in empathy; adequate empathy is required for intact social functioning. Oxytocin is a neuropeptide that helps regulate complex social cognition and behavior. Prior research has found that oxytocin administration enhances emotion regulation and empathy. Women with BPD have been observed to have lower levels of oxytocin. Domes et al⁶ conducted an RCT to see if oxytocin could have a beneficial effect on social approach and social cognition in women with BPD.

Study design

• In a double-blind, placebo-controlled, between-subject trial, 61 women who met DSM-IV criteria for BPD and 68 matched healthy controls were randomized to receive intranasal oxytocin, 24 IU, or placebo 45 minutes before completing an empathy task.
• An extended version of the Multifaceted Empathy Test was used to assess empathy and approach motivation.

Outcomes

• For cognitive empathy, patients with BPD exhibited significantly lower overall performance compared to controls. There was no effect of oxytocin on this performance in either group.
• Patients with BPD had significantly lower affective empathy compared with controls. After oxytocin administration, patients with BPD had significantly higher affective empathy than those with BPD who received placebo, reaching the level of healthy controls who received placebo.
• For positive stimuli, patients with BPD showed lower affective empathy than controls. Oxytocin treatment increased affective empathy in both groups.
• For negative stimuli, oxytocin increased affective empathy more in patients with BPD than in controls.
• Patients with BPD demonstrated less approach motivation than controls. Oxytocin increased approach motivation more in patients with BPD than in controls. For approach motivation toward positive stimuli, oxytocin had a significant effect on patients with BPD.

Continue to: Conclusions/limitations...

Conclusions/limitations

• Patients with BPD showed reduced cognitive and affective empathy and less approach behavior motivation than healthy controls.
• Patients with BPD who received oxytocin attained a level of affective empathy and approach motivation similar to that of healthy controls who received placebo. For positive stimuli, both groups exhibited comparable improvements from oxytocin. For negative stimuli, patients with BPD patients showed significant improvement with oxytocin, whereas healthy controls received no such benefit.
• Limitations include the use of self-report scales, lack of a control group, and inclusion of patients using psychotherapeutic medications. The study lacks generalizability because only women were included; the effect of exogenous oxytocin on men may differ.

5. Bozzatello P, Rocca P, Uscinska M, et al. Efficacy and tolerability of asenapine compared with olanzapine in borderline personality disorder: an open-label randomized controlled trial. CNS Drugs. 2017;31(9):809-819. doi: 10.1007/s40263-017-0458-4

The last decade has seen a noticeable shift in clinical practice from the use of antidepressants to mood stabilizers and second-generation antipsychotics (SGAs) in the treatment of BPD. Studies have demonstrated therapeutic effects of antipsychotic drugs across a wide range of BPD symptoms. Among SGAs, olanzapine is the most extensively studied across case reports, open-label studies, and RCTs of patients with BPD. In an RCT, Bozzatello et al⁷ compared the efficacy and tolerability of asenapine to olanzapine.

Study design

• In this open-label RCT, adults who met DSM-5 criteria for BPD were assigned to receive asenapine (n = 25) or olanzapine (n = 26) for 12 weeks.
• Study measurements included the Clinical Global Impression Scale, Severity item, HAM-D, HAM-A, Social and Occupational Functioning Assessment Scale, Borderline Personality Disorder Severity Index (BPDSI), BIS-11, Modified Overt Aggression Scale, and Dosage Record Treatment Emergent Symptom Scale.

Outcomes

• Asenapine and olanzapine had similar effects on BPD-related psychopathology, anxiety, and social and occupational functioning.
• Neither medication significantly decreased depressive or aggressive symptoms.
• Asenapine was superior to olanzapine in reducing the affective instability score of the BPDSI.
• Akathisia and restlessness/anxiety were more common with asenapine, and somnolence and fatigue were more common with olanzapine.

Conclusions/limitations

• The overall efficacy of asenapine was not different from olanzapine, and both medications were well-tolerated.
• Neither medication led to an improvement in depression or aggression, but asenapine was superior to olanzapine in reducing the severity of affective instability.
• Limitations include an open-label design, lack of placebo group, small sample size, high drop-out rate, exclusion of participants with co-occurring MDD and substance abuse/dependence, lack of data on prior pharmacotherapies and psychotherapies, and lack of power to detect a difference on the dissociation/paranoid ideation item of BPDSI.

6. Kulkarni J, Thomas N, Hudaib AR, et al. Effect of the glutamate NMDA receptor antagonist memantine as adjunctive treatment in borderline personality disorder: an exploratory, randomised, double-blind, placebo-controlled trial. CNS Drugs. 2018;32(2):179-187. doi: 10.1007/s40263-018-0506-8

It has been hypothesized that glutamate dysregulation and excitotoxicity are crucial to the development of the cognitive disturbances that underlie BPD. As such, glutamate modulators such as memantine hold promise for the treatment of BPD. In this RCT, Kulkarni et al⁸ examined the efficacy and tolerability of memantine compared with treatment as usual in patients with BPD.

Continue to: Study design...

• In an 8-week, double-blind, placebo-controlled trial, adults diagnosed with BPD according to the Diagnostic Interview for Borderline Patients were randomized to receive memantine (n = 17) or placebo (n = 16) in addition to treatment as usual. Treatment as usual included the use of antidepressants, mood stabilizers, and antipsychotics as well as psychotherapy and other psychosocial interventions.
• Patients were initiated on placebo or memantine, 10 mg/d. Memantine was increased to 20 mg/d after 7 days.
• ZAN-BPD score was the primary outcome and was measured at baseline and 2, 4, 6, and 8 weeks. An adverse effects questionnaire was administered every 2 weeks to assess tolerability.

Outcomes

• During the first 2 weeks of treatment, there were no significant improvements in ZAN-BPD score in the memantine group compared with the placebo group.
• Beginning with Week 2, compared with the placebo group, the memantine group experienced a significant reduction in total symptoms as measured by ZAN-BPD.
• There were no statistically significant differences in adverse events between groups.

Conclusions/limitations

• Memantine appears to be a well-tolerated treatment option for patients with BPD and merits further study.
• Limitations include a small sample size, and an inability to reach plateau of ZAN-BPD total score in either group. Also, there is considerable individual variability in memantine steady-state plasma concentrations, but plasma levels were not measured in this study.

Bottom Line

Findings from small randomized controlled trials suggest that transcranial direct current stimulation, oxytocin, asenapine, olanzapine, and memantine may have beneficial effects on some core symptoms of borderline personality disorder. These findings need to be replicated in larger studies.

FIRST OF 2 PARTS

Borderline personality disorder (BPD) is marked by an ongoing pattern of mood instability, cognitive distortions, problems with self-image, and impulsive behavior, often resulting in problems in relationships. BPD is associated with serious impairment in psychosocial functioning.¹ Patients with BPD tend to use more mental health services than patients with other personality disorders or those with major depressive disorder (MDD).² However, there has been little consensus on the best treatment(s) for this serious and debilitating disorder, and some clinicians view BPD as difficult to treat.

Current treatments for BPD include psychological and pharmacological interventions. Neuromodulation techniques, such as repetitive transcranial magnetic stimulation, may also positively affect BPD symptomatology. In recent years, there have been some promising findings in the treatment of BPD. In this 2-part article, we focus on current (within the last 5 years) findings from randomized controlled trials (RCTs) of BPD treatments. Here in Part 1, we focus on 6 studies that evaluated biological interventions (Table,^3-8). In Part 2, we will focus on RCTs that investigated psychological interventions.

1. Lisoni J, Miotto P, Barlati S, et al. Change in core symptoms of borderline personality disorder by tDCS: a pilot study. Psychiatry Res. 2020;291:113261. doi: 10.1016/j.psychres.2020.113261

Impulsivity has been described as the core feature of BPD that best explains its behavioral, cognitive, and clinical manifestations. Studies have repeatedly demonstrated the role of the prefrontal cortex in modulating impulsivity. Dysfunction of the dorsolateral prefrontal cortex (DLPFC) has been implicated in BPD. DLPFC transcranial direct current stimulation (tDCS) is a well-tolerated, noninvasive neurostimulation technique that can be used to alter cortical brain activity. Lisoni et al³ examined whether a bilateral right anodal/left cathodal tDCS montage could modulate the psychopathology of BPD.

Continue to: Study design...

• In a double-blind, sham-controlled trial, adults who met DSM-IV-TR criteria for BPD were randomized to 3 weeks (15 sessions) of right anodal/left cathodal DLPFC tCDS (n = 15) or sham tDCS (n = 15). This study included patients with comorbid psychiatric disorders, including substance use disorders. Discontinuation or alteration of existing medications was not allowed.
• The presence, severity, and change over time of BPD core symptoms was assessed at baseline and after 3 weeks using several clinical scales, self-questionnaires, and neuropsychological tests, including the Barratt Impulsiveness Scale-11 (BIS-11), Buss-Perry Aggression Questionnaire (BP-AQ), Difficulties in Emotion Regulation Scale (DERS), Hamilton Depression Rating Scale (HAM-D), Beck Depression Inventory (BDI), Hamilton Anxiety Rating Scale (HAM-A), Irritability-Depression Anxiety Scale (IDA), Visual Analog Scales (VAS), and Iowa Gambling Task.

Outcomes

• Participants in the active tDCS group experienced significant reductions in impulsivity, aggression, and craving as measured by the BIS-11, BP-AQ, and VAS.
• Compared to the sham group, the active tDCS group had greater reductions in HAM-D and BDI scores.
• HAM-A and IDA scores were improved in both groups, although the active tDCS group showed greater reductions in IDA scores compared with the sham group.
• As measured by DERS, active tDCS did not improve affective dysregulation more than sham tDCS.

Conclusions/limitations

• Bilateral tDCS targeting the right DLPFC with anodal stimulation is a safe, well-tolerated technique that may modulate core dimensions of BPD, including impulsivity, aggression, and craving.
• Excitatory anodal stimulation of the right DLFPC coupled with inhibitory cathodal stimulation on the left DLPFC may be an effective montage for targeting impulsivity in patients with BPD.
• Study limitations include a small sample size, use of targeted questionnaires only, inclusion of patients with BPD who also had certain comorbid psychiatric disorders, lack of analysis of the contributions of medications, lack of functional neuroimaging, and lack of a follow-up phase.

2. Molavi P, Aziziaram S, Basharpoor S, et al. Repeated transcranial direct current stimulation of dorsolateral-prefrontal cortex improves executive functions, cognitive reappraisal emotion regulation, and control over emotional processing in borderline personality disorder: a randomized, sham-controlled, parallel-group study. J Affect Disord. 2020;274:93-102. doi: 10.1016/j.jad.2020.05.007

Emotional dysregulation is considered a core feature of BPD psychopathology and is closely associated with executive dysfunction and cognitive control. Manifestations of executive dysfunction include aggressiveness, impulsive decision-making, disinhibition, and self-destructive behaviors. Neuroimaging of patients with BPD has shown enhanced activity in the insula, posterior cingulate cortex, and amygdala, with reduced activity in the medial PFC, subgenual anterior cingulate cortex, and DLPFC. Molavi et al⁴ postulated that increasing DLPFC activation with left anodal tDCS would result in improved executive functioning and emotion dysregulation in patients with BPD.

Study design

• In this single-blind, sham-controlled, parallel-group study, adults who met DSM-5 criteria for BPD were randomized to receive 10 consecutive daily sessions of left anodal/right cathodal DLPFC tDCS (n = 16) or sham tDCS (n = 16).
• The effect of tDCS on executive dysfunction, emotion dysregulation, and emotional processing was measured using the Executive Skills Questionnaire for Adults (ESQ), Emotion Regulation Questionnaire (ERQ), and Emotional Processing Scale (EPS). Measurements occurred at baseline and after 10 sessions of active or sham tDCS.

Outcomes

• Participants who received active tDCS experienced significant improvements in ESQ overall score and most of the executive function domains measured by the ESQ.
• Those in the active tDCS group also experienced significant improvement in emotion regulation as measured by the cognitive reappraisal subscale (but not the expressive suppression subscale) of the ERQ after the intervention.
• Overall emotional processing as measured by the EPS was significantly improved in the active tDCS group following the intervention.

Conclusions/limitations

• Repeated bilateral left anodal/right cathodal tDCS stimulation of the DLPFC significantly improved executive functioning and aspects of emotion regulation and emotional processing in patients with BPD. This improvement was presumed to be the result of increased activity of left DLPFC.
• Study limitations include a single-blind design, lack of follow-up to assess durability and stability of response over time, reliance on self-report measures, lack of functional neuroimaging, and limited focality of tDCS.

3. Crawford MJ, Sanatinia R, Barrett B, et al; LABILE study team. The clinical effectiveness and cost-effectiveness of lamotrigine in borderline personality disorder: a randomized placebo-controlled trial. Am J Psychiatry. 2018;175(8):756-764. doi: 10.1176/appi.ajp.2018.17091006

One of the hallmark symptoms of BPD is mood dysregulation. Current treatment guidelines recommend the use of mood stabilizers for BPD despite limited quality evidence of effectiveness and a lack of FDA-approved medications with this indication. In this RCT, Crawford et al⁵ examined whether lamotrigine is a clinically effective and cost-effective treatment for people with BPD.

Continue to: Study design...

• In this 2-arm, parallel-group, double-blind, placebo-controlled trial, 276 adults who met DSM-IV criteria for BPD were randomized to receive lamotrigine (up to 400 mg/d) or placebo for 52 weeks.
• The primary outcome was the score on the Zanarini Rating Scale for Borderline Personality Disorder (ZAN-BPD) at 52 weeks. Secondary outcomes included depressive symptoms, deliberate self-harm, social functioning, health-related quality of life, resource use and costs, treatment adverse effects, and adverse events. These were assessed using the BDI; Acts of Deliberate Self-Harm Inventory; Social Functioning Questionnaire; Alcohol, Smoking, and Substance Involvement Screening Test; and the EQ-5D-3L.

Outcomes

• Mean ZAN-BPD score decreased at 12 weeks in both groups, after which time the score remained stable.
• There was no difference in ZAN-BPD scores at 52 weeks between treatment arms. No difference was found in any secondary outcome measures.
• Difference in costs between groups was not significant.

Conclusions/limitations

• There was no evidence that lamotrigine led to clinical improvements in BPD symptomatology, social functioning, health-related quality of life, or substance use.
• Lamotrigine is neither clinically effective nor a cost-effective use of resources in the treatment of BPD.
• Limitations include a low level of adherence.

4. Domes G, Ower N, von Dawans B, et al. Effects of intranasal oxytocin administration on empathy and approach motivation in women with borderline personality disorder: a randomized controlled trial. Transl Psychiatry. 2019;9(1):328. doi: 10.1038/s41398-019-0658-4

A core feature of BPD is impairment in empathy; adequate empathy is required for intact social functioning. Oxytocin is a neuropeptide that helps regulate complex social cognition and behavior. Prior research has found that oxytocin administration enhances emotion regulation and empathy. Women with BPD have been observed to have lower levels of oxytocin. Domes et al⁶ conducted an RCT to see if oxytocin could have a beneficial effect on social approach and social cognition in women with BPD.

Study design

• In a double-blind, placebo-controlled, between-subject trial, 61 women who met DSM-IV criteria for BPD and 68 matched healthy controls were randomized to receive intranasal oxytocin, 24 IU, or placebo 45 minutes before completing an empathy task.
• An extended version of the Multifaceted Empathy Test was used to assess empathy and approach motivation.

Outcomes

• For cognitive empathy, patients with BPD exhibited significantly lower overall performance compared to controls. There was no effect of oxytocin on this performance in either group.
• Patients with BPD had significantly lower affective empathy compared with controls. After oxytocin administration, patients with BPD had significantly higher affective empathy than those with BPD who received placebo, reaching the level of healthy controls who received placebo.
• For positive stimuli, patients with BPD showed lower affective empathy than controls. Oxytocin treatment increased affective empathy in both groups.
• For negative stimuli, oxytocin increased affective empathy more in patients with BPD than in controls.
• Patients with BPD demonstrated less approach motivation than controls. Oxytocin increased approach motivation more in patients with BPD than in controls. For approach motivation toward positive stimuli, oxytocin had a significant effect on patients with BPD.

Continue to: Conclusions/limitations...

Conclusions/limitations

• Patients with BPD showed reduced cognitive and affective empathy and less approach behavior motivation than healthy controls.
• Patients with BPD who received oxytocin attained a level of affective empathy and approach motivation similar to that of healthy controls who received placebo. For positive stimuli, both groups exhibited comparable improvements from oxytocin. For negative stimuli, patients with BPD patients showed significant improvement with oxytocin, whereas healthy controls received no such benefit.
• Limitations include the use of self-report scales, lack of a control group, and inclusion of patients using psychotherapeutic medications. The study lacks generalizability because only women were included; the effect of exogenous oxytocin on men may differ.

5. Bozzatello P, Rocca P, Uscinska M, et al. Efficacy and tolerability of asenapine compared with olanzapine in borderline personality disorder: an open-label randomized controlled trial. CNS Drugs. 2017;31(9):809-819. doi: 10.1007/s40263-017-0458-4

The last decade has seen a noticeable shift in clinical practice from the use of antidepressants to mood stabilizers and second-generation antipsychotics (SGAs) in the treatment of BPD. Studies have demonstrated therapeutic effects of antipsychotic drugs across a wide range of BPD symptoms. Among SGAs, olanzapine is the most extensively studied across case reports, open-label studies, and RCTs of patients with BPD. In an RCT, Bozzatello et al⁷ compared the efficacy and tolerability of asenapine to olanzapine.

Study design

• In this open-label RCT, adults who met DSM-5 criteria for BPD were assigned to receive asenapine (n = 25) or olanzapine (n = 26) for 12 weeks.
• Study measurements included the Clinical Global Impression Scale, Severity item, HAM-D, HAM-A, Social and Occupational Functioning Assessment Scale, Borderline Personality Disorder Severity Index (BPDSI), BIS-11, Modified Overt Aggression Scale, and Dosage Record Treatment Emergent Symptom Scale.

Outcomes

• Asenapine and olanzapine had similar effects on BPD-related psychopathology, anxiety, and social and occupational functioning.
• Neither medication significantly decreased depressive or aggressive symptoms.
• Asenapine was superior to olanzapine in reducing the affective instability score of the BPDSI.
• Akathisia and restlessness/anxiety were more common with asenapine, and somnolence and fatigue were more common with olanzapine.

Conclusions/limitations

• The overall efficacy of asenapine was not different from olanzapine, and both medications were well-tolerated.
• Neither medication led to an improvement in depression or aggression, but asenapine was superior to olanzapine in reducing the severity of affective instability.
• Limitations include an open-label design, lack of placebo group, small sample size, high drop-out rate, exclusion of participants with co-occurring MDD and substance abuse/dependence, lack of data on prior pharmacotherapies and psychotherapies, and lack of power to detect a difference on the dissociation/paranoid ideation item of BPDSI.

6. Kulkarni J, Thomas N, Hudaib AR, et al. Effect of the glutamate NMDA receptor antagonist memantine as adjunctive treatment in borderline personality disorder: an exploratory, randomised, double-blind, placebo-controlled trial. CNS Drugs. 2018;32(2):179-187. doi: 10.1007/s40263-018-0506-8

It has been hypothesized that glutamate dysregulation and excitotoxicity are crucial to the development of the cognitive disturbances that underlie BPD. As such, glutamate modulators such as memantine hold promise for the treatment of BPD. In this RCT, Kulkarni et al⁸ examined the efficacy and tolerability of memantine compared with treatment as usual in patients with BPD.

Continue to: Study design...

• In an 8-week, double-blind, placebo-controlled trial, adults diagnosed with BPD according to the Diagnostic Interview for Borderline Patients were randomized to receive memantine (n = 17) or placebo (n = 16) in addition to treatment as usual. Treatment as usual included the use of antidepressants, mood stabilizers, and antipsychotics as well as psychotherapy and other psychosocial interventions.
• Patients were initiated on placebo or memantine, 10 mg/d. Memantine was increased to 20 mg/d after 7 days.
• ZAN-BPD score was the primary outcome and was measured at baseline and 2, 4, 6, and 8 weeks. An adverse effects questionnaire was administered every 2 weeks to assess tolerability.

Outcomes

• During the first 2 weeks of treatment, there were no significant improvements in ZAN-BPD score in the memantine group compared with the placebo group.
• Beginning with Week 2, compared with the placebo group, the memantine group experienced a significant reduction in total symptoms as measured by ZAN-BPD.
• There were no statistically significant differences in adverse events between groups.

Conclusions/limitations

• Memantine appears to be a well-tolerated treatment option for patients with BPD and merits further study.
• Limitations include a small sample size, and an inability to reach plateau of ZAN-BPD total score in either group. Also, there is considerable individual variability in memantine steady-state plasma concentrations, but plasma levels were not measured in this study.

Bottom Line

Findings from small randomized controlled trials suggest that transcranial direct current stimulation, oxytocin, asenapine, olanzapine, and memantine may have beneficial effects on some core symptoms of borderline personality disorder. These findings need to be replicated in larger studies.

For centuries, animals, especially dogs, have assisted humans in a variety of ways in their daily lives. Animals that assist people with disabilities fall into 2 broad categories: disability service animals, and emotional support animals (ESAs). Often there is confusion in how these categories differ because of the animal’s role and the laws related to them.

This article describes the differences between disability service animals and ESAs, and outlines the forensic and ethical concerns you should consider before agreeing to write a letter for a patient outlining their need for a disability service animal or ESA. A letter may protect a patient and their service animal or ESA in situations where laws and regulations typically prohibit animals, such as on a flight or when renting an apartment or house. Note that a description of how to conduct the formal patient evaluation before writing a verification letter is beyond the scope of this article.

The differences between disability service animals and ESAs

Purpose and training. Disability service animals, or service animals, are dogs of any breed (and in some cases miniature horses) that are specially trained to perform tasks for an individual with a disability (physical, sensory, psychiatric, intellectual, or other mental disability).^1-3 These tasks must be directly related to the individual’s disability.^1,2 On the other hand, ESAs, which can be any species of animal, provide support and minimize the impact of an individual’s emotional or psychological disability based on their presence alone. Unlike disability service animals, ESAs are not trained to perform a specific task or duty.^2,3

There is no legal requirement for service animals to know specific commands, and professional training is not required—individuals can train the animals themselves.¹ Service animals, mainly dogs, can be trained to perform numerous tasks, including⁴:

• attending to an individual’s mobility and activities of daily living
• guiding an individual who is deaf or hearing impaired
• helping to remind an individual to take their medications
• assisting an individual during and/or after a seizure
• alerting individuals with diabetes in advance of low or high blood sugar episodes
• supporting an individual with autism
• assisting an individual with a psychiatric or mental disability
• applying sensory commands such as lying on the person or resting their head on the individual’s lap to help the individual regain behavioral control.

Service dog verification works via an honor system, which can be problematic, especially in the case of psychiatric service dogs, whose handlers may not have a visible disability (Box 1^1,5).
 

Box 1

Legal protections. Under the Americans with Disabilities Act (ADA), individuals with disabilities can bring their service animals into buildings or facilities where members of the public, program participants, clients, customers, patrons, or invitees are allowed.² This does not include private clubs, religious organizations, or places of worship that are not open to the public.^6,7 ESAs do not qualify as service animals under the ADA and are not given the same legal accommodations as service animals.^1,3 Although ESAs were initially covered by the Air Carrier Access Act, they are no longer allowed in aircraft cabins after the US Department of Transportation revised this Act’s regulations in December 2020. ESAs are covered under the Fair Housing Act. Box 2^1-3,6-15 further discusses these laws and protections.

Evidence. In 1998, Dogs for Good (formerly Dogs for the Disabled), an organization based in the United Kingdom, conducted a survey that assessed the satisfaction of owners who were provided with trained assistance dogs.¹⁶ The results suggested that service dogs improved their owners’ mobility and helped ease the completion of tasks, thereby helping their owners integrate further on a society level and gain a strong bond with their animal.¹⁶ Another survey compared quality of life scores of individuals who owned a service dog vs individuals who were eligible to receive a dog, but did not yet have one.¹⁷ It found that service animals were able to help their owners gain a greater degree of freedom and enhance their ability to participate in everyday outings or tasks that may otherwise have been a struggle, or impossible, if the owner were alone.¹⁷ In addition to boosting confidence, self-esteem, and improving social integration, service dogs have been shown to improve their owners’ quality of life.¹⁷

Due to the difficulty in reconciling inconsistent definitions for ESAs, there is limited high-quality data pertaining to the potential benefits and risks of ESAs.⁹ Currently, ESAs are not an evidence-based treatment for psychiatric disorders. To date, a handful of small studies have focused on ESAs. However, data from actual tests of the clinical risks and benefits of ESAs do not exist.⁹ In practice, ESAs are equivalent to pets. It stands to reason that similar to pets, ESAs could reduce loneliness, improve life satisfaction, and provide a sense of well-being.⁹ A systematic review suggested that pets provide benefits to patients with mental health conditions “through the intensity of connectivity with their owners and the contribution they make to emotional support in times of crises together with their ability to help manage symptoms when they arise.”¹⁸ In response to a congressional mandate, the US Department of Veterans Affairs launched a multi-site study from December 2014 to June 2019 to examine how limitations on activity and quality of life in veterans with posttraumatic stress disorder are impacted by the provision of a service dog vs an emotional support dog.¹⁹ As of October 14, 2021, results had not been published.¹⁹

Continue to: What’s in a disability service animal/ESA letter?

What’s in a disability service animal/ESA letter?

If you decide to write a letter advocating for your patient to have a service animal or ESA, the letter should appear on letterhead, be written by a licensed mental health professional, and include the following^2,20:

• statement that the letter is being written at the patient’s request and is being given directly to the patient for use as the patient sees fit
• confirmation of the patient’s DSM-5 mental health diagnosis
• explanation of how the animal helps alleviate symptoms of the patient’s condition, briefly describing any interaction(s) between the animal and patient that you may have observed, and if applicable, a mention of any training the animal may have received from a qualified trainer if applicable
• explanation of the possible negative effects of the patient not having the animal with him or her
• statement that you are not vouching for the animal’s behavior
• verification of your involvement in your patient’s treatment and your assessment of the patient as their licensed mental health professional (including details such as date and type of license you have and the state/other jurisdiction where it was issued).

In a letter for a service animal, also indicate that your patient is psychiatrically disabled to the extent that your patient is not able to perform at least one major life task without the daily assistance of a service animal.²Should you write your patient a letter?

Writing a letter advocating for a patient to have a service animal or ESA may appear innocuous, but doing so may have serious ramifications. Writing a letter certifying a dog as a service animal does not make that animal a service animal; the dog must be specifically trained for a task or tasks directly related to that individual’s disability. There are no current standards for conducting evaluations to determine the need a patient has for a service animal or ESA. How to conduct such evaluations is beyond the scope of this article. There are meager opportunities for formal education and training on how to conduct these evaluations.⁹ Online resources may be incomplete or inaccurate, and this information is often produced by lay animal enthusiasts and organizations, which can lead to a biased depiction of these animals.⁹

If you decide to write a letter for your patient, consider the following forensic and ethical concerns.

Remain objective. As an advocate for your patient, you may find it difficult to remain neutral and objective when asked to determine if your patient has a disability, the severity of the disability, the impact of the disability on your patient’s life, and the need for a service animal or ESA. Ensure that your advocacy for your patient does not impair your objectivity; if that is difficult, consider referring your patient to a third party who can conduct an objective evaluation.

Understand the risks. If you make written recommendations for special accommodations in a letter and those recommendations are disputed by an agency, that agency could initiate legal action and you may be called to justify your recommendations in a deposition or open court.^9,21 Before writing the letter, ask yourself, “Can I defend my determination that my patient is disabled by a DSM-5 disorder and that this disability requires the presence of an animal in exception to existing policy?”²¹ Be prepared to state in a legal proceeding that the presence of a service animal or ESA is necessary. If you are unwilling to risk exposure to a legal action, then you should likely refrain from writing the letter. It is a crime to fraudulently certify an animal as a service animal in some jurisdictions, and such conduct could result in disciplinary action by your licensing board.²¹

Conduct a systematic examination. When you write a letter for your patient, you are explicitly declaring your patient has a disability or condition. Comprehensive disability determinations are complex and are best conducted by assessing for objective evidence of psychiatric disorders and impairment through the use of standard, systematic examination methods.²² Unstandardized measures (eg, asking patients open-ended questions and then relying on your clinical judgement and interpretation in arriving at conclusions) are not as effective.²² In addition, consider the possibility that your patient may malinger their symptoms in an effort to obtain a letter supporting a service animal or ESA. Assessing for malingering is essential to making a disability determination, especially if a disability claim is based primarily on self-report.²²

Anticipate pushback. Problems can arise when a patient wants a letter that you cannot or will not provide due to your scope of practice. Consider how you would resolve the situation when you do not believe your patient has a disability that requires the presence of a service animal or ESA—or you believe that your patient no longer needs a service animal or ESA—and the patient disagrees.²¹ Disagreeing with your patient’s assessment could result in a conflict of interest that could damage the therapeutic relationship.²¹

Box 2

Bottom Line 

Disability service animals and emotional support animals (ESAs) differ in their roles and legal protections. Before writing a letter in support of a patient’s request for a service animal or ESA, take into account the forensic and ethical implications of doing so.

Related Resources

• US Department of Justice. Civil Rights Division. Disability Rights Section. ADA requirements. Service animals. Updated February 24, 2020. https://www.ada.gov/service_ animals_2010.htm

• American Veterinary Medical Association. Service, emotional support and therapy animals. https://www. avma.org/resources-tools/animal-health-welfare/ service-emotional-support-and-therapy-animals

• US Department of Transportation. US Department of Transportation announces final rule on traveling by air with service animals. https://www.transportation.gov/briefingroom/us-department-transportation-announces-finalrule-traveling-air-service-animals

For centuries, animals, especially dogs, have assisted humans in a variety of ways in their daily lives. Animals that assist people with disabilities fall into 2 broad categories: disability service animals, and emotional support animals (ESAs). Often there is confusion in how these categories differ because of the animal’s role and the laws related to them.

This article describes the differences between disability service animals and ESAs, and outlines the forensic and ethical concerns you should consider before agreeing to write a letter for a patient outlining their need for a disability service animal or ESA. A letter may protect a patient and their service animal or ESA in situations where laws and regulations typically prohibit animals, such as on a flight or when renting an apartment or house. Note that a description of how to conduct the formal patient evaluation before writing a verification letter is beyond the scope of this article.

The differences between disability service animals and ESAs

Purpose and training. Disability service animals, or service animals, are dogs of any breed (and in some cases miniature horses) that are specially trained to perform tasks for an individual with a disability (physical, sensory, psychiatric, intellectual, or other mental disability).^1-3 These tasks must be directly related to the individual’s disability.^1,2 On the other hand, ESAs, which can be any species of animal, provide support and minimize the impact of an individual’s emotional or psychological disability based on their presence alone. Unlike disability service animals, ESAs are not trained to perform a specific task or duty.^2,3

There is no legal requirement for service animals to know specific commands, and professional training is not required—individuals can train the animals themselves.¹ Service animals, mainly dogs, can be trained to perform numerous tasks, including⁴:

• attending to an individual’s mobility and activities of daily living
• guiding an individual who is deaf or hearing impaired
• helping to remind an individual to take their medications
• assisting an individual during and/or after a seizure
• alerting individuals with diabetes in advance of low or high blood sugar episodes
• supporting an individual with autism
• assisting an individual with a psychiatric or mental disability
• applying sensory commands such as lying on the person or resting their head on the individual’s lap to help the individual regain behavioral control.

Service dog verification works via an honor system, which can be problematic, especially in the case of psychiatric service dogs, whose handlers may not have a visible disability (Box 1^1,5).
 

Box 1

Legal protections. Under the Americans with Disabilities Act (ADA), individuals with disabilities can bring their service animals into buildings or facilities where members of the public, program participants, clients, customers, patrons, or invitees are allowed.² This does not include private clubs, religious organizations, or places of worship that are not open to the public.^6,7 ESAs do not qualify as service animals under the ADA and are not given the same legal accommodations as service animals.^1,3 Although ESAs were initially covered by the Air Carrier Access Act, they are no longer allowed in aircraft cabins after the US Department of Transportation revised this Act’s regulations in December 2020. ESAs are covered under the Fair Housing Act. Box 2^1-3,6-15 further discusses these laws and protections.

Evidence. In 1998, Dogs for Good (formerly Dogs for the Disabled), an organization based in the United Kingdom, conducted a survey that assessed the satisfaction of owners who were provided with trained assistance dogs.¹⁶ The results suggested that service dogs improved their owners’ mobility and helped ease the completion of tasks, thereby helping their owners integrate further on a society level and gain a strong bond with their animal.¹⁶ Another survey compared quality of life scores of individuals who owned a service dog vs individuals who were eligible to receive a dog, but did not yet have one.¹⁷ It found that service animals were able to help their owners gain a greater degree of freedom and enhance their ability to participate in everyday outings or tasks that may otherwise have been a struggle, or impossible, if the owner were alone.¹⁷ In addition to boosting confidence, self-esteem, and improving social integration, service dogs have been shown to improve their owners’ quality of life.¹⁷

Due to the difficulty in reconciling inconsistent definitions for ESAs, there is limited high-quality data pertaining to the potential benefits and risks of ESAs.⁹ Currently, ESAs are not an evidence-based treatment for psychiatric disorders. To date, a handful of small studies have focused on ESAs. However, data from actual tests of the clinical risks and benefits of ESAs do not exist.⁹ In practice, ESAs are equivalent to pets. It stands to reason that similar to pets, ESAs could reduce loneliness, improve life satisfaction, and provide a sense of well-being.⁹ A systematic review suggested that pets provide benefits to patients with mental health conditions “through the intensity of connectivity with their owners and the contribution they make to emotional support in times of crises together with their ability to help manage symptoms when they arise.”¹⁸ In response to a congressional mandate, the US Department of Veterans Affairs launched a multi-site study from December 2014 to June 2019 to examine how limitations on activity and quality of life in veterans with posttraumatic stress disorder are impacted by the provision of a service dog vs an emotional support dog.¹⁹ As of October 14, 2021, results had not been published.¹⁹

Continue to: What’s in a disability service animal/ESA letter?

What’s in a disability service animal/ESA letter?

If you decide to write a letter advocating for your patient to have a service animal or ESA, the letter should appear on letterhead, be written by a licensed mental health professional, and include the following^2,20:

• statement that the letter is being written at the patient’s request and is being given directly to the patient for use as the patient sees fit
• confirmation of the patient’s DSM-5 mental health diagnosis
• explanation of how the animal helps alleviate symptoms of the patient’s condition, briefly describing any interaction(s) between the animal and patient that you may have observed, and if applicable, a mention of any training the animal may have received from a qualified trainer if applicable
• explanation of the possible negative effects of the patient not having the animal with him or her
• statement that you are not vouching for the animal’s behavior
• verification of your involvement in your patient’s treatment and your assessment of the patient as their licensed mental health professional (including details such as date and type of license you have and the state/other jurisdiction where it was issued).

In a letter for a service animal, also indicate that your patient is psychiatrically disabled to the extent that your patient is not able to perform at least one major life task without the daily assistance of a service animal.²Should you write your patient a letter?

Writing a letter advocating for a patient to have a service animal or ESA may appear innocuous, but doing so may have serious ramifications. Writing a letter certifying a dog as a service animal does not make that animal a service animal; the dog must be specifically trained for a task or tasks directly related to that individual’s disability. There are no current standards for conducting evaluations to determine the need a patient has for a service animal or ESA. How to conduct such evaluations is beyond the scope of this article. There are meager opportunities for formal education and training on how to conduct these evaluations.⁹ Online resources may be incomplete or inaccurate, and this information is often produced by lay animal enthusiasts and organizations, which can lead to a biased depiction of these animals.⁹

If you decide to write a letter for your patient, consider the following forensic and ethical concerns.

Remain objective. As an advocate for your patient, you may find it difficult to remain neutral and objective when asked to determine if your patient has a disability, the severity of the disability, the impact of the disability on your patient’s life, and the need for a service animal or ESA. Ensure that your advocacy for your patient does not impair your objectivity; if that is difficult, consider referring your patient to a third party who can conduct an objective evaluation.

Understand the risks. If you make written recommendations for special accommodations in a letter and those recommendations are disputed by an agency, that agency could initiate legal action and you may be called to justify your recommendations in a deposition or open court.^9,21 Before writing the letter, ask yourself, “Can I defend my determination that my patient is disabled by a DSM-5 disorder and that this disability requires the presence of an animal in exception to existing policy?”²¹ Be prepared to state in a legal proceeding that the presence of a service animal or ESA is necessary. If you are unwilling to risk exposure to a legal action, then you should likely refrain from writing the letter. It is a crime to fraudulently certify an animal as a service animal in some jurisdictions, and such conduct could result in disciplinary action by your licensing board.²¹

Conduct a systematic examination. When you write a letter for your patient, you are explicitly declaring your patient has a disability or condition. Comprehensive disability determinations are complex and are best conducted by assessing for objective evidence of psychiatric disorders and impairment through the use of standard, systematic examination methods.²² Unstandardized measures (eg, asking patients open-ended questions and then relying on your clinical judgement and interpretation in arriving at conclusions) are not as effective.²² In addition, consider the possibility that your patient may malinger their symptoms in an effort to obtain a letter supporting a service animal or ESA. Assessing for malingering is essential to making a disability determination, especially if a disability claim is based primarily on self-report.²²

Anticipate pushback. Problems can arise when a patient wants a letter that you cannot or will not provide due to your scope of practice. Consider how you would resolve the situation when you do not believe your patient has a disability that requires the presence of a service animal or ESA—or you believe that your patient no longer needs a service animal or ESA—and the patient disagrees.²¹ Disagreeing with your patient’s assessment could result in a conflict of interest that could damage the therapeutic relationship.²¹

Box 2

Bottom Line 

Disability service animals and emotional support animals (ESAs) differ in their roles and legal protections. Before writing a letter in support of a patient’s request for a service animal or ESA, take into account the forensic and ethical implications of doing so.

Related Resources

• US Department of Justice. Civil Rights Division. Disability Rights Section. ADA requirements. Service animals. Updated February 24, 2020. https://www.ada.gov/service_ animals_2010.htm

• American Veterinary Medical Association. Service, emotional support and therapy animals. https://www. avma.org/resources-tools/animal-health-welfare/ service-emotional-support-and-therapy-animals

• US Department of Transportation. US Department of Transportation announces final rule on traveling by air with service animals. https://www.transportation.gov/briefingroom/us-department-transportation-announces-finalrule-traveling-air-service-animals

For centuries, animals, especially dogs, have assisted humans in a variety of ways in their daily lives. Animals that assist people with disabilities fall into 2 broad categories: disability service animals, and emotional support animals (ESAs). Often there is confusion in how these categories differ because of the animal’s role and the laws related to them.

This article describes the differences between disability service animals and ESAs, and outlines the forensic and ethical concerns you should consider before agreeing to write a letter for a patient outlining their need for a disability service animal or ESA. A letter may protect a patient and their service animal or ESA in situations where laws and regulations typically prohibit animals, such as on a flight or when renting an apartment or house. Note that a description of how to conduct the formal patient evaluation before writing a verification letter is beyond the scope of this article.

The differences between disability service animals and ESAs

Purpose and training. Disability service animals, or service animals, are dogs of any breed (and in some cases miniature horses) that are specially trained to perform tasks for an individual with a disability (physical, sensory, psychiatric, intellectual, or other mental disability).^1-3 These tasks must be directly related to the individual’s disability.^1,2 On the other hand, ESAs, which can be any species of animal, provide support and minimize the impact of an individual’s emotional or psychological disability based on their presence alone. Unlike disability service animals, ESAs are not trained to perform a specific task or duty.^2,3

There is no legal requirement for service animals to know specific commands, and professional training is not required—individuals can train the animals themselves.¹ Service animals, mainly dogs, can be trained to perform numerous tasks, including⁴:

• attending to an individual’s mobility and activities of daily living
• guiding an individual who is deaf or hearing impaired
• helping to remind an individual to take their medications
• assisting an individual during and/or after a seizure
• alerting individuals with diabetes in advance of low or high blood sugar episodes
• supporting an individual with autism
• assisting an individual with a psychiatric or mental disability
• applying sensory commands such as lying on the person or resting their head on the individual’s lap to help the individual regain behavioral control.

Service dog verification works via an honor system, which can be problematic, especially in the case of psychiatric service dogs, whose handlers may not have a visible disability (Box 1^1,5).
 

Box 1

Legal protections. Under the Americans with Disabilities Act (ADA), individuals with disabilities can bring their service animals into buildings or facilities where members of the public, program participants, clients, customers, patrons, or invitees are allowed.² This does not include private clubs, religious organizations, or places of worship that are not open to the public.^6,7 ESAs do not qualify as service animals under the ADA and are not given the same legal accommodations as service animals.^1,3 Although ESAs were initially covered by the Air Carrier Access Act, they are no longer allowed in aircraft cabins after the US Department of Transportation revised this Act’s regulations in December 2020. ESAs are covered under the Fair Housing Act. Box 2^1-3,6-15 further discusses these laws and protections.

Evidence. In 1998, Dogs for Good (formerly Dogs for the Disabled), an organization based in the United Kingdom, conducted a survey that assessed the satisfaction of owners who were provided with trained assistance dogs.¹⁶ The results suggested that service dogs improved their owners’ mobility and helped ease the completion of tasks, thereby helping their owners integrate further on a society level and gain a strong bond with their animal.¹⁶ Another survey compared quality of life scores of individuals who owned a service dog vs individuals who were eligible to receive a dog, but did not yet have one.¹⁷ It found that service animals were able to help their owners gain a greater degree of freedom and enhance their ability to participate in everyday outings or tasks that may otherwise have been a struggle, or impossible, if the owner were alone.¹⁷ In addition to boosting confidence, self-esteem, and improving social integration, service dogs have been shown to improve their owners’ quality of life.¹⁷

Due to the difficulty in reconciling inconsistent definitions for ESAs, there is limited high-quality data pertaining to the potential benefits and risks of ESAs.⁹ Currently, ESAs are not an evidence-based treatment for psychiatric disorders. To date, a handful of small studies have focused on ESAs. However, data from actual tests of the clinical risks and benefits of ESAs do not exist.⁹ In practice, ESAs are equivalent to pets. It stands to reason that similar to pets, ESAs could reduce loneliness, improve life satisfaction, and provide a sense of well-being.⁹ A systematic review suggested that pets provide benefits to patients with mental health conditions “through the intensity of connectivity with their owners and the contribution they make to emotional support in times of crises together with their ability to help manage symptoms when they arise.”¹⁸ In response to a congressional mandate, the US Department of Veterans Affairs launched a multi-site study from December 2014 to June 2019 to examine how limitations on activity and quality of life in veterans with posttraumatic stress disorder are impacted by the provision of a service dog vs an emotional support dog.¹⁹ As of October 14, 2021, results had not been published.¹⁹

Continue to: What’s in a disability service animal/ESA letter?

What’s in a disability service animal/ESA letter?

If you decide to write a letter advocating for your patient to have a service animal or ESA, the letter should appear on letterhead, be written by a licensed mental health professional, and include the following^2,20:

• statement that the letter is being written at the patient’s request and is being given directly to the patient for use as the patient sees fit
• confirmation of the patient’s DSM-5 mental health diagnosis
• explanation of how the animal helps alleviate symptoms of the patient’s condition, briefly describing any interaction(s) between the animal and patient that you may have observed, and if applicable, a mention of any training the animal may have received from a qualified trainer if applicable
• explanation of the possible negative effects of the patient not having the animal with him or her
• statement that you are not vouching for the animal’s behavior
• verification of your involvement in your patient’s treatment and your assessment of the patient as their licensed mental health professional (including details such as date and type of license you have and the state/other jurisdiction where it was issued).

In a letter for a service animal, also indicate that your patient is psychiatrically disabled to the extent that your patient is not able to perform at least one major life task without the daily assistance of a service animal.²Should you write your patient a letter?

Writing a letter advocating for a patient to have a service animal or ESA may appear innocuous, but doing so may have serious ramifications. Writing a letter certifying a dog as a service animal does not make that animal a service animal; the dog must be specifically trained for a task or tasks directly related to that individual’s disability. There are no current standards for conducting evaluations to determine the need a patient has for a service animal or ESA. How to conduct such evaluations is beyond the scope of this article. There are meager opportunities for formal education and training on how to conduct these evaluations.⁹ Online resources may be incomplete or inaccurate, and this information is often produced by lay animal enthusiasts and organizations, which can lead to a biased depiction of these animals.⁹

If you decide to write a letter for your patient, consider the following forensic and ethical concerns.

Remain objective. As an advocate for your patient, you may find it difficult to remain neutral and objective when asked to determine if your patient has a disability, the severity of the disability, the impact of the disability on your patient’s life, and the need for a service animal or ESA. Ensure that your advocacy for your patient does not impair your objectivity; if that is difficult, consider referring your patient to a third party who can conduct an objective evaluation.

Understand the risks. If you make written recommendations for special accommodations in a letter and those recommendations are disputed by an agency, that agency could initiate legal action and you may be called to justify your recommendations in a deposition or open court.^9,21 Before writing the letter, ask yourself, “Can I defend my determination that my patient is disabled by a DSM-5 disorder and that this disability requires the presence of an animal in exception to existing policy?”²¹ Be prepared to state in a legal proceeding that the presence of a service animal or ESA is necessary. If you are unwilling to risk exposure to a legal action, then you should likely refrain from writing the letter. It is a crime to fraudulently certify an animal as a service animal in some jurisdictions, and such conduct could result in disciplinary action by your licensing board.²¹

Conduct a systematic examination. When you write a letter for your patient, you are explicitly declaring your patient has a disability or condition. Comprehensive disability determinations are complex and are best conducted by assessing for objective evidence of psychiatric disorders and impairment through the use of standard, systematic examination methods.²² Unstandardized measures (eg, asking patients open-ended questions and then relying on your clinical judgement and interpretation in arriving at conclusions) are not as effective.²² In addition, consider the possibility that your patient may malinger their symptoms in an effort to obtain a letter supporting a service animal or ESA. Assessing for malingering is essential to making a disability determination, especially if a disability claim is based primarily on self-report.²²

Anticipate pushback. Problems can arise when a patient wants a letter that you cannot or will not provide due to your scope of practice. Consider how you would resolve the situation when you do not believe your patient has a disability that requires the presence of a service animal or ESA—or you believe that your patient no longer needs a service animal or ESA—and the patient disagrees.²¹ Disagreeing with your patient’s assessment could result in a conflict of interest that could damage the therapeutic relationship.²¹

Box 2

Bottom Line 

Disability service animals and emotional support animals (ESAs) differ in their roles and legal protections. Before writing a letter in support of a patient’s request for a service animal or ESA, take into account the forensic and ethical implications of doing so.

Related Resources

• US Department of Justice. Civil Rights Division. Disability Rights Section. ADA requirements. Service animals. Updated February 24, 2020. https://www.ada.gov/service_ animals_2010.htm

• American Veterinary Medical Association. Service, emotional support and therapy animals. https://www. avma.org/resources-tools/animal-health-welfare/ service-emotional-support-and-therapy-animals

• US Department of Transportation. US Department of Transportation announces final rule on traveling by air with service animals. https://www.transportation.gov/briefingroom/us-department-transportation-announces-finalrule-traveling-air-service-animals

Neuroscience research over the past half century has failed to significantly advance the treatment of severe mental illness.^1,2 Hence, evidence that a longer duration of untreated psychosis (DUP) aggravates—and early intervention with medication and social supports ameliorates—the long-term adverse consequences of psychotic disorders generated a great deal of interest.^3,4 This knowledge led to the development of diverse early intervention services worldwide aimed at this putative “critical window.” It raised the possibility that appropriate interventions could prevent the long-term disability that makes chronic psychosis one of the most debilitating disorders.^5,6 However, even beyond the varied cultural and economic confounds, it is difficult to assess, compare, and optimize program effectiveness.⁷ Obstacles include paucity of sufficiently powered, well-designed randomized controlled trials (RCTs), the absence of diagnostic biomarkers or other prognostic indicators to better account for the inherent heterogeneity in the population and associated outcomes, and the absence of modifiable risk factors that can guide interventions and provide intermediate outcomes.^4,8-10

To better appreciate these issues, it is important to distinguish whether a program is designed to prevent psychosis, or to mitigate the effects of psychosis. Two models include the:

• Prevention model, which focuses on young individuals who are not yet overtly psychotic but at high risk
• First-episode recovery model, which focuses on those who have experienced a first episode of psychosis (FEP) but have not yet developed a chronic disorder.

Both models share long-term goals and are hampered by many of the same issues summarized above. They both deviate markedly from the standard medical model by including psychosocial services designed to promote restoration of a self-defined trajectory to greater independence.^11-14 The 2 differ, however, in the challenges they must overcome to produce their sample populations and establish effective interventions.^10,15,16

In this article, we provide a succinct overview of these issues and a set of recommendations based on a “strength-based” approach. This approach focuses on finding common ground between patients, their support system, and the treatment team in the service of empowering patients to resume responsibility for transition to adulthood.

The prevention model

While most prevention initiatives in medicine rely on the growing ability to target specific pathophysiologic pathways,³ preventing psychosis relies on clinical evidence showing that DUP and early interventions predict a better course of severe mental illness.¹⁷ In contrast, initiatives such as normalizing neonatal neuronal pathways are more consistent with the strategy utilized in other fields but have yet to yield a pathophysiologic target for psychosis.^3,18

Initial efforts to identify ‘at-risk’ individuals

The prevention model of psychosis is based on the ability to identify young individuals at high risk for developing a psychotic disorder (Figure). The first screening measures were focused on prodromal psychosis (eg, significant loss of function, family history, and “intermittent” and “attenuated” psychotic symptoms). When applied to referred (ie, pre-screened) samples, 30% to 40% of this group who met criteria transitioned to psychosis over the next 1 to 3 years despite antidepressant and psychosocial interventions.¹⁹ Comprising 8 academic medical centers, the North American Prodrome Longitudinal Study (NAPLS) produced similar results using the Structured Interview for Prodromal Syndromes (SIPS).¹⁷ Thus, 30% to 50% of pre-screened individuals referred by school counselors and mental health professionals met SIPS criteria, and 35% of these individuals transitioned to psychosis over 30 months. The validity of this measure was further supported by the fact that higher baseline levels of unusual thought content, suspicion/paranoia, social impairment, and substance abuse successfully distinguished approximately 80% of those who transitioned to psychosis. The results of this first generation of screening studies were exciting because they seemed to demonstrate that highly concentrated samples of young persons at high risk of developing psychosis could be identified, and that fine-tuning the screening criteria could produce even more enriched samples (ie, positive predictive power).

Initial interventions produced promising results

The development of effective screening measures led to reports of effective treatment interventions. These were largely applied in a clinical staging model that restricted antipsychotic medications to those who failed to improve after receiving potentially “less toxic” interventions (eg, omega-3 polyunsaturated fatty acids and other antioxidants; psychotherapy; cognitive-behavioral therapy [CBT]; family therapy).⁵ While study designs were typically quasi-experimental, the interventions appeared to dramatically diminish the transition to psychosis (ie, approximately 50%).

Continue to: The first generation...

The first generation of RCTs appeared to confirm these results, although sample sizes were small, and most study designs assessed only a single intervention. Initial meta-analyses of these data reported that both CBT and antipsychotics appeared to prevent approximately one-half of individuals from becoming psychotic at 12 months, and more than one-third at 2 to 4 years, compared with treatment as usual.²⁰

While some researchers challenged the validity of these findings,^21-23 the results generated tremendous international enthusiasm and calls for widespread implementation.⁶ The number of early intervention services (EIS) centers increased dramatically worldwide, and in 2014 the National Institute for Health and Care Excellence released standards for interventions to prevent transition to psychosis.²⁴ These included close monitoring, CBT and family interventions, and avoiding antipsychotics when possible.²⁴

Focusing on sensitivity over specificity

The first generation of studies generated by the prevention model relied on outreach programs or referrals, which produced small samples of carefully selected, pre-screened individuals (Figure, Pre-screened) who were then screened again to establish the high-risk sample.²⁵ While approximately 33% of these individuals became psychotic, the screening process required a very efficient means of eliminating those not at high-risk (given the ultimate target population represented only approximately .5% of young people) (Figure). The pre-screening and screening processes in these first-generation studies were labor-intensive but could only identify approximately 5% of those individuals destined to become psychotic over the next 2 or 3 years. Thus, alternative methods to enhance sensitivity were needed to extend programming to the general population.

Second-generation pre-screening (Figure; Step 1). New pre-screening methods were identified that captured more individuals destined to become psychotic. For example, approximately 90% of this population were registered in health care organizations (eg, health maintenance organizations) and received a psychiatric diagnosis in the year prior to the onset of psychosis (true positives).⁸ These samples, however, contained a much higher percentage of persons not destined to become psychotic, and somehow the issue of specificity (decreasing false positives) was minimized.^8,9 For example, pre-screened samples contained 20 to 50 individuals not destined to become psychotic for each one who did.²⁶ Since screening measures could only eliminate approximately 20% of this group (Figure, Step 2, page 25), second-generation transition rates fell from 30% to 40% to 2% to 10%.^27,28

Other pre-screening approaches were introduced, but they also focused on capturing more of those destined to become psychotic (sensitivity) than eliminating those who would not (specificity). For instance, Australia opened more than 100 “Headspace” community centers nationwide designed to promote engagement and self-esteem in youth experiencing anxiety; depression; stress; relationship, work, or school problems; or bullying.¹³ Most services were free and included mental health staff who screened for psychosis and provided a wide range of services in a destigmatized setting. These methods identified at least an additional 5% to 7% of individuals destined to become psychotic, but to our knowledge, no data have been published on whether they helped eliminate those who did not.

Continue to: Second-generation screening

Second-generation screening (Figure, Step 2). A second screening aims to retain those pre-screened individuals who will become psychotic (ie, minimizing false negatives) while further minimizing those who do not (ie, minimizing false positives). The addition of cognitive, neural (eg, structural MRI; neurophysiologic), and biochemical (eg, inflammatory immune and stress) markers to the risk calculators have produced a sensitivity close to 100%.^8,9 Unfortunately, these studies downplayed specificity, which remained approximately 20%.^8,9 Specificity is critical not just because of concerns about stigma (ie, labeling people as pre-psychotic when they are not) but also because of the adverse effects of antipsychotic medications and the effects on future program development (interventions are costly and labor-intensive). Also, diluting the pool with individuals not at risk makes it nearly impossible to identify effective interventions (ie, power).^27,28

While some studies focused on increasing specificity (to approximately 75%), this leads to an unacceptable loss of sensitivity (from 90% to 60%),²⁹ with 40% of pre-screened individuals who would become psychotic being eliminated from the study population. The addition of other biological markers (eg, salivary cortisol)³⁰ and use of learning health systems may be able to enhance these numbers (initial reports of specificity = 87% and sensitivity = 85%).^8,9 This is accomplished by integrating artificial and human intelligence measures of clinical (symptom and neurocognitive measures) and biological (eg, polygenetic risk scores; gray matter volume) variables.³¹ However, even if these results are replicated, more effective pre-screening measures will be required.

Identifying a suitable sample population for prevention program studies is clearly more complicated than for FEP studies, where one can usually identify many of those in the at-risk population by their first hospitalization for psychotic symptoms. The issues of false positives (eg, substance-induced psychosis) and negatives (eg, slow deterioration, prominent negative symptoms) are important concerns, but proportionately far less significant.

Prevention and FEP interventions

Once a study sample is constituted, 1 to 3 years of treatment interventions are initiated. Interventions for prevention programs typically include CBT directed at attenuated psychosis (eg, reframing or de-catastrophizing unusual thoughts and minimizing distress associated with unusual perceptions); case management to facilitate personal, educational, and vocational goals; and family therapy in single or multi-group formats to educate one’s support system about the risk state and to minimize adverse familial responses.¹⁴ Many programs also include supported education or employment services to promote reintegration in age-appropriate activities; group therapy focused on substance abuse and social skills training; cognitive remediation to ameliorate the cognitive dysfunction; and an array of pharmacologic interventions designed to delay or prevent transition to psychosis or to alleviate symptoms. While most interventions are similar, FEP programs have recently included peer support staff. This appears to instill hope in newly diagnosed patients, provide role models, and provide peer supporters an opportunity to use their experiences to help others and earn income.³²

The breadth and depth of these services are critical because retention in the program is highly dependent on participant engagement, which in turn is highly dependent on whether the program can help individuals get what they want (eg, friends, employment, education, more autonomy, physical health). The setting and atmosphere of the treatment program and the willingness/ability of staff to meet participants in the community are also important elements.^11,12 In this context, the Headspace community centers are having an impact far beyond Australia and may prove to be a particularly good model.¹³

Continue to: Assessing prevention and FEP interventions

The second generation of studies of prevention programs has not confirmed, let alone extended, the earlier findings and meta-analyses. A 2020 report concluded CBT was still the most promising intervention; it was more effective than control treatments at 12 and 18 months, although not at 6, 24, or 48 months.³³ This review included controlled, open-label, and naturalistic studies that assessed family therapy; omega-3 polyunsaturated fatty acids; integrated psychological therapy (a package of interventions that included family education, CBT, social skills training, and cognitive remediation); N-methyl-D-aspartate receptor modulators; mood stabilizers; and antipsychotics. In addition to the evidence supporting CBT, the results also indicated nonsignificant trends favoring family and integrated psychological therapy. Neither a 2019 Cochrane review³⁴ nor a 2020 “umbrella” assessment of 42 meta-analyses⁹ found convincing evidence for the efficacy of any program components.

While these disappointing findings are at least partly attributable to the methodological challenges described above and in the Figure, other factors may hinder establishing effective interventions. In contrast to FEP studies, those focused on prevention had a very ambitious agenda (eliminating psychosis) and tended to downplay more modest intermediate outcomes. These studies also tended to assess new ideas with small samples rather than pursue promising findings with larger multi-site studies focused on a group of interventions. The authors of a Cochrane review observed “There is the impression that in this whole area there is a triumph of hope over adversity. There is the repeated hope invested in another—often unique—study question and then a study of fewer than 100 participants are completed. This results in the set of comparisons reported here, all 9 of which are too underpowered to really highlight clear differences.”³⁴ To use a baseball analogy, it seems that investigators are “swinging for the fence” when a few singles are what’s really needed.

From the outset, the goals of FEP studies were more modest, largely ignoring the task of developing consensus definitions of recovery that require following patients for up to 5 to 10 years. Instead, they use intermediate endpoints based on adapting treatments that already appeared effective in patients with chronic mental disorders.³⁵ As a consequence, researchers examining FEP demonstrated clear, albeit limited, salutary effects using large multi-site trials and previously established outcome measures.^3,10,36 For instance, the Recovery After an Initial Schizophrenia Episode-Early Treatment Program (RAISE-ETP) study was a 2-year, multi-site RCT (N = 404) funded by the National Institute of Mental Health (NIMH). The investigators reported improved indices of social function (eg, quality of life; education and work participation) and total ratings of psychopathology and depression compared with treatment as usual. Furthermore, they established that DUP predicted treatment response.³⁵ The latter finding was underscored by improvement being limited to the 50% with <74 weeks DUP. Annual costs of the program per 1 standard deviation improvement in quality of life were approximately $1,000 for patients with <74 weeks DUP and $40,000 for those with >74 weeks DUP. Concurrent meta-analyses confirmed and extended these findings,¹⁶ showing higher remission rates; diminished relapses and hospital admissions; greater engagement in programming; greater involvement in work and school; improved quality of life; and other steps toward recovery. These studies were also able to establish a clear benefit of antipsychotic medications, particularly a high acceptance of long-acting injectable antipsychotic formulations, which promoted adherence and decreased some adverse events³⁷; and early use of clozapine therapy, which improved remission rates and longer-term outcomes.³⁸ Other findings underscored the need to anticipate and address new problems associated with effective antipsychotic therapy (eg, antipsychotic response correlates with weight gain, a particularly intolerable adverse event for this age group).³⁹ Providing pre-emptive strategies such as exercise groups and nutritional education may be necessary to maintain adherence.

Limitations of FEP studies

The effect sizes in these FEP studies were small to medium on outcome measures tracking recovery and associated indicators (eg, global functioning, school/work participation, treatment engagement); the number needed to treat for each of these was >10. There is no clear evidence that recovery programs such as RAISE-ETP actually reduce longer-term disability. Most studies showed disability payments increased while clinical benefits tended to fade over time. In addition, by grouping interventions together, the studies made it difficult to identify effective vs ineffective treatments, let alone determine how best to personalize therapy for participants in future studies.

The next generation of FEP studies

While limited in scope, the results of the recent FEP studies justify a next generation of recovery interventions designed to address these shortcomings and optimize program outcomes.³⁹ Most previous FEP studies were conducted in community mental health center settings, thus eliminating the need to transition services developed in academia into the “real world.” The next generation of NIMH studies will be primarily conducted in analogous settings under the Early Psychosis Intervention Network (EPINET).⁴⁰ EPINET’s study design echoes that responsible for the stepwise successes in the late 20th century that produced cures for the deadliest childhood cancer, acute lymphoblastic leukemia (ALL). This disease was successfully treated by modifying diverse evidence-based practices without relying on pharmacologic or other major treatment breakthroughs. Despite this, the effort yielded successful personalized interventions that were not obtainable for other severe childhood conditions.⁴⁰ EPINET hopes to automate much of these stepwise advances with a learning health system. This program relies on data routinely collected in clinical practice to drive the process of scientific discovery. Specifically, it determines the relationships between clinical features, biologic measures, treatment characteristics, and symptomatic and functional outcomes. EPINET aims to accelerate our understanding of biomarkers of psychosis risk and onset, as well as factors associated with recovery and cure. Dashboard displays of outcomes will allow for real-time comparisons within and across early intervention clinics. This in turn identifies performance gaps and drives continuous quality improvement.

Continue to: Barriers to optimizing program efficacy for both models

Unfortunately, there are stark differences between ALL and severe mental disorders that potentially jeopardize the achievement of these aims, despite the advances in data analytic abilities that drive the learning health system. Specifically, the heterogeneity of psychotic illnesses and the absence of reliable prognostic and modifiable risk markers (responsible for failed efforts to enhance treatment of serious mental illness over the last half century^1,2,41) are unlikely to be resolved by a learning health system. These measures are vital to determine whether specific interventions are effective, particularly given the absence of a randomized control group in the EPINET/learning health system design. Fortunately, however, the National Institutes for Health has recently initiated the Accelerating Medicines Partnership–Schizophrenia (AMP-SCZ). This approach seeks “promising biological markers that can help identify those at risk of developing schizophrenia as early as possible, track the progression of symptoms and other outcomes and ultimately define targets for treatment development.”⁴² The Box^{1,4,9,10,36,41,43-45} describes some of the challenges involved in identifying biomarkers of severe mental illness.

Box

A strength-based approach

The absence of sufficiently powered RCTs for prevention studies and the reliance on intermediate outcomes for FEP studies leaves unanswered whether such programs can effectively prevent chronic psychosis at a cost society is willing to pay. Still, substantial evidence indicates that outreach, long-acting injectable antipsychotics, early consideration of clozapine, family therapy, CBT for psychosis/attenuated psychosis, and services focused on competitive employment can preserve social and occupational functioning.^16,34 Until these broader questions are more definitively addressed, it seems reasonable to apply what we have learned (Table^{11,12,35,37-39,46}).

Simply avoiding the most divisive aspects of the medical model that inadvertently promote stigma and undercut self-confidence may help maintain patients’ willingness to learn how best to apply their strengths and manage their limitations.¹¹ The progression to enduring psychotic features (eg, fixed delusions) may reflect ongoing social isolation and alienation. A strength-based approach seeks first to establish common goals (eg, school, work, friends, family support, housing, leaving home) and then works to empower the patient to successfully reach those goals.³⁵ This typically involves giving them the opportunity to fail, avoiding criticism when they do, and focusing on these experiences as learning opportunities from which success can ultimately result.

It is difficult to offer all these services in a typical private practice setting. Instead, it may make more sense to use one of the hundreds of early intervention services programs in the United States.⁴⁶ If a psychiatric clinician is dedicated to working with this population, it may also be possible to establish ongoing relationships with primary care physicians, family and CBT therapists, family support services (eg, National Alliance on Mental Illness), caseworkers and employment counselors. In essence, a psychiatrist may be able re-create a multidisciplinary effort by taking advantage of the expertise of these various professionals. The challenge is to create a consistent message for patients and families in the absence of regular meetings with the clinical team, although the recent reliance on and improved sophistication of virtual meetings may help. Psychiatrists often play a critical role even when the patient is not prescribed medication, partly because they are most comfortable handling the risks and may have the most comprehensive understanding of the issues at play. When medications are appropriate and patients with FEP are willing to take them, early consideration of long-acting injectable antipsychotics and clozapine may provide better stabilization and diminish the risk of earlier and more frequent relapses.

Bottom Line

Early interventions for psychosis include the prevention model and the first-episode recovery model. It is difficult to assess, compare, and optimize the effectiveness of such programs. Current evidence supports a ‘strength-based’ approach focused on finding common ground between patients, their support system, and the treatment team.

Related Resources

• Early Assessment and Support Alliance. National Early Psychosis Directory. https://easacommunity.org/nationaldirectory.php
• Kane JM, Robinson DG, Schooler NR, et al. Comprehensive versus usual community care for first-episode psychosis: 2-year outcomes from the NIMH RAISE Early Treatment Program. Am J Psychiatry. 2016 ;173(4):362-372

Drug Brand Name

Clozapine • Clozaril

Neuroscience research over the past half century has failed to significantly advance the treatment of severe mental illness.^1,2 Hence, evidence that a longer duration of untreated psychosis (DUP) aggravates—and early intervention with medication and social supports ameliorates—the long-term adverse consequences of psychotic disorders generated a great deal of interest.^3,4 This knowledge led to the development of diverse early intervention services worldwide aimed at this putative “critical window.” It raised the possibility that appropriate interventions could prevent the long-term disability that makes chronic psychosis one of the most debilitating disorders.^5,6 However, even beyond the varied cultural and economic confounds, it is difficult to assess, compare, and optimize program effectiveness.⁷ Obstacles include paucity of sufficiently powered, well-designed randomized controlled trials (RCTs), the absence of diagnostic biomarkers or other prognostic indicators to better account for the inherent heterogeneity in the population and associated outcomes, and the absence of modifiable risk factors that can guide interventions and provide intermediate outcomes.^4,8-10

To better appreciate these issues, it is important to distinguish whether a program is designed to prevent psychosis, or to mitigate the effects of psychosis. Two models include the:

• Prevention model, which focuses on young individuals who are not yet overtly psychotic but at high risk
• First-episode recovery model, which focuses on those who have experienced a first episode of psychosis (FEP) but have not yet developed a chronic disorder.

Both models share long-term goals and are hampered by many of the same issues summarized above. They both deviate markedly from the standard medical model by including psychosocial services designed to promote restoration of a self-defined trajectory to greater independence.^11-14 The 2 differ, however, in the challenges they must overcome to produce their sample populations and establish effective interventions.^10,15,16

In this article, we provide a succinct overview of these issues and a set of recommendations based on a “strength-based” approach. This approach focuses on finding common ground between patients, their support system, and the treatment team in the service of empowering patients to resume responsibility for transition to adulthood.

The prevention model

While most prevention initiatives in medicine rely on the growing ability to target specific pathophysiologic pathways,³ preventing psychosis relies on clinical evidence showing that DUP and early interventions predict a better course of severe mental illness.¹⁷ In contrast, initiatives such as normalizing neonatal neuronal pathways are more consistent with the strategy utilized in other fields but have yet to yield a pathophysiologic target for psychosis.^3,18

Initial efforts to identify ‘at-risk’ individuals

The prevention model of psychosis is based on the ability to identify young individuals at high risk for developing a psychotic disorder (Figure). The first screening measures were focused on prodromal psychosis (eg, significant loss of function, family history, and “intermittent” and “attenuated” psychotic symptoms). When applied to referred (ie, pre-screened) samples, 30% to 40% of this group who met criteria transitioned to psychosis over the next 1 to 3 years despite antidepressant and psychosocial interventions.¹⁹ Comprising 8 academic medical centers, the North American Prodrome Longitudinal Study (NAPLS) produced similar results using the Structured Interview for Prodromal Syndromes (SIPS).¹⁷ Thus, 30% to 50% of pre-screened individuals referred by school counselors and mental health professionals met SIPS criteria, and 35% of these individuals transitioned to psychosis over 30 months. The validity of this measure was further supported by the fact that higher baseline levels of unusual thought content, suspicion/paranoia, social impairment, and substance abuse successfully distinguished approximately 80% of those who transitioned to psychosis. The results of this first generation of screening studies were exciting because they seemed to demonstrate that highly concentrated samples of young persons at high risk of developing psychosis could be identified, and that fine-tuning the screening criteria could produce even more enriched samples (ie, positive predictive power).

Initial interventions produced promising results

The development of effective screening measures led to reports of effective treatment interventions. These were largely applied in a clinical staging model that restricted antipsychotic medications to those who failed to improve after receiving potentially “less toxic” interventions (eg, omega-3 polyunsaturated fatty acids and other antioxidants; psychotherapy; cognitive-behavioral therapy [CBT]; family therapy).⁵ While study designs were typically quasi-experimental, the interventions appeared to dramatically diminish the transition to psychosis (ie, approximately 50%).

Continue to: The first generation...

The first generation of RCTs appeared to confirm these results, although sample sizes were small, and most study designs assessed only a single intervention. Initial meta-analyses of these data reported that both CBT and antipsychotics appeared to prevent approximately one-half of individuals from becoming psychotic at 12 months, and more than one-third at 2 to 4 years, compared with treatment as usual.²⁰

While some researchers challenged the validity of these findings,^21-23 the results generated tremendous international enthusiasm and calls for widespread implementation.⁶ The number of early intervention services (EIS) centers increased dramatically worldwide, and in 2014 the National Institute for Health and Care Excellence released standards for interventions to prevent transition to psychosis.²⁴ These included close monitoring, CBT and family interventions, and avoiding antipsychotics when possible.²⁴

Focusing on sensitivity over specificity

The first generation of studies generated by the prevention model relied on outreach programs or referrals, which produced small samples of carefully selected, pre-screened individuals (Figure, Pre-screened) who were then screened again to establish the high-risk sample.²⁵ While approximately 33% of these individuals became psychotic, the screening process required a very efficient means of eliminating those not at high-risk (given the ultimate target population represented only approximately .5% of young people) (Figure). The pre-screening and screening processes in these first-generation studies were labor-intensive but could only identify approximately 5% of those individuals destined to become psychotic over the next 2 or 3 years. Thus, alternative methods to enhance sensitivity were needed to extend programming to the general population.

Second-generation pre-screening (Figure; Step 1). New pre-screening methods were identified that captured more individuals destined to become psychotic. For example, approximately 90% of this population were registered in health care organizations (eg, health maintenance organizations) and received a psychiatric diagnosis in the year prior to the onset of psychosis (true positives).⁸ These samples, however, contained a much higher percentage of persons not destined to become psychotic, and somehow the issue of specificity (decreasing false positives) was minimized.^8,9 For example, pre-screened samples contained 20 to 50 individuals not destined to become psychotic for each one who did.²⁶ Since screening measures could only eliminate approximately 20% of this group (Figure, Step 2, page 25), second-generation transition rates fell from 30% to 40% to 2% to 10%.^27,28

Other pre-screening approaches were introduced, but they also focused on capturing more of those destined to become psychotic (sensitivity) than eliminating those who would not (specificity). For instance, Australia opened more than 100 “Headspace” community centers nationwide designed to promote engagement and self-esteem in youth experiencing anxiety; depression; stress; relationship, work, or school problems; or bullying.¹³ Most services were free and included mental health staff who screened for psychosis and provided a wide range of services in a destigmatized setting. These methods identified at least an additional 5% to 7% of individuals destined to become psychotic, but to our knowledge, no data have been published on whether they helped eliminate those who did not.

Continue to: Second-generation screening

Second-generation screening (Figure, Step 2). A second screening aims to retain those pre-screened individuals who will become psychotic (ie, minimizing false negatives) while further minimizing those who do not (ie, minimizing false positives). The addition of cognitive, neural (eg, structural MRI; neurophysiologic), and biochemical (eg, inflammatory immune and stress) markers to the risk calculators have produced a sensitivity close to 100%.^8,9 Unfortunately, these studies downplayed specificity, which remained approximately 20%.^8,9 Specificity is critical not just because of concerns about stigma (ie, labeling people as pre-psychotic when they are not) but also because of the adverse effects of antipsychotic medications and the effects on future program development (interventions are costly and labor-intensive). Also, diluting the pool with individuals not at risk makes it nearly impossible to identify effective interventions (ie, power).^27,28

While some studies focused on increasing specificity (to approximately 75%), this leads to an unacceptable loss of sensitivity (from 90% to 60%),²⁹ with 40% of pre-screened individuals who would become psychotic being eliminated from the study population. The addition of other biological markers (eg, salivary cortisol)³⁰ and use of learning health systems may be able to enhance these numbers (initial reports of specificity = 87% and sensitivity = 85%).^8,9 This is accomplished by integrating artificial and human intelligence measures of clinical (symptom and neurocognitive measures) and biological (eg, polygenetic risk scores; gray matter volume) variables.³¹ However, even if these results are replicated, more effective pre-screening measures will be required.

Identifying a suitable sample population for prevention program studies is clearly more complicated than for FEP studies, where one can usually identify many of those in the at-risk population by their first hospitalization for psychotic symptoms. The issues of false positives (eg, substance-induced psychosis) and negatives (eg, slow deterioration, prominent negative symptoms) are important concerns, but proportionately far less significant.

Prevention and FEP interventions

Once a study sample is constituted, 1 to 3 years of treatment interventions are initiated. Interventions for prevention programs typically include CBT directed at attenuated psychosis (eg, reframing or de-catastrophizing unusual thoughts and minimizing distress associated with unusual perceptions); case management to facilitate personal, educational, and vocational goals; and family therapy in single or multi-group formats to educate one’s support system about the risk state and to minimize adverse familial responses.¹⁴ Many programs also include supported education or employment services to promote reintegration in age-appropriate activities; group therapy focused on substance abuse and social skills training; cognitive remediation to ameliorate the cognitive dysfunction; and an array of pharmacologic interventions designed to delay or prevent transition to psychosis or to alleviate symptoms. While most interventions are similar, FEP programs have recently included peer support staff. This appears to instill hope in newly diagnosed patients, provide role models, and provide peer supporters an opportunity to use their experiences to help others and earn income.³²

The breadth and depth of these services are critical because retention in the program is highly dependent on participant engagement, which in turn is highly dependent on whether the program can help individuals get what they want (eg, friends, employment, education, more autonomy, physical health). The setting and atmosphere of the treatment program and the willingness/ability of staff to meet participants in the community are also important elements.^11,12 In this context, the Headspace community centers are having an impact far beyond Australia and may prove to be a particularly good model.¹³

Continue to: Assessing prevention and FEP interventions

The second generation of studies of prevention programs has not confirmed, let alone extended, the earlier findings and meta-analyses. A 2020 report concluded CBT was still the most promising intervention; it was more effective than control treatments at 12 and 18 months, although not at 6, 24, or 48 months.³³ This review included controlled, open-label, and naturalistic studies that assessed family therapy; omega-3 polyunsaturated fatty acids; integrated psychological therapy (a package of interventions that included family education, CBT, social skills training, and cognitive remediation); N-methyl-D-aspartate receptor modulators; mood stabilizers; and antipsychotics. In addition to the evidence supporting CBT, the results also indicated nonsignificant trends favoring family and integrated psychological therapy. Neither a 2019 Cochrane review³⁴ nor a 2020 “umbrella” assessment of 42 meta-analyses⁹ found convincing evidence for the efficacy of any program components.

While these disappointing findings are at least partly attributable to the methodological challenges described above and in the Figure, other factors may hinder establishing effective interventions. In contrast to FEP studies, those focused on prevention had a very ambitious agenda (eliminating psychosis) and tended to downplay more modest intermediate outcomes. These studies also tended to assess new ideas with small samples rather than pursue promising findings with larger multi-site studies focused on a group of interventions. The authors of a Cochrane review observed “There is the impression that in this whole area there is a triumph of hope over adversity. There is the repeated hope invested in another—often unique—study question and then a study of fewer than 100 participants are completed. This results in the set of comparisons reported here, all 9 of which are too underpowered to really highlight clear differences.”³⁴ To use a baseball analogy, it seems that investigators are “swinging for the fence” when a few singles are what’s really needed.

From the outset, the goals of FEP studies were more modest, largely ignoring the task of developing consensus definitions of recovery that require following patients for up to 5 to 10 years. Instead, they use intermediate endpoints based on adapting treatments that already appeared effective in patients with chronic mental disorders.³⁵ As a consequence, researchers examining FEP demonstrated clear, albeit limited, salutary effects using large multi-site trials and previously established outcome measures.^3,10,36 For instance, the Recovery After an Initial Schizophrenia Episode-Early Treatment Program (RAISE-ETP) study was a 2-year, multi-site RCT (N = 404) funded by the National Institute of Mental Health (NIMH). The investigators reported improved indices of social function (eg, quality of life; education and work participation) and total ratings of psychopathology and depression compared with treatment as usual. Furthermore, they established that DUP predicted treatment response.³⁵ The latter finding was underscored by improvement being limited to the 50% with <74 weeks DUP. Annual costs of the program per 1 standard deviation improvement in quality of life were approximately $1,000 for patients with <74 weeks DUP and $40,000 for those with >74 weeks DUP. Concurrent meta-analyses confirmed and extended these findings,¹⁶ showing higher remission rates; diminished relapses and hospital admissions; greater engagement in programming; greater involvement in work and school; improved quality of life; and other steps toward recovery. These studies were also able to establish a clear benefit of antipsychotic medications, particularly a high acceptance of long-acting injectable antipsychotic formulations, which promoted adherence and decreased some adverse events³⁷; and early use of clozapine therapy, which improved remission rates and longer-term outcomes.³⁸ Other findings underscored the need to anticipate and address new problems associated with effective antipsychotic therapy (eg, antipsychotic response correlates with weight gain, a particularly intolerable adverse event for this age group).³⁹ Providing pre-emptive strategies such as exercise groups and nutritional education may be necessary to maintain adherence.

Limitations of FEP studies

The effect sizes in these FEP studies were small to medium on outcome measures tracking recovery and associated indicators (eg, global functioning, school/work participation, treatment engagement); the number needed to treat for each of these was >10. There is no clear evidence that recovery programs such as RAISE-ETP actually reduce longer-term disability. Most studies showed disability payments increased while clinical benefits tended to fade over time. In addition, by grouping interventions together, the studies made it difficult to identify effective vs ineffective treatments, let alone determine how best to personalize therapy for participants in future studies.

The next generation of FEP studies

While limited in scope, the results of the recent FEP studies justify a next generation of recovery interventions designed to address these shortcomings and optimize program outcomes.³⁹ Most previous FEP studies were conducted in community mental health center settings, thus eliminating the need to transition services developed in academia into the “real world.” The next generation of NIMH studies will be primarily conducted in analogous settings under the Early Psychosis Intervention Network (EPINET).⁴⁰ EPINET’s study design echoes that responsible for the stepwise successes in the late 20th century that produced cures for the deadliest childhood cancer, acute lymphoblastic leukemia (ALL). This disease was successfully treated by modifying diverse evidence-based practices without relying on pharmacologic or other major treatment breakthroughs. Despite this, the effort yielded successful personalized interventions that were not obtainable for other severe childhood conditions.⁴⁰ EPINET hopes to automate much of these stepwise advances with a learning health system. This program relies on data routinely collected in clinical practice to drive the process of scientific discovery. Specifically, it determines the relationships between clinical features, biologic measures, treatment characteristics, and symptomatic and functional outcomes. EPINET aims to accelerate our understanding of biomarkers of psychosis risk and onset, as well as factors associated with recovery and cure. Dashboard displays of outcomes will allow for real-time comparisons within and across early intervention clinics. This in turn identifies performance gaps and drives continuous quality improvement.

Continue to: Barriers to optimizing program efficacy for both models

Unfortunately, there are stark differences between ALL and severe mental disorders that potentially jeopardize the achievement of these aims, despite the advances in data analytic abilities that drive the learning health system. Specifically, the heterogeneity of psychotic illnesses and the absence of reliable prognostic and modifiable risk markers (responsible for failed efforts to enhance treatment of serious mental illness over the last half century^1,2,41) are unlikely to be resolved by a learning health system. These measures are vital to determine whether specific interventions are effective, particularly given the absence of a randomized control group in the EPINET/learning health system design. Fortunately, however, the National Institutes for Health has recently initiated the Accelerating Medicines Partnership–Schizophrenia (AMP-SCZ). This approach seeks “promising biological markers that can help identify those at risk of developing schizophrenia as early as possible, track the progression of symptoms and other outcomes and ultimately define targets for treatment development.”⁴² The Box^{1,4,9,10,36,41,43-45} describes some of the challenges involved in identifying biomarkers of severe mental illness.

Box

A strength-based approach

The absence of sufficiently powered RCTs for prevention studies and the reliance on intermediate outcomes for FEP studies leaves unanswered whether such programs can effectively prevent chronic psychosis at a cost society is willing to pay. Still, substantial evidence indicates that outreach, long-acting injectable antipsychotics, early consideration of clozapine, family therapy, CBT for psychosis/attenuated psychosis, and services focused on competitive employment can preserve social and occupational functioning.^16,34 Until these broader questions are more definitively addressed, it seems reasonable to apply what we have learned (Table^{11,12,35,37-39,46}).

Simply avoiding the most divisive aspects of the medical model that inadvertently promote stigma and undercut self-confidence may help maintain patients’ willingness to learn how best to apply their strengths and manage their limitations.¹¹ The progression to enduring psychotic features (eg, fixed delusions) may reflect ongoing social isolation and alienation. A strength-based approach seeks first to establish common goals (eg, school, work, friends, family support, housing, leaving home) and then works to empower the patient to successfully reach those goals.³⁵ This typically involves giving them the opportunity to fail, avoiding criticism when they do, and focusing on these experiences as learning opportunities from which success can ultimately result.

It is difficult to offer all these services in a typical private practice setting. Instead, it may make more sense to use one of the hundreds of early intervention services programs in the United States.⁴⁶ If a psychiatric clinician is dedicated to working with this population, it may also be possible to establish ongoing relationships with primary care physicians, family and CBT therapists, family support services (eg, National Alliance on Mental Illness), caseworkers and employment counselors. In essence, a psychiatrist may be able re-create a multidisciplinary effort by taking advantage of the expertise of these various professionals. The challenge is to create a consistent message for patients and families in the absence of regular meetings with the clinical team, although the recent reliance on and improved sophistication of virtual meetings may help. Psychiatrists often play a critical role even when the patient is not prescribed medication, partly because they are most comfortable handling the risks and may have the most comprehensive understanding of the issues at play. When medications are appropriate and patients with FEP are willing to take them, early consideration of long-acting injectable antipsychotics and clozapine may provide better stabilization and diminish the risk of earlier and more frequent relapses.

Bottom Line

Early interventions for psychosis include the prevention model and the first-episode recovery model. It is difficult to assess, compare, and optimize the effectiveness of such programs. Current evidence supports a ‘strength-based’ approach focused on finding common ground between patients, their support system, and the treatment team.

Related Resources

• Early Assessment and Support Alliance. National Early Psychosis Directory. https://easacommunity.org/nationaldirectory.php
• Kane JM, Robinson DG, Schooler NR, et al. Comprehensive versus usual community care for first-episode psychosis: 2-year outcomes from the NIMH RAISE Early Treatment Program. Am J Psychiatry. 2016 ;173(4):362-372

Drug Brand Name

Clozapine • Clozaril

Neuroscience research over the past half century has failed to significantly advance the treatment of severe mental illness.^1,2 Hence, evidence that a longer duration of untreated psychosis (DUP) aggravates—and early intervention with medication and social supports ameliorates—the long-term adverse consequences of psychotic disorders generated a great deal of interest.^3,4 This knowledge led to the development of diverse early intervention services worldwide aimed at this putative “critical window.” It raised the possibility that appropriate interventions could prevent the long-term disability that makes chronic psychosis one of the most debilitating disorders.^5,6 However, even beyond the varied cultural and economic confounds, it is difficult to assess, compare, and optimize program effectiveness.⁷ Obstacles include paucity of sufficiently powered, well-designed randomized controlled trials (RCTs), the absence of diagnostic biomarkers or other prognostic indicators to better account for the inherent heterogeneity in the population and associated outcomes, and the absence of modifiable risk factors that can guide interventions and provide intermediate outcomes.^4,8-10

To better appreciate these issues, it is important to distinguish whether a program is designed to prevent psychosis, or to mitigate the effects of psychosis. Two models include the:

• Prevention model, which focuses on young individuals who are not yet overtly psychotic but at high risk
• First-episode recovery model, which focuses on those who have experienced a first episode of psychosis (FEP) but have not yet developed a chronic disorder.

Both models share long-term goals and are hampered by many of the same issues summarized above. They both deviate markedly from the standard medical model by including psychosocial services designed to promote restoration of a self-defined trajectory to greater independence.^11-14 The 2 differ, however, in the challenges they must overcome to produce their sample populations and establish effective interventions.^10,15,16

In this article, we provide a succinct overview of these issues and a set of recommendations based on a “strength-based” approach. This approach focuses on finding common ground between patients, their support system, and the treatment team in the service of empowering patients to resume responsibility for transition to adulthood.

The prevention model

While most prevention initiatives in medicine rely on the growing ability to target specific pathophysiologic pathways,³ preventing psychosis relies on clinical evidence showing that DUP and early interventions predict a better course of severe mental illness.¹⁷ In contrast, initiatives such as normalizing neonatal neuronal pathways are more consistent with the strategy utilized in other fields but have yet to yield a pathophysiologic target for psychosis.^3,18

Initial efforts to identify ‘at-risk’ individuals

The prevention model of psychosis is based on the ability to identify young individuals at high risk for developing a psychotic disorder (Figure). The first screening measures were focused on prodromal psychosis (eg, significant loss of function, family history, and “intermittent” and “attenuated” psychotic symptoms). When applied to referred (ie, pre-screened) samples, 30% to 40% of this group who met criteria transitioned to psychosis over the next 1 to 3 years despite antidepressant and psychosocial interventions.¹⁹ Comprising 8 academic medical centers, the North American Prodrome Longitudinal Study (NAPLS) produced similar results using the Structured Interview for Prodromal Syndromes (SIPS).¹⁷ Thus, 30% to 50% of pre-screened individuals referred by school counselors and mental health professionals met SIPS criteria, and 35% of these individuals transitioned to psychosis over 30 months. The validity of this measure was further supported by the fact that higher baseline levels of unusual thought content, suspicion/paranoia, social impairment, and substance abuse successfully distinguished approximately 80% of those who transitioned to psychosis. The results of this first generation of screening studies were exciting because they seemed to demonstrate that highly concentrated samples of young persons at high risk of developing psychosis could be identified, and that fine-tuning the screening criteria could produce even more enriched samples (ie, positive predictive power).

Initial interventions produced promising results

The development of effective screening measures led to reports of effective treatment interventions. These were largely applied in a clinical staging model that restricted antipsychotic medications to those who failed to improve after receiving potentially “less toxic” interventions (eg, omega-3 polyunsaturated fatty acids and other antioxidants; psychotherapy; cognitive-behavioral therapy [CBT]; family therapy).⁵ While study designs were typically quasi-experimental, the interventions appeared to dramatically diminish the transition to psychosis (ie, approximately 50%).

Continue to: The first generation...

The first generation of RCTs appeared to confirm these results, although sample sizes were small, and most study designs assessed only a single intervention. Initial meta-analyses of these data reported that both CBT and antipsychotics appeared to prevent approximately one-half of individuals from becoming psychotic at 12 months, and more than one-third at 2 to 4 years, compared with treatment as usual.²⁰

While some researchers challenged the validity of these findings,^21-23 the results generated tremendous international enthusiasm and calls for widespread implementation.⁶ The number of early intervention services (EIS) centers increased dramatically worldwide, and in 2014 the National Institute for Health and Care Excellence released standards for interventions to prevent transition to psychosis.²⁴ These included close monitoring, CBT and family interventions, and avoiding antipsychotics when possible.²⁴

Focusing on sensitivity over specificity

The first generation of studies generated by the prevention model relied on outreach programs or referrals, which produced small samples of carefully selected, pre-screened individuals (Figure, Pre-screened) who were then screened again to establish the high-risk sample.²⁵ While approximately 33% of these individuals became psychotic, the screening process required a very efficient means of eliminating those not at high-risk (given the ultimate target population represented only approximately .5% of young people) (Figure). The pre-screening and screening processes in these first-generation studies were labor-intensive but could only identify approximately 5% of those individuals destined to become psychotic over the next 2 or 3 years. Thus, alternative methods to enhance sensitivity were needed to extend programming to the general population.

Second-generation pre-screening (Figure; Step 1). New pre-screening methods were identified that captured more individuals destined to become psychotic. For example, approximately 90% of this population were registered in health care organizations (eg, health maintenance organizations) and received a psychiatric diagnosis in the year prior to the onset of psychosis (true positives).⁸ These samples, however, contained a much higher percentage of persons not destined to become psychotic, and somehow the issue of specificity (decreasing false positives) was minimized.^8,9 For example, pre-screened samples contained 20 to 50 individuals not destined to become psychotic for each one who did.²⁶ Since screening measures could only eliminate approximately 20% of this group (Figure, Step 2, page 25), second-generation transition rates fell from 30% to 40% to 2% to 10%.^27,28

Other pre-screening approaches were introduced, but they also focused on capturing more of those destined to become psychotic (sensitivity) than eliminating those who would not (specificity). For instance, Australia opened more than 100 “Headspace” community centers nationwide designed to promote engagement and self-esteem in youth experiencing anxiety; depression; stress; relationship, work, or school problems; or bullying.¹³ Most services were free and included mental health staff who screened for psychosis and provided a wide range of services in a destigmatized setting. These methods identified at least an additional 5% to 7% of individuals destined to become psychotic, but to our knowledge, no data have been published on whether they helped eliminate those who did not.

Continue to: Second-generation screening

Second-generation screening (Figure, Step 2). A second screening aims to retain those pre-screened individuals who will become psychotic (ie, minimizing false negatives) while further minimizing those who do not (ie, minimizing false positives). The addition of cognitive, neural (eg, structural MRI; neurophysiologic), and biochemical (eg, inflammatory immune and stress) markers to the risk calculators have produced a sensitivity close to 100%.^8,9 Unfortunately, these studies downplayed specificity, which remained approximately 20%.^8,9 Specificity is critical not just because of concerns about stigma (ie, labeling people as pre-psychotic when they are not) but also because of the adverse effects of antipsychotic medications and the effects on future program development (interventions are costly and labor-intensive). Also, diluting the pool with individuals not at risk makes it nearly impossible to identify effective interventions (ie, power).^27,28

While some studies focused on increasing specificity (to approximately 75%), this leads to an unacceptable loss of sensitivity (from 90% to 60%),²⁹ with 40% of pre-screened individuals who would become psychotic being eliminated from the study population. The addition of other biological markers (eg, salivary cortisol)³⁰ and use of learning health systems may be able to enhance these numbers (initial reports of specificity = 87% and sensitivity = 85%).^8,9 This is accomplished by integrating artificial and human intelligence measures of clinical (symptom and neurocognitive measures) and biological (eg, polygenetic risk scores; gray matter volume) variables.³¹ However, even if these results are replicated, more effective pre-screening measures will be required.

Identifying a suitable sample population for prevention program studies is clearly more complicated than for FEP studies, where one can usually identify many of those in the at-risk population by their first hospitalization for psychotic symptoms. The issues of false positives (eg, substance-induced psychosis) and negatives (eg, slow deterioration, prominent negative symptoms) are important concerns, but proportionately far less significant.

Prevention and FEP interventions

Once a study sample is constituted, 1 to 3 years of treatment interventions are initiated. Interventions for prevention programs typically include CBT directed at attenuated psychosis (eg, reframing or de-catastrophizing unusual thoughts and minimizing distress associated with unusual perceptions); case management to facilitate personal, educational, and vocational goals; and family therapy in single or multi-group formats to educate one’s support system about the risk state and to minimize adverse familial responses.¹⁴ Many programs also include supported education or employment services to promote reintegration in age-appropriate activities; group therapy focused on substance abuse and social skills training; cognitive remediation to ameliorate the cognitive dysfunction; and an array of pharmacologic interventions designed to delay or prevent transition to psychosis or to alleviate symptoms. While most interventions are similar, FEP programs have recently included peer support staff. This appears to instill hope in newly diagnosed patients, provide role models, and provide peer supporters an opportunity to use their experiences to help others and earn income.³²

The breadth and depth of these services are critical because retention in the program is highly dependent on participant engagement, which in turn is highly dependent on whether the program can help individuals get what they want (eg, friends, employment, education, more autonomy, physical health). The setting and atmosphere of the treatment program and the willingness/ability of staff to meet participants in the community are also important elements.^11,12 In this context, the Headspace community centers are having an impact far beyond Australia and may prove to be a particularly good model.¹³

Continue to: Assessing prevention and FEP interventions

The second generation of studies of prevention programs has not confirmed, let alone extended, the earlier findings and meta-analyses. A 2020 report concluded CBT was still the most promising intervention; it was more effective than control treatments at 12 and 18 months, although not at 6, 24, or 48 months.³³ This review included controlled, open-label, and naturalistic studies that assessed family therapy; omega-3 polyunsaturated fatty acids; integrated psychological therapy (a package of interventions that included family education, CBT, social skills training, and cognitive remediation); N-methyl-D-aspartate receptor modulators; mood stabilizers; and antipsychotics. In addition to the evidence supporting CBT, the results also indicated nonsignificant trends favoring family and integrated psychological therapy. Neither a 2019 Cochrane review³⁴ nor a 2020 “umbrella” assessment of 42 meta-analyses⁹ found convincing evidence for the efficacy of any program components.

While these disappointing findings are at least partly attributable to the methodological challenges described above and in the Figure, other factors may hinder establishing effective interventions. In contrast to FEP studies, those focused on prevention had a very ambitious agenda (eliminating psychosis) and tended to downplay more modest intermediate outcomes. These studies also tended to assess new ideas with small samples rather than pursue promising findings with larger multi-site studies focused on a group of interventions. The authors of a Cochrane review observed “There is the impression that in this whole area there is a triumph of hope over adversity. There is the repeated hope invested in another—often unique—study question and then a study of fewer than 100 participants are completed. This results in the set of comparisons reported here, all 9 of which are too underpowered to really highlight clear differences.”³⁴ To use a baseball analogy, it seems that investigators are “swinging for the fence” when a few singles are what’s really needed.

From the outset, the goals of FEP studies were more modest, largely ignoring the task of developing consensus definitions of recovery that require following patients for up to 5 to 10 years. Instead, they use intermediate endpoints based on adapting treatments that already appeared effective in patients with chronic mental disorders.³⁵ As a consequence, researchers examining FEP demonstrated clear, albeit limited, salutary effects using large multi-site trials and previously established outcome measures.^3,10,36 For instance, the Recovery After an Initial Schizophrenia Episode-Early Treatment Program (RAISE-ETP) study was a 2-year, multi-site RCT (N = 404) funded by the National Institute of Mental Health (NIMH). The investigators reported improved indices of social function (eg, quality of life; education and work participation) and total ratings of psychopathology and depression compared with treatment as usual. Furthermore, they established that DUP predicted treatment response.³⁵ The latter finding was underscored by improvement being limited to the 50% with <74 weeks DUP. Annual costs of the program per 1 standard deviation improvement in quality of life were approximately $1,000 for patients with <74 weeks DUP and $40,000 for those with >74 weeks DUP. Concurrent meta-analyses confirmed and extended these findings,¹⁶ showing higher remission rates; diminished relapses and hospital admissions; greater engagement in programming; greater involvement in work and school; improved quality of life; and other steps toward recovery. These studies were also able to establish a clear benefit of antipsychotic medications, particularly a high acceptance of long-acting injectable antipsychotic formulations, which promoted adherence and decreased some adverse events³⁷; and early use of clozapine therapy, which improved remission rates and longer-term outcomes.³⁸ Other findings underscored the need to anticipate and address new problems associated with effective antipsychotic therapy (eg, antipsychotic response correlates with weight gain, a particularly intolerable adverse event for this age group).³⁹ Providing pre-emptive strategies such as exercise groups and nutritional education may be necessary to maintain adherence.

Limitations of FEP studies

The effect sizes in these FEP studies were small to medium on outcome measures tracking recovery and associated indicators (eg, global functioning, school/work participation, treatment engagement); the number needed to treat for each of these was >10. There is no clear evidence that recovery programs such as RAISE-ETP actually reduce longer-term disability. Most studies showed disability payments increased while clinical benefits tended to fade over time. In addition, by grouping interventions together, the studies made it difficult to identify effective vs ineffective treatments, let alone determine how best to personalize therapy for participants in future studies.

The next generation of FEP studies

While limited in scope, the results of the recent FEP studies justify a next generation of recovery interventions designed to address these shortcomings and optimize program outcomes.³⁹ Most previous FEP studies were conducted in community mental health center settings, thus eliminating the need to transition services developed in academia into the “real world.” The next generation of NIMH studies will be primarily conducted in analogous settings under the Early Psychosis Intervention Network (EPINET).⁴⁰ EPINET’s study design echoes that responsible for the stepwise successes in the late 20th century that produced cures for the deadliest childhood cancer, acute lymphoblastic leukemia (ALL). This disease was successfully treated by modifying diverse evidence-based practices without relying on pharmacologic or other major treatment breakthroughs. Despite this, the effort yielded successful personalized interventions that were not obtainable for other severe childhood conditions.⁴⁰ EPINET hopes to automate much of these stepwise advances with a learning health system. This program relies on data routinely collected in clinical practice to drive the process of scientific discovery. Specifically, it determines the relationships between clinical features, biologic measures, treatment characteristics, and symptomatic and functional outcomes. EPINET aims to accelerate our understanding of biomarkers of psychosis risk and onset, as well as factors associated with recovery and cure. Dashboard displays of outcomes will allow for real-time comparisons within and across early intervention clinics. This in turn identifies performance gaps and drives continuous quality improvement.

Continue to: Barriers to optimizing program efficacy for both models

Unfortunately, there are stark differences between ALL and severe mental disorders that potentially jeopardize the achievement of these aims, despite the advances in data analytic abilities that drive the learning health system. Specifically, the heterogeneity of psychotic illnesses and the absence of reliable prognostic and modifiable risk markers (responsible for failed efforts to enhance treatment of serious mental illness over the last half century^1,2,41) are unlikely to be resolved by a learning health system. These measures are vital to determine whether specific interventions are effective, particularly given the absence of a randomized control group in the EPINET/learning health system design. Fortunately, however, the National Institutes for Health has recently initiated the Accelerating Medicines Partnership–Schizophrenia (AMP-SCZ). This approach seeks “promising biological markers that can help identify those at risk of developing schizophrenia as early as possible, track the progression of symptoms and other outcomes and ultimately define targets for treatment development.”⁴² The Box^{1,4,9,10,36,41,43-45} describes some of the challenges involved in identifying biomarkers of severe mental illness.

Box

A strength-based approach

The absence of sufficiently powered RCTs for prevention studies and the reliance on intermediate outcomes for FEP studies leaves unanswered whether such programs can effectively prevent chronic psychosis at a cost society is willing to pay. Still, substantial evidence indicates that outreach, long-acting injectable antipsychotics, early consideration of clozapine, family therapy, CBT for psychosis/attenuated psychosis, and services focused on competitive employment can preserve social and occupational functioning.^16,34 Until these broader questions are more definitively addressed, it seems reasonable to apply what we have learned (Table^{11,12,35,37-39,46}).

Simply avoiding the most divisive aspects of the medical model that inadvertently promote stigma and undercut self-confidence may help maintain patients’ willingness to learn how best to apply their strengths and manage their limitations.¹¹ The progression to enduring psychotic features (eg, fixed delusions) may reflect ongoing social isolation and alienation. A strength-based approach seeks first to establish common goals (eg, school, work, friends, family support, housing, leaving home) and then works to empower the patient to successfully reach those goals.³⁵ This typically involves giving them the opportunity to fail, avoiding criticism when they do, and focusing on these experiences as learning opportunities from which success can ultimately result.

It is difficult to offer all these services in a typical private practice setting. Instead, it may make more sense to use one of the hundreds of early intervention services programs in the United States.⁴⁶ If a psychiatric clinician is dedicated to working with this population, it may also be possible to establish ongoing relationships with primary care physicians, family and CBT therapists, family support services (eg, National Alliance on Mental Illness), caseworkers and employment counselors. In essence, a psychiatrist may be able re-create a multidisciplinary effort by taking advantage of the expertise of these various professionals. The challenge is to create a consistent message for patients and families in the absence of regular meetings with the clinical team, although the recent reliance on and improved sophistication of virtual meetings may help. Psychiatrists often play a critical role even when the patient is not prescribed medication, partly because they are most comfortable handling the risks and may have the most comprehensive understanding of the issues at play. When medications are appropriate and patients with FEP are willing to take them, early consideration of long-acting injectable antipsychotics and clozapine may provide better stabilization and diminish the risk of earlier and more frequent relapses.

Bottom Line

Early interventions for psychosis include the prevention model and the first-episode recovery model. It is difficult to assess, compare, and optimize the effectiveness of such programs. Current evidence supports a ‘strength-based’ approach focused on finding common ground between patients, their support system, and the treatment team.

Related Resources

• Early Assessment and Support Alliance. National Early Psychosis Directory. https://easacommunity.org/nationaldirectory.php
• Kane JM, Robinson DG, Schooler NR, et al. Comprehensive versus usual community care for first-episode psychosis: 2-year outcomes from the NIMH RAISE Early Treatment Program. Am J Psychiatry. 2016 ;173(4):362-372

Drug Brand Name

Clozapine • Clozaril

When patients with major depressive disorder (MDD) do not achieve optimal outcomes after FDA-approved first-line treatments and standard adjunctive strategies, clinicians look for additional approaches to alleviate their patients’ symptoms. Recent research suggests that several “nontraditional” treatments used primarily as adjuncts to standard antidepressants have promise for treatment-resistant depression.

In Part 1 of this article (Current Psychiatry, September 2021), we examined off-label medications. In Part 2, we will review other nontraditional approaches to treatment-resistant depression, including herbal/nutraceutical agents, anti-inflammatory/immune system therapies, device-based treatments, and other alternative approaches. Importantly, some treatments also demonstrate adverse effects (Table^1-32). With a careful consideration of the risk/benefit balance, this article reviews some of the better-studied nontraditional treatment options for patients with treatment-resistant depression.

Herbal/nutraceutical agents

This category encompasses a variety of commonly available “natural” options patients often ask about and at times self-prescribe. Examples evaluated in clinical trials include:

• vitamin D
• essential fatty acids (omega-3, omega-6)
• S-adenosyl-L-methionine (SAMe)
• hypericum perforatum (St. John’s Wort)
• probiotics.

Vitamin D deficiency has been linked to depression, possibly by lowering serotonin, norepinephrine, and dopamine concentrations.^1-3

A meta-analysis of 3 prospective, observational studies (N = 8,815) found an elevated risk of affective disorders in patients with low vitamin D levels.⁴ In addition, a systematic review and meta-analysis supported a potential role for vitamin D supplementation for patients with treatment-resistant depresssion.⁵

Toxicity can occur at levels >100 ng/mL, and resulting adverse effects may include weakness, fatigue, sleepiness, headache, loss of appetite, dry mouth, metallic taste, nausea, and vomiting. This vitamin can be considered as an adjunct to standard antidepressants, particularly in patients with treatment-resistant depression who have low vitamin D levels, but regular monitoring is necessary to avoid toxicity.

Essential fatty acids. Protein receptors embedded in lipid membranes and their binding affinities are influenced by omega-3 and omega-6 polyunsaturated fatty acids. Thus, essential fatty acids may benefit depression by maintaining membrane integrity and fluidity, as well as via their anti-inflammatory activity.

Continue to: Although results from...

Although results from controlled trials are mixed, a systematic review and meta-analysis of adjunctive nutraceuticals supported a potential role for essential fatty acids, primarily eicosapentaenoic acid (EPA), by itself or in combination with docosahexaenoic acid (DHA), with total EPA >60%.⁵ A second meta-analysis of 26 studies (N = 2,160) that considered only essential fatty acids concluded that EPA ≥60% at ≤1 g/d could benefit depression.⁶ Furthermore, omega-3 fatty acids may be helpful as an add-on agent for postpartum depression.⁷

Be aware that a diet rich in omega-6 greatly increases oxidized low-density lipoprotein levels in adipose tissue, potentially posing a cardiac risk factor. Clinicians need to be aware that self-prescribed use of essential fatty acids is common, and to ask about and monitor their patients’ use of these agents.

S-adenosyl-L-methionine (SAMe) is an intracellular amino acid and methyl donor. Among other actions, it is involved in the biosynthesis of hormones and neurotransmitters. There is promising but limited preliminary evidence of its efficacy and safety as a monotherapy or for antidepressant augmentation.⁸ For example, when compared with placebo for depressive symptoms in 19 randomized controlled trials (RCTs) (N = 878) ⁸:

• Five out of 6 earlier controlled studies reported SAMe IV (200 to 400 mg/d) or IM (45 to 50 mg/d) was more effective than placebo
• When the above studies were added to 14 subsequent studies for a meta-analysis, 12 of 19 RCTs reported that parenteral or oral SAMe was significantly more effective than placebo for depression (P < .05).

Overall, the safety and tolerability of SAMe are good. Common adverse effects include nausea, mild insomnia, dizziness, irritability, and anxiety. This is another compound widely available without a prescription and at times self-prescribed. It carries an acceptable risk/benefit balance, with decades of experience.

Hypericum perforatum (St. John’s Wort) is widely prescribed for depression in China and Europe, typically in doses ranging from 500 to 900 mg/d. Its mechanism of action in depression may relate to inhibition of serotonin, dopamine, and norepinephrine uptake from the synaptic cleft of these interconnecting neurotransmitter systems.

Continue to: A meta-analysis of 7 clinical trials...

A meta-analysis of 7 clinical trials (N = 3,808) comparing St. John’s Wort with various selective serotonin reuptake inhibitors (SSRIs) reported comparable rates of response (pooled relative risk .983, 95% CI .924 to 1.042; P < .001) and remission (pooled relative risk 1.013, 95% CI .892 to 1.134; P < .001).⁹ Further, there were significantly lower discontinuation/dropout rates (pooled odds ratio .587, 95% CI .478 to 0.697; P < .001) for St. John’s Wort compared with the SSRIs.

Existing evidence on the long-term efficacy and safety is limited (studies ranged from 4 to 12 weeks), as is evidence for patients with more severe depression or high suicidality.

Serious drug interactions include the potential for serotonin syndrome when St. John’s Wort is combined with certain antidepressants, compromised efficacy of benzodiazepines and standard antidepressants, and severe skin reactions to sun exposure. In addition, St. John’s Wort may not be safe to use during pregnancy or while breastfeeding. Because potential drug interactions can be serious and individuals often self-prescribe this agent, it is important to ask patients about their use of St. John’s Wort, and to be vigilant for such potential adverse interactions.

Probiotics. These agents produce neuroactive substances that act on the brain/gut axis. Preliminary evidence suggests that these “psychobiotics” confer mental health benefits.^10-12 Relative to other approaches, their low-risk profile make them an attractive option for some patients.

Anti-inflammatory/immune system therapies

Inflammation is linked to various medical and brain disorders. For example, patients with depression often demonstrate increased levels of peripheral blood inflammatory biomarkers (such as C-reactive protein and interleukin-6 and -17) that are known to alter norepinephrine, neuroendocrine (eg, the hypothalamic-pituitary-adrenal axis), and microglia function in addition to neuro­plasticity. Thus, targeting inflammation may facilitate the development of novel antidepressants. In addition, these agents may benefit depression associated with comorbid autoimmune disorders, such as psoriasis or rheumatoid arthritis. A systematic review and meta-analysis of 36 RCTs (N = 10,000) found 5 out of 6 anti-inflammatory agents improved depression.^13,14 In general, reported disadvantages of anti-inflammatories/immunosuppressants include the potential to block the antidepressant effect of some agents, the risk of opportunistic infections, and an increased risk of suicide.

Continue to: Statins

In a meta-analysis of 3 randomized, double-blind trials, 3 statins (lovastatin, atorvastatin, and simvastatin) significantly improved depression scores when used as an adjunctive therapy to fluoxetine and citalopram, compared with adjunctive placebo (N = 165, P < .001).¹⁵

Specific adverse effects of statins include headaches, muscle pain (rarely rhabdomyolysis), dizziness, rash, and liver damage. Statins also have the potential for adverse interactions with other medications. Given the limited efficacy literature on statins for depression and the potential for serious adverse effects, these agents probably should be limited to patients with treatment-resistant depression for whom a statin is indicated for a comorbid medical disorder, such as hypercholesteremia.

Neurosteroids

Brexanolone is FDA-approved for the treatment of postpartum depression. It is an IV formulation of the neuroactive steroid hormone allopregnanolone (a metabolite of progesterone), which acts as a positive allosteric modulator of the GABA-A receptor. Unfortunately, the infusion needs to occur over a 60-hour period.

Ganaxolone is an oral analog formulation of allopregnanolone. In an uncontrolled, open-label pilot study, this medication was administered for 8 weeks as an adjunct to an adequately dosed antidepressant to 10 postmenopausal women with persistent MDD.¹⁶ Of the 9 women who completed the study, 4 (44%) improved significantly (P < .019) and the benefit was sustained for 2 additional weeks.¹⁶ Adverse effects of ganaxolone included dizziness in 60% of participants, and sleepiness and fatigue in all of them with twice-daily dosing. If the FDA approves ganaxolone, it would become an easier-to-administer option to brexanolone.

Zuranolone is an investigational agent being studied as a treatment for postpartum depression. In a double-blind RCT that evaluated 151 women with postpartum depression, those who took oral zuranolone, 30 mg daily at bedtime for 2 weeks, experienced significant reductions in Hamilton Depression Rating Scale-17 (HDRS-17) scores compared with placebo (P < .003).¹⁷ Improvement in core depression symptom ratings was seen as early as Day 3 and persisted through Day 45.

Continue to: The most common...

The most common (≥5%) treatment-emergent adverse effects were somnolence (15%), headache (9%), dizziness (8%), upper respiratory tract infection (8%), diarrhea (6%), and sedation (5%). Two patients experienced a serious adverse event: one who received zuranolone (confusional state) and one who received placebo (pancreatitis). One patient discontinued zuranolone due to adverse effects vs no discontinuations among those who received placebo. The risk of taking zuranolone while breastfeeding is not known.

Device-based strategies

In addition to FDA-cleared approaches (eg, electroconvulsive therapy [ECT], vagus nerve stimulation [VNS], transcranial magnetic stimulation [TMS]), other devices have also demonstrated promising results.

Transcranial direct current stimulation (tDCS) involves delivering weak electrical current to the cerebral cortex through small scalp electrodes to produce the following effects:

• anodal tDCS enhances cortical excitability
• cathodal tDCS reduces cortical excitability.

A typical protocol consists of delivering 1 to 2 mA over 20 minutes with scalp electrodes placed in different configurations based on the targeted symptom(s).

While tDCS has been evaluated as a treatment for various neuropsychiatric disorders, including bipolar depression, Parkinson’s disease, and schizophrenia, most trials have looked at its use for treating depression. Results have been promising but mixed. For example, 1 meta-analysis of 6 RCTs (comprising 96 active and 80 sham tDCS courses) reported that active tDCS was superior to a sham procedure (Hedges’ g = 0.743) for symptoms of depression.¹⁸ By contrast, another meta-analysis of 6 RCTs (N = 200) did not find a significant difference between active and sham tDCS for response and remission rates.¹⁹ More recently, a group of experts created an evidence-based guideline using a systematic review of the controlled trial literature. These authors concluded there is “probable efficacy for anodal tDCS of the left dorsolateral prefrontal cortex (DLPFC) (with right orbitofrontal cathode) in major depressive episodes without drug resistance but probable inefficacy for drug-resistant major depressive episodes.”²⁰

Continue to: Adverse effects of tDCS...

Adverse effects of tDCS are typically mild but may include persistent skin lesions similar to burns; mania or hypomania; and one reported seizure in a pediatric patient.

Because various over-the-counter direct current stimulation devices are available for purchase at modest cost, clinicians should ask patients if they have been self-administering this treatment.

Chronotherapy strategies

Agomelatine combines serotonergic (5-HT2B and 5-HT2C antagonist) and melatonergic (MT1-MT2 agonist in the suprachiasmatic nucleus) actions that contribute to stabilization of circadian rhythms and subsequent improvement in sleep patterns. Agomelatine (n = 1,274) significantly lowered depression symptoms compared with placebo (n = 689) (standardized mean difference −0.26; P < 3.48×10^-11), but the clinical relevance was questionable.²¹ A recent review of the literature and expert opinion suggest this agent may also have efficacy for anhedonia; however, in placebo-controlled, relapse prevention studies, its long-term efficacy was not consistent.²²

Common adverse effects include anxiety; nausea, vomiting, and stomach pain; abnormal dreams and insomnia; dizziness; drowsiness and fatigue; and weight gain. Some reviewers have expressed concerns about agomelatine’s potential for hepatotoxicity and the need for repeated clinical laboratory tests. Although agomelatine is approved outside of the United States, limited efficacy data and the potential for serious adverse effects have precluded FDA approval of this agent.

Sleep deprivation as a treatment technique for depression has been developed over the past 50 years. With total sleep deprivation (TSD) over 1 cycle, patients stay awake for approximately 36 hours, from daytime until the next day’s evening. While 1 to 6 cycles can produce acute antidepressant effects, prompt relapse after sleep recovery is common.

Continue to: In a systematic review...

In a systematic review and meta-analysis of 7 studies that included a total of 311 patients with bipolar depression²³:

• TSD plus medications resulted in a significant decrease in depressive symptoms at 1 week compared with medications alone
• higher response rates were maintained after 3 months with lithium.

Adverse effects commonly include general fatigue and headaches; possible switch into mania with bipolar depression; and rarely, seizures or other unexpected medical conditions (eg, acute coronary syndrome). Presently, this approach is limited to research laboratories with the appropriate sophistication to safely conduct such trials.

Other nontraditional strategies

Cardiovascular exercise, resistance training, mindfulness, and yoga have been shown to decrease severe depressive symptoms when used as adjuncts for patients with treatment-resistant depression, or as monotherapy to treat patients with milder depression.

Exercise. The significant benefits of exercise in various forms as treatment for mild to moderate depression are well described in the literature, but it is less clear if it is effective for treatment-resistant depression. A 2013 Cochrane report²⁴ (39 studies with 2,326 participants total) and 2 meta-analyses undertaken in 2015 (Kvam et al²⁵ included 23 studies with 977 participants, and Schuh et al²⁶ included 25 trials with 1,487 participants) reported that various types of exercise ameliorate depression of differing subtypes and severity, with effect sizes ranging from small to large. Schuh et al²⁶ found that publication bias underestimated effect size. Also, not surprisingly, separate analysis of only higher-quality trials decreased effect size.^24-26 A meta-analysis that included tai chi and yoga in addition to aerobic exercise and strength training (25 trials with 2,083 participants) found low to moderate benefit for exercise and yoga.²⁷ Finally, a meta-analysis by Cramer et al²⁸ that included 12 RCTs (N = 619) supported the use of yoga plus controlled breathing techniques as an ancillary treatment for depression.

Two small exercise trials specifically evaluated patients with treatment-resistant depression.^29,30 Mota-Pereira et al²⁹ compared 22 participants who walked for 30 to 45 minutes, 5 days a week for 12 weeks in addition to pharmacotherapy with 11 patients who received pharmacotherapy only. Exercise improved all outcomes, including HDRS score (both compared to baseline and to the control group). Moreover, 26% of the exercise group went into remission. Pilu et al³⁰ evaluated strength training as an adjunctive treatment. Participants received 1 hour of strength training twice weekly for 8 months (n = 10), or pharmacotherapy only (n = 20). The adjunct strength training group had a statistically significant (P < .0001) improvement in HDRS scores at the end of the 8 months, whereas the control group did not (P < .28).

Continue to: Adverse effects...

Adverse effects of exercise are typically limited to sprains or strains; rarely, participants experience serious injuries.

Mindfulness-based interventions involve purposely paying attention in the present moment to enhance self-understanding and decrease anxiety about the future and regrets about the past, both of which complicate depression. A meta-analysis of 12 RCTs (N = 578) found this approach significantly reduced depression severity when used as an adjunctive therapy.³¹ There may be risks if mindfulness-based interventions are practiced incorrectly. For example, some reports have linked mindfulness-based interventions to psychotic episodes, meditation addiction, and antisocial or asocial behavior.³²

Bottom Line

Nonpharmacologic options for patients with treatment-resistant depression include herbal/nutraceuticals, anti-inflammatory/immune system therapies, and devices. While research suggests some of these approaches are promising, clinicians need to carefully consider potential adverse effects, some of which may be serious.

Related Resources

• Kaur M, Sanches M. Experimental therapeutics in treatmentresistant major depressive disorder. J Exp Pharmacol. 2021;13:181-196.
• Janicak PG. What’s new in transcranial magnetic stimulation. Current Psychiatry. 2019;18(3):10-16.

Drug Brand Names

Atorvastatin • Lipitor
Brexanolone • Zulresso
Citalopram • Celexa
Fluoxetine • Prozac
Lithium • Eskalith, Lithobid
Lovastatin • Altoprev, Mevacor
Minocycline • Dynacin, Minocin
Simvastatin • Flolipid, Zocor

When patients with major depressive disorder (MDD) do not achieve optimal outcomes after FDA-approved first-line treatments and standard adjunctive strategies, clinicians look for additional approaches to alleviate their patients’ symptoms. Recent research suggests that several “nontraditional” treatments used primarily as adjuncts to standard antidepressants have promise for treatment-resistant depression.

In Part 1 of this article (Current Psychiatry, September 2021), we examined off-label medications. In Part 2, we will review other nontraditional approaches to treatment-resistant depression, including herbal/nutraceutical agents, anti-inflammatory/immune system therapies, device-based treatments, and other alternative approaches. Importantly, some treatments also demonstrate adverse effects (Table^1-32). With a careful consideration of the risk/benefit balance, this article reviews some of the better-studied nontraditional treatment options for patients with treatment-resistant depression.

Herbal/nutraceutical agents

This category encompasses a variety of commonly available “natural” options patients often ask about and at times self-prescribe. Examples evaluated in clinical trials include:

• vitamin D
• essential fatty acids (omega-3, omega-6)
• S-adenosyl-L-methionine (SAMe)
• hypericum perforatum (St. John’s Wort)
• probiotics.

Vitamin D deficiency has been linked to depression, possibly by lowering serotonin, norepinephrine, and dopamine concentrations.^1-3

A meta-analysis of 3 prospective, observational studies (N = 8,815) found an elevated risk of affective disorders in patients with low vitamin D levels.⁴ In addition, a systematic review and meta-analysis supported a potential role for vitamin D supplementation for patients with treatment-resistant depresssion.⁵

Toxicity can occur at levels >100 ng/mL, and resulting adverse effects may include weakness, fatigue, sleepiness, headache, loss of appetite, dry mouth, metallic taste, nausea, and vomiting. This vitamin can be considered as an adjunct to standard antidepressants, particularly in patients with treatment-resistant depression who have low vitamin D levels, but regular monitoring is necessary to avoid toxicity.

Essential fatty acids. Protein receptors embedded in lipid membranes and their binding affinities are influenced by omega-3 and omega-6 polyunsaturated fatty acids. Thus, essential fatty acids may benefit depression by maintaining membrane integrity and fluidity, as well as via their anti-inflammatory activity.

Continue to: Although results from...

Although results from controlled trials are mixed, a systematic review and meta-analysis of adjunctive nutraceuticals supported a potential role for essential fatty acids, primarily eicosapentaenoic acid (EPA), by itself or in combination with docosahexaenoic acid (DHA), with total EPA >60%.⁵ A second meta-analysis of 26 studies (N = 2,160) that considered only essential fatty acids concluded that EPA ≥60% at ≤1 g/d could benefit depression.⁶ Furthermore, omega-3 fatty acids may be helpful as an add-on agent for postpartum depression.⁷

Be aware that a diet rich in omega-6 greatly increases oxidized low-density lipoprotein levels in adipose tissue, potentially posing a cardiac risk factor. Clinicians need to be aware that self-prescribed use of essential fatty acids is common, and to ask about and monitor their patients’ use of these agents.

S-adenosyl-L-methionine (SAMe) is an intracellular amino acid and methyl donor. Among other actions, it is involved in the biosynthesis of hormones and neurotransmitters. There is promising but limited preliminary evidence of its efficacy and safety as a monotherapy or for antidepressant augmentation.⁸ For example, when compared with placebo for depressive symptoms in 19 randomized controlled trials (RCTs) (N = 878) ⁸:

• Five out of 6 earlier controlled studies reported SAMe IV (200 to 400 mg/d) or IM (45 to 50 mg/d) was more effective than placebo
• When the above studies were added to 14 subsequent studies for a meta-analysis, 12 of 19 RCTs reported that parenteral or oral SAMe was significantly more effective than placebo for depression (P < .05).

Overall, the safety and tolerability of SAMe are good. Common adverse effects include nausea, mild insomnia, dizziness, irritability, and anxiety. This is another compound widely available without a prescription and at times self-prescribed. It carries an acceptable risk/benefit balance, with decades of experience.

Hypericum perforatum (St. John’s Wort) is widely prescribed for depression in China and Europe, typically in doses ranging from 500 to 900 mg/d. Its mechanism of action in depression may relate to inhibition of serotonin, dopamine, and norepinephrine uptake from the synaptic cleft of these interconnecting neurotransmitter systems.

Continue to: A meta-analysis of 7 clinical trials...

A meta-analysis of 7 clinical trials (N = 3,808) comparing St. John’s Wort with various selective serotonin reuptake inhibitors (SSRIs) reported comparable rates of response (pooled relative risk .983, 95% CI .924 to 1.042; P < .001) and remission (pooled relative risk 1.013, 95% CI .892 to 1.134; P < .001).⁹ Further, there were significantly lower discontinuation/dropout rates (pooled odds ratio .587, 95% CI .478 to 0.697; P < .001) for St. John’s Wort compared with the SSRIs.

Existing evidence on the long-term efficacy and safety is limited (studies ranged from 4 to 12 weeks), as is evidence for patients with more severe depression or high suicidality.

Serious drug interactions include the potential for serotonin syndrome when St. John’s Wort is combined with certain antidepressants, compromised efficacy of benzodiazepines and standard antidepressants, and severe skin reactions to sun exposure. In addition, St. John’s Wort may not be safe to use during pregnancy or while breastfeeding. Because potential drug interactions can be serious and individuals often self-prescribe this agent, it is important to ask patients about their use of St. John’s Wort, and to be vigilant for such potential adverse interactions.

Probiotics. These agents produce neuroactive substances that act on the brain/gut axis. Preliminary evidence suggests that these “psychobiotics” confer mental health benefits.^10-12 Relative to other approaches, their low-risk profile make them an attractive option for some patients.

Anti-inflammatory/immune system therapies

Inflammation is linked to various medical and brain disorders. For example, patients with depression often demonstrate increased levels of peripheral blood inflammatory biomarkers (such as C-reactive protein and interleukin-6 and -17) that are known to alter norepinephrine, neuroendocrine (eg, the hypothalamic-pituitary-adrenal axis), and microglia function in addition to neuro­plasticity. Thus, targeting inflammation may facilitate the development of novel antidepressants. In addition, these agents may benefit depression associated with comorbid autoimmune disorders, such as psoriasis or rheumatoid arthritis. A systematic review and meta-analysis of 36 RCTs (N = 10,000) found 5 out of 6 anti-inflammatory agents improved depression.^13,14 In general, reported disadvantages of anti-inflammatories/immunosuppressants include the potential to block the antidepressant effect of some agents, the risk of opportunistic infections, and an increased risk of suicide.

Continue to: Statins

In a meta-analysis of 3 randomized, double-blind trials, 3 statins (lovastatin, atorvastatin, and simvastatin) significantly improved depression scores when used as an adjunctive therapy to fluoxetine and citalopram, compared with adjunctive placebo (N = 165, P < .001).¹⁵

Specific adverse effects of statins include headaches, muscle pain (rarely rhabdomyolysis), dizziness, rash, and liver damage. Statins also have the potential for adverse interactions with other medications. Given the limited efficacy literature on statins for depression and the potential for serious adverse effects, these agents probably should be limited to patients with treatment-resistant depression for whom a statin is indicated for a comorbid medical disorder, such as hypercholesteremia.

Neurosteroids

Brexanolone is FDA-approved for the treatment of postpartum depression. It is an IV formulation of the neuroactive steroid hormone allopregnanolone (a metabolite of progesterone), which acts as a positive allosteric modulator of the GABA-A receptor. Unfortunately, the infusion needs to occur over a 60-hour period.

Ganaxolone is an oral analog formulation of allopregnanolone. In an uncontrolled, open-label pilot study, this medication was administered for 8 weeks as an adjunct to an adequately dosed antidepressant to 10 postmenopausal women with persistent MDD.¹⁶ Of the 9 women who completed the study, 4 (44%) improved significantly (P < .019) and the benefit was sustained for 2 additional weeks.¹⁶ Adverse effects of ganaxolone included dizziness in 60% of participants, and sleepiness and fatigue in all of them with twice-daily dosing. If the FDA approves ganaxolone, it would become an easier-to-administer option to brexanolone.

Zuranolone is an investigational agent being studied as a treatment for postpartum depression. In a double-blind RCT that evaluated 151 women with postpartum depression, those who took oral zuranolone, 30 mg daily at bedtime for 2 weeks, experienced significant reductions in Hamilton Depression Rating Scale-17 (HDRS-17) scores compared with placebo (P < .003).¹⁷ Improvement in core depression symptom ratings was seen as early as Day 3 and persisted through Day 45.

Continue to: The most common...

The most common (≥5%) treatment-emergent adverse effects were somnolence (15%), headache (9%), dizziness (8%), upper respiratory tract infection (8%), diarrhea (6%), and sedation (5%). Two patients experienced a serious adverse event: one who received zuranolone (confusional state) and one who received placebo (pancreatitis). One patient discontinued zuranolone due to adverse effects vs no discontinuations among those who received placebo. The risk of taking zuranolone while breastfeeding is not known.

Device-based strategies

In addition to FDA-cleared approaches (eg, electroconvulsive therapy [ECT], vagus nerve stimulation [VNS], transcranial magnetic stimulation [TMS]), other devices have also demonstrated promising results.

Transcranial direct current stimulation (tDCS) involves delivering weak electrical current to the cerebral cortex through small scalp electrodes to produce the following effects:

• anodal tDCS enhances cortical excitability
• cathodal tDCS reduces cortical excitability.

A typical protocol consists of delivering 1 to 2 mA over 20 minutes with scalp electrodes placed in different configurations based on the targeted symptom(s).

While tDCS has been evaluated as a treatment for various neuropsychiatric disorders, including bipolar depression, Parkinson’s disease, and schizophrenia, most trials have looked at its use for treating depression. Results have been promising but mixed. For example, 1 meta-analysis of 6 RCTs (comprising 96 active and 80 sham tDCS courses) reported that active tDCS was superior to a sham procedure (Hedges’ g = 0.743) for symptoms of depression.¹⁸ By contrast, another meta-analysis of 6 RCTs (N = 200) did not find a significant difference between active and sham tDCS for response and remission rates.¹⁹ More recently, a group of experts created an evidence-based guideline using a systematic review of the controlled trial literature. These authors concluded there is “probable efficacy for anodal tDCS of the left dorsolateral prefrontal cortex (DLPFC) (with right orbitofrontal cathode) in major depressive episodes without drug resistance but probable inefficacy for drug-resistant major depressive episodes.”²⁰

Continue to: Adverse effects of tDCS...

Adverse effects of tDCS are typically mild but may include persistent skin lesions similar to burns; mania or hypomania; and one reported seizure in a pediatric patient.

Because various over-the-counter direct current stimulation devices are available for purchase at modest cost, clinicians should ask patients if they have been self-administering this treatment.

Chronotherapy strategies

Agomelatine combines serotonergic (5-HT2B and 5-HT2C antagonist) and melatonergic (MT1-MT2 agonist in the suprachiasmatic nucleus) actions that contribute to stabilization of circadian rhythms and subsequent improvement in sleep patterns. Agomelatine (n = 1,274) significantly lowered depression symptoms compared with placebo (n = 689) (standardized mean difference −0.26; P < 3.48×10^-11), but the clinical relevance was questionable.²¹ A recent review of the literature and expert opinion suggest this agent may also have efficacy for anhedonia; however, in placebo-controlled, relapse prevention studies, its long-term efficacy was not consistent.²²

Common adverse effects include anxiety; nausea, vomiting, and stomach pain; abnormal dreams and insomnia; dizziness; drowsiness and fatigue; and weight gain. Some reviewers have expressed concerns about agomelatine’s potential for hepatotoxicity and the need for repeated clinical laboratory tests. Although agomelatine is approved outside of the United States, limited efficacy data and the potential for serious adverse effects have precluded FDA approval of this agent.

Sleep deprivation as a treatment technique for depression has been developed over the past 50 years. With total sleep deprivation (TSD) over 1 cycle, patients stay awake for approximately 36 hours, from daytime until the next day’s evening. While 1 to 6 cycles can produce acute antidepressant effects, prompt relapse after sleep recovery is common.

Continue to: In a systematic review...

In a systematic review and meta-analysis of 7 studies that included a total of 311 patients with bipolar depression²³:

• TSD plus medications resulted in a significant decrease in depressive symptoms at 1 week compared with medications alone
• higher response rates were maintained after 3 months with lithium.

Adverse effects commonly include general fatigue and headaches; possible switch into mania with bipolar depression; and rarely, seizures or other unexpected medical conditions (eg, acute coronary syndrome). Presently, this approach is limited to research laboratories with the appropriate sophistication to safely conduct such trials.

Other nontraditional strategies

Cardiovascular exercise, resistance training, mindfulness, and yoga have been shown to decrease severe depressive symptoms when used as adjuncts for patients with treatment-resistant depression, or as monotherapy to treat patients with milder depression.

Exercise. The significant benefits of exercise in various forms as treatment for mild to moderate depression are well described in the literature, but it is less clear if it is effective for treatment-resistant depression. A 2013 Cochrane report²⁴ (39 studies with 2,326 participants total) and 2 meta-analyses undertaken in 2015 (Kvam et al²⁵ included 23 studies with 977 participants, and Schuh et al²⁶ included 25 trials with 1,487 participants) reported that various types of exercise ameliorate depression of differing subtypes and severity, with effect sizes ranging from small to large. Schuh et al²⁶ found that publication bias underestimated effect size. Also, not surprisingly, separate analysis of only higher-quality trials decreased effect size.^24-26 A meta-analysis that included tai chi and yoga in addition to aerobic exercise and strength training (25 trials with 2,083 participants) found low to moderate benefit for exercise and yoga.²⁷ Finally, a meta-analysis by Cramer et al²⁸ that included 12 RCTs (N = 619) supported the use of yoga plus controlled breathing techniques as an ancillary treatment for depression.

Two small exercise trials specifically evaluated patients with treatment-resistant depression.^29,30 Mota-Pereira et al²⁹ compared 22 participants who walked for 30 to 45 minutes, 5 days a week for 12 weeks in addition to pharmacotherapy with 11 patients who received pharmacotherapy only. Exercise improved all outcomes, including HDRS score (both compared to baseline and to the control group). Moreover, 26% of the exercise group went into remission. Pilu et al³⁰ evaluated strength training as an adjunctive treatment. Participants received 1 hour of strength training twice weekly for 8 months (n = 10), or pharmacotherapy only (n = 20). The adjunct strength training group had a statistically significant (P < .0001) improvement in HDRS scores at the end of the 8 months, whereas the control group did not (P < .28).

Continue to: Adverse effects...

Adverse effects of exercise are typically limited to sprains or strains; rarely, participants experience serious injuries.

Mindfulness-based interventions involve purposely paying attention in the present moment to enhance self-understanding and decrease anxiety about the future and regrets about the past, both of which complicate depression. A meta-analysis of 12 RCTs (N = 578) found this approach significantly reduced depression severity when used as an adjunctive therapy.³¹ There may be risks if mindfulness-based interventions are practiced incorrectly. For example, some reports have linked mindfulness-based interventions to psychotic episodes, meditation addiction, and antisocial or asocial behavior.³²

Bottom Line

Nonpharmacologic options for patients with treatment-resistant depression include herbal/nutraceuticals, anti-inflammatory/immune system therapies, and devices. While research suggests some of these approaches are promising, clinicians need to carefully consider potential adverse effects, some of which may be serious.

Related Resources

• Kaur M, Sanches M. Experimental therapeutics in treatmentresistant major depressive disorder. J Exp Pharmacol. 2021;13:181-196.
• Janicak PG. What’s new in transcranial magnetic stimulation. Current Psychiatry. 2019;18(3):10-16.

Drug Brand Names

Atorvastatin • Lipitor
Brexanolone • Zulresso
Citalopram • Celexa
Fluoxetine • Prozac
Lithium • Eskalith, Lithobid
Lovastatin • Altoprev, Mevacor
Minocycline • Dynacin, Minocin
Simvastatin • Flolipid, Zocor

When patients with major depressive disorder (MDD) do not achieve optimal outcomes after FDA-approved first-line treatments and standard adjunctive strategies, clinicians look for additional approaches to alleviate their patients’ symptoms. Recent research suggests that several “nontraditional” treatments used primarily as adjuncts to standard antidepressants have promise for treatment-resistant depression.

In Part 1 of this article (Current Psychiatry, September 2021), we examined off-label medications. In Part 2, we will review other nontraditional approaches to treatment-resistant depression, including herbal/nutraceutical agents, anti-inflammatory/immune system therapies, device-based treatments, and other alternative approaches. Importantly, some treatments also demonstrate adverse effects (Table^1-32). With a careful consideration of the risk/benefit balance, this article reviews some of the better-studied nontraditional treatment options for patients with treatment-resistant depression.

Herbal/nutraceutical agents

This category encompasses a variety of commonly available “natural” options patients often ask about and at times self-prescribe. Examples evaluated in clinical trials include:

• vitamin D
• essential fatty acids (omega-3, omega-6)
• S-adenosyl-L-methionine (SAMe)
• hypericum perforatum (St. John’s Wort)
• probiotics.

Vitamin D deficiency has been linked to depression, possibly by lowering serotonin, norepinephrine, and dopamine concentrations.^1-3

A meta-analysis of 3 prospective, observational studies (N = 8,815) found an elevated risk of affective disorders in patients with low vitamin D levels.⁴ In addition, a systematic review and meta-analysis supported a potential role for vitamin D supplementation for patients with treatment-resistant depresssion.⁵

Toxicity can occur at levels >100 ng/mL, and resulting adverse effects may include weakness, fatigue, sleepiness, headache, loss of appetite, dry mouth, metallic taste, nausea, and vomiting. This vitamin can be considered as an adjunct to standard antidepressants, particularly in patients with treatment-resistant depression who have low vitamin D levels, but regular monitoring is necessary to avoid toxicity.

Essential fatty acids. Protein receptors embedded in lipid membranes and their binding affinities are influenced by omega-3 and omega-6 polyunsaturated fatty acids. Thus, essential fatty acids may benefit depression by maintaining membrane integrity and fluidity, as well as via their anti-inflammatory activity.

Continue to: Although results from...

Although results from controlled trials are mixed, a systematic review and meta-analysis of adjunctive nutraceuticals supported a potential role for essential fatty acids, primarily eicosapentaenoic acid (EPA), by itself or in combination with docosahexaenoic acid (DHA), with total EPA >60%.⁵ A second meta-analysis of 26 studies (N = 2,160) that considered only essential fatty acids concluded that EPA ≥60% at ≤1 g/d could benefit depression.⁶ Furthermore, omega-3 fatty acids may be helpful as an add-on agent for postpartum depression.⁷

Be aware that a diet rich in omega-6 greatly increases oxidized low-density lipoprotein levels in adipose tissue, potentially posing a cardiac risk factor. Clinicians need to be aware that self-prescribed use of essential fatty acids is common, and to ask about and monitor their patients’ use of these agents.

S-adenosyl-L-methionine (SAMe) is an intracellular amino acid and methyl donor. Among other actions, it is involved in the biosynthesis of hormones and neurotransmitters. There is promising but limited preliminary evidence of its efficacy and safety as a monotherapy or for antidepressant augmentation.⁸ For example, when compared with placebo for depressive symptoms in 19 randomized controlled trials (RCTs) (N = 878) ⁸:

• Five out of 6 earlier controlled studies reported SAMe IV (200 to 400 mg/d) or IM (45 to 50 mg/d) was more effective than placebo
• When the above studies were added to 14 subsequent studies for a meta-analysis, 12 of 19 RCTs reported that parenteral or oral SAMe was significantly more effective than placebo for depression (P < .05).

Overall, the safety and tolerability of SAMe are good. Common adverse effects include nausea, mild insomnia, dizziness, irritability, and anxiety. This is another compound widely available without a prescription and at times self-prescribed. It carries an acceptable risk/benefit balance, with decades of experience.

Hypericum perforatum (St. John’s Wort) is widely prescribed for depression in China and Europe, typically in doses ranging from 500 to 900 mg/d. Its mechanism of action in depression may relate to inhibition of serotonin, dopamine, and norepinephrine uptake from the synaptic cleft of these interconnecting neurotransmitter systems.

Continue to: A meta-analysis of 7 clinical trials...

A meta-analysis of 7 clinical trials (N = 3,808) comparing St. John’s Wort with various selective serotonin reuptake inhibitors (SSRIs) reported comparable rates of response (pooled relative risk .983, 95% CI .924 to 1.042; P < .001) and remission (pooled relative risk 1.013, 95% CI .892 to 1.134; P < .001).⁹ Further, there were significantly lower discontinuation/dropout rates (pooled odds ratio .587, 95% CI .478 to 0.697; P < .001) for St. John’s Wort compared with the SSRIs.

Existing evidence on the long-term efficacy and safety is limited (studies ranged from 4 to 12 weeks), as is evidence for patients with more severe depression or high suicidality.

Serious drug interactions include the potential for serotonin syndrome when St. John’s Wort is combined with certain antidepressants, compromised efficacy of benzodiazepines and standard antidepressants, and severe skin reactions to sun exposure. In addition, St. John’s Wort may not be safe to use during pregnancy or while breastfeeding. Because potential drug interactions can be serious and individuals often self-prescribe this agent, it is important to ask patients about their use of St. John’s Wort, and to be vigilant for such potential adverse interactions.

Probiotics. These agents produce neuroactive substances that act on the brain/gut axis. Preliminary evidence suggests that these “psychobiotics” confer mental health benefits.^10-12 Relative to other approaches, their low-risk profile make them an attractive option for some patients.

Anti-inflammatory/immune system therapies

Inflammation is linked to various medical and brain disorders. For example, patients with depression often demonstrate increased levels of peripheral blood inflammatory biomarkers (such as C-reactive protein and interleukin-6 and -17) that are known to alter norepinephrine, neuroendocrine (eg, the hypothalamic-pituitary-adrenal axis), and microglia function in addition to neuro­plasticity. Thus, targeting inflammation may facilitate the development of novel antidepressants. In addition, these agents may benefit depression associated with comorbid autoimmune disorders, such as psoriasis or rheumatoid arthritis. A systematic review and meta-analysis of 36 RCTs (N = 10,000) found 5 out of 6 anti-inflammatory agents improved depression.^13,14 In general, reported disadvantages of anti-inflammatories/immunosuppressants include the potential to block the antidepressant effect of some agents, the risk of opportunistic infections, and an increased risk of suicide.

Continue to: Statins

In a meta-analysis of 3 randomized, double-blind trials, 3 statins (lovastatin, atorvastatin, and simvastatin) significantly improved depression scores when used as an adjunctive therapy to fluoxetine and citalopram, compared with adjunctive placebo (N = 165, P < .001).¹⁵

Specific adverse effects of statins include headaches, muscle pain (rarely rhabdomyolysis), dizziness, rash, and liver damage. Statins also have the potential for adverse interactions with other medications. Given the limited efficacy literature on statins for depression and the potential for serious adverse effects, these agents probably should be limited to patients with treatment-resistant depression for whom a statin is indicated for a comorbid medical disorder, such as hypercholesteremia.

Neurosteroids

Brexanolone is FDA-approved for the treatment of postpartum depression. It is an IV formulation of the neuroactive steroid hormone allopregnanolone (a metabolite of progesterone), which acts as a positive allosteric modulator of the GABA-A receptor. Unfortunately, the infusion needs to occur over a 60-hour period.

Ganaxolone is an oral analog formulation of allopregnanolone. In an uncontrolled, open-label pilot study, this medication was administered for 8 weeks as an adjunct to an adequately dosed antidepressant to 10 postmenopausal women with persistent MDD.¹⁶ Of the 9 women who completed the study, 4 (44%) improved significantly (P < .019) and the benefit was sustained for 2 additional weeks.¹⁶ Adverse effects of ganaxolone included dizziness in 60% of participants, and sleepiness and fatigue in all of them with twice-daily dosing. If the FDA approves ganaxolone, it would become an easier-to-administer option to brexanolone.

Zuranolone is an investigational agent being studied as a treatment for postpartum depression. In a double-blind RCT that evaluated 151 women with postpartum depression, those who took oral zuranolone, 30 mg daily at bedtime for 2 weeks, experienced significant reductions in Hamilton Depression Rating Scale-17 (HDRS-17) scores compared with placebo (P < .003).¹⁷ Improvement in core depression symptom ratings was seen as early as Day 3 and persisted through Day 45.

Continue to: The most common...

The most common (≥5%) treatment-emergent adverse effects were somnolence (15%), headache (9%), dizziness (8%), upper respiratory tract infection (8%), diarrhea (6%), and sedation (5%). Two patients experienced a serious adverse event: one who received zuranolone (confusional state) and one who received placebo (pancreatitis). One patient discontinued zuranolone due to adverse effects vs no discontinuations among those who received placebo. The risk of taking zuranolone while breastfeeding is not known.

Device-based strategies

In addition to FDA-cleared approaches (eg, electroconvulsive therapy [ECT], vagus nerve stimulation [VNS], transcranial magnetic stimulation [TMS]), other devices have also demonstrated promising results.

Transcranial direct current stimulation (tDCS) involves delivering weak electrical current to the cerebral cortex through small scalp electrodes to produce the following effects:

• anodal tDCS enhances cortical excitability
• cathodal tDCS reduces cortical excitability.

A typical protocol consists of delivering 1 to 2 mA over 20 minutes with scalp electrodes placed in different configurations based on the targeted symptom(s).

While tDCS has been evaluated as a treatment for various neuropsychiatric disorders, including bipolar depression, Parkinson’s disease, and schizophrenia, most trials have looked at its use for treating depression. Results have been promising but mixed. For example, 1 meta-analysis of 6 RCTs (comprising 96 active and 80 sham tDCS courses) reported that active tDCS was superior to a sham procedure (Hedges’ g = 0.743) for symptoms of depression.¹⁸ By contrast, another meta-analysis of 6 RCTs (N = 200) did not find a significant difference between active and sham tDCS for response and remission rates.¹⁹ More recently, a group of experts created an evidence-based guideline using a systematic review of the controlled trial literature. These authors concluded there is “probable efficacy for anodal tDCS of the left dorsolateral prefrontal cortex (DLPFC) (with right orbitofrontal cathode) in major depressive episodes without drug resistance but probable inefficacy for drug-resistant major depressive episodes.”²⁰

Continue to: Adverse effects of tDCS...

Adverse effects of tDCS are typically mild but may include persistent skin lesions similar to burns; mania or hypomania; and one reported seizure in a pediatric patient.

Because various over-the-counter direct current stimulation devices are available for purchase at modest cost, clinicians should ask patients if they have been self-administering this treatment.

Chronotherapy strategies

Agomelatine combines serotonergic (5-HT2B and 5-HT2C antagonist) and melatonergic (MT1-MT2 agonist in the suprachiasmatic nucleus) actions that contribute to stabilization of circadian rhythms and subsequent improvement in sleep patterns. Agomelatine (n = 1,274) significantly lowered depression symptoms compared with placebo (n = 689) (standardized mean difference −0.26; P < 3.48×10^-11), but the clinical relevance was questionable.²¹ A recent review of the literature and expert opinion suggest this agent may also have efficacy for anhedonia; however, in placebo-controlled, relapse prevention studies, its long-term efficacy was not consistent.²²

Common adverse effects include anxiety; nausea, vomiting, and stomach pain; abnormal dreams and insomnia; dizziness; drowsiness and fatigue; and weight gain. Some reviewers have expressed concerns about agomelatine’s potential for hepatotoxicity and the need for repeated clinical laboratory tests. Although agomelatine is approved outside of the United States, limited efficacy data and the potential for serious adverse effects have precluded FDA approval of this agent.

Sleep deprivation as a treatment technique for depression has been developed over the past 50 years. With total sleep deprivation (TSD) over 1 cycle, patients stay awake for approximately 36 hours, from daytime until the next day’s evening. While 1 to 6 cycles can produce acute antidepressant effects, prompt relapse after sleep recovery is common.

Continue to: In a systematic review...

In a systematic review and meta-analysis of 7 studies that included a total of 311 patients with bipolar depression²³:

• TSD plus medications resulted in a significant decrease in depressive symptoms at 1 week compared with medications alone
• higher response rates were maintained after 3 months with lithium.

Adverse effects commonly include general fatigue and headaches; possible switch into mania with bipolar depression; and rarely, seizures or other unexpected medical conditions (eg, acute coronary syndrome). Presently, this approach is limited to research laboratories with the appropriate sophistication to safely conduct such trials.

Other nontraditional strategies

Cardiovascular exercise, resistance training, mindfulness, and yoga have been shown to decrease severe depressive symptoms when used as adjuncts for patients with treatment-resistant depression, or as monotherapy to treat patients with milder depression.

Exercise. The significant benefits of exercise in various forms as treatment for mild to moderate depression are well described in the literature, but it is less clear if it is effective for treatment-resistant depression. A 2013 Cochrane report²⁴ (39 studies with 2,326 participants total) and 2 meta-analyses undertaken in 2015 (Kvam et al²⁵ included 23 studies with 977 participants, and Schuh et al²⁶ included 25 trials with 1,487 participants) reported that various types of exercise ameliorate depression of differing subtypes and severity, with effect sizes ranging from small to large. Schuh et al²⁶ found that publication bias underestimated effect size. Also, not surprisingly, separate analysis of only higher-quality trials decreased effect size.^24-26 A meta-analysis that included tai chi and yoga in addition to aerobic exercise and strength training (25 trials with 2,083 participants) found low to moderate benefit for exercise and yoga.²⁷ Finally, a meta-analysis by Cramer et al²⁸ that included 12 RCTs (N = 619) supported the use of yoga plus controlled breathing techniques as an ancillary treatment for depression.

Two small exercise trials specifically evaluated patients with treatment-resistant depression.^29,30 Mota-Pereira et al²⁹ compared 22 participants who walked for 30 to 45 minutes, 5 days a week for 12 weeks in addition to pharmacotherapy with 11 patients who received pharmacotherapy only. Exercise improved all outcomes, including HDRS score (both compared to baseline and to the control group). Moreover, 26% of the exercise group went into remission. Pilu et al³⁰ evaluated strength training as an adjunctive treatment. Participants received 1 hour of strength training twice weekly for 8 months (n = 10), or pharmacotherapy only (n = 20). The adjunct strength training group had a statistically significant (P < .0001) improvement in HDRS scores at the end of the 8 months, whereas the control group did not (P < .28).

Continue to: Adverse effects...

Adverse effects of exercise are typically limited to sprains or strains; rarely, participants experience serious injuries.

Mindfulness-based interventions involve purposely paying attention in the present moment to enhance self-understanding and decrease anxiety about the future and regrets about the past, both of which complicate depression. A meta-analysis of 12 RCTs (N = 578) found this approach significantly reduced depression severity when used as an adjunctive therapy.³¹ There may be risks if mindfulness-based interventions are practiced incorrectly. For example, some reports have linked mindfulness-based interventions to psychotic episodes, meditation addiction, and antisocial or asocial behavior.³²

Bottom Line

Nonpharmacologic options for patients with treatment-resistant depression include herbal/nutraceuticals, anti-inflammatory/immune system therapies, and devices. While research suggests some of these approaches are promising, clinicians need to carefully consider potential adverse effects, some of which may be serious.

Related Resources

• Kaur M, Sanches M. Experimental therapeutics in treatmentresistant major depressive disorder. J Exp Pharmacol. 2021;13:181-196.
• Janicak PG. What’s new in transcranial magnetic stimulation. Current Psychiatry. 2019;18(3):10-16.

Drug Brand Names

Atorvastatin • Lipitor
Brexanolone • Zulresso
Citalopram • Celexa
Fluoxetine • Prozac
Lithium • Eskalith, Lithobid
Lovastatin • Altoprev, Mevacor
Minocycline • Dynacin, Minocin
Simvastatin • Flolipid, Zocor

Presently, FDA-approved first-line treatments and standard adjunctive strategies (eg, lithium, thyroid supplementation, stimulants, second-generation antipsychotics) for major depressive disorder (MDD) often produce less-than-desired outcomes while carrying a potentially substantial safety and tolerability burden. The lack of clinically useful and individual-based biomarkers (eg, genetic, neurophysiological, imaging) is a major obstacle to enhancing treatment efficacy and/or decreasing associated adverse effects (AEs). While the discovery of such tools is being aggressively pursued and ultimately will facilitate a more precision-based choice of therapy, empirical strategies remain our primary approach.

In controlled trials, several nontraditional treatments used primarily as adjuncts to standard antidepressants have shown promise. These include “repurposed” (off-label) medications, herbal/nutraceuticals, anti-inflammatory/immune system therapies, device-based treatments, and other alternative approaches.

Importantly, some nontraditional treatments also demonstrate AEs (Table^1-16). With a careful consideration of the risk/benefit balance, this article reviews some of the better-studied treatment options for patients with treatment-resistant depression (TRD). In Part 1, we will examine off-label medications. In Part 2, we will review other nontraditional approaches to TRD, including herbal/nutraceuticals, anti-inflammatory/immune system therapies, device-based treatments, and other alternative approaches.

We believe this review will help clinicians who need to formulate a different approach after their patient with depression is not helped by traditional first-, second-, and third-line treatments. The potential options discussed in Part 1 of this article are categorized based on their putative mechanism of action (MOA) for depression.

Serotonergic and noradrenergic strategies

Pimavanserin is FDA-approved for treatment of Parkinson’s psychosis. Its potential MOA as an adjunctive strategy for MDD may involve 5-HT2A antagonist and inverse agonist receptor activity, as well as lesser effects at the 5-HT2Creceptor.

A 2-stage, 5-week randomized controlled trial (RCT) (CLARITY; N = 207) found adjunctive pimavanserin (34 mg/d) produced a robust antidepressant effect vs placebo in patients whose depression did not respond to selective serotonin reuptake inhibitors (SSRIs) or serotonin-norepinephrine reuptake inhibitors (SNRIs).¹ Furthermore, a secondary analysis of the data suggested that pimavanserin also improved sleepiness (P < .0003) and daily functioning (P < .014) at Week 5.²

Unfortunately, two 6-week, Phase III RCTs (CLARITY-2 and -3; N = 298) did not find a statistically significant difference between active treatment and placebo. This was based on change in the primary outcome measure (Hamilton Depression Rating Scale-17 score) when adjunctive pimavanserin (34 mg/d) was added to an SSRI or SNRI in patients with TRD.³ There was, however, a significant difference favoring active treatment over placebo based on the Clinical Global Impression–Severity score.

Continue to: In these trials...

In these trials, pimavanserin was generally well-tolerated. The most common AEs were dry mouth, nausea, and headache. Pimavanserin has minimal activity at norepinephrine, dopamine, histamine, or acetylcholine receptors, thus avoiding AEs associated with these receptor interactions.

Given the mixed efficacy results of existing trials, further studies are needed to clarify this agent’s overall risk/benefit in the context of TRD.

Antihypertensive medications

Emerging data suggest that some beta-adrenergic blockers, angiotensin-inhibiting agents, and calcium antagonists are associated with a decreased incidence of depression. A large 2020 study (N = 3,747,190) used population-based Danish registries (2005 to 2015) to evaluate if any of the 41 most commonly prescribed antihypertensive medications were associated with the diagnosis of depressive disorder or use of antidepressants.⁴ These researchers found that enalapril, ramipril, amlodipine, propranolol, atenolol, bisoprolol, carvedilol (P < .001), and verapamil (P < .004) were strongly associated with a decreased risk of depression.⁴

Adverse effects across these different classes of antihypertensives are well characterized, can be substantial, and commonly are related to their impact on cardiovascular function (eg, hypotension). Clinically, these agents may be potential adjuncts for patients with TRD who need antihypertensive therapy. Their use and the choice of specific agent should only be determined in consultation with the patient’s primary care physician (PCP) or appropriate specialist.

Glutamatergic strategies

Ketamine is a dissociative anesthetic and analgesic. Its MOA for treating depression appears to occur primarily through antagonist activity at the N-methyl-D-aspartate ionotropic receptor of the glutamatergic system. There is preliminary evidence that its opioid receptor actions also may contribute to its antidepressant effect.⁵

Continue to: Many published studies...

Many published studies and reviews have described ketamine’s role for treating MDD. Several studies have reported that low-dose (0.5 mg/kg) IV ketamine infusions can rapidly attenuate severe episodes of MDD as well as associated suicidality. For example, a meta-analysis of 9 RCTs (N = 368) comparing ketamine to placebo for acute treatment of unipolar and bipolar depression reported superior therapeutic effects with active treatment at 24 hours, 72 hours, and 7 days.⁶ The response and remission rates for ketamine were 52% and 21% at 24 hours; 48% and 24% at 72 hours; and 40% and 26% at 7 days, respectively.⁶

The most commonly reported AEs during the 4 hours after ketamine infusion included⁷:

• drowsiness, dizziness, poor coordination
• blurred vision, feeling strange or unreal
• hemodynamic changes (approximately 33%)
• small but significant (P < .05) increases in psychotomimetic and dissociative symptoms.

Because some individuals use ketamine recreationally, this agent also carries the risk of abuse.

Research is ongoing on strategies for long-term maintenance ketamine treatment, and the results of both short- and long-term trials will require careful scrutiny to better assess this agent’s safety and tolerability. Clinicians should first consider esketamine—the S-enantiomer of ketamine—because an intranasal formulation of this agent is FDA-approved for treating patients with TRD or MDD with suicidality when administered in a Risk Evaluation and Mitigation Strategy–certified setting.

Cholinergic strategies

Scopolamine is a potent muscarinic receptor antagonist used to prevent nausea and vomiting caused by motion sickness or medications used during surgery. Its use for MDD is based on the theory that muscarinic receptors may be hypersensitive in mood disorders.

Continue to: Several double-blind RCTs...

Several double-blind RCTs of patients with unipolar or bipolar depression that used 3 pulsed IV infusions (4.0 mcg/kg) over 15 minutes found a rapid, robust antidepressant effect with scopolamine vs placebo.^8,9 The oral formulation might also be effective, but would not have a rapid onset.

Common adverse effects of scopolamine include agitation, dry mouth, urinary retention, and cognitive clouding. Given scopolamine’s substantial AE profile, it should be considered only for patients with TRD who could also benefit from the oral formulation for the medical indications noted above, should generally be avoided in older patients, and should be prescribed in consultation with the patient’s PCP.

Botulinum toxin. This neurotoxin inhibits acetylcholine release. It is used to treat disorders characterized by abnormal muscular contraction, such as strabismus, blepharospasm, and chronic pain syndromes. Its MOA for depression may involve its paralytic effects after injection into the glabella forehead muscle (based on the facial feedback hypothesis), as well as modulation of neurotransmitters implicated in the pathophysiology of depression.

In several small trials, injectable botulinum toxin type A (BTA) (29 units) demonstrated antidepressant effects. A recent review that considered 6 trials (N = 235; 4 of the 6 studies were RCTs, 3 of which were rated as high quality) concluded that BTA may be a promising treatment for MDD.¹⁰ Limitations of this review included lack of a priori hypotheses, small sample sizes, gender bias, and difficulty in blinding.

In clinical trials, the most common AEs included local irritation at the injection site and transient headache. This agent’s relatively mild AE profile and possible overlap when used for some of the medical indications noted above opens its potential use as an adjunct in patients with comorbid TRD.

Continue to: Endocrine strategies

Endocrine strategies

Mifepristone (RU486). This anti-glucocorticoid receptor antagonist is used as an abortifacient. Based on the theory that hyperactivity of the hypothalamic-pituitary-adrenal axis is implicated in the pathophysiology of MDD with psychotic features (psychotic depression), this agent has been studied as a treatment for this indication.

An analysis of 5 double-blind RCTs (N = 1,460) found that 7 days of mifepristone, 1,200 mg/d, was superior to placebo (P < .004) in reducing psychotic symptoms of depression.¹¹ Plasma concentrations ≥1,600 ng/mL may be required to maximize benefit.¹¹

Overall, this agent demonstrated a good safety profile in clinical trials, with treatment-emergent AEs reported in 556 (66.7%) patients who received mifepristone vs 386 (61.6%) patients who received placebo.¹¹ Common AEs included gastrointestinal (GI) symptoms, headache, and dizziness. However, 3 deaths occurred: 2 patients who received mifepristone and 1 patient who received placebo. Given this potential for a fatal outcome, clinicians should first consider prescribing an adjunctive antipsychotic agent or electroconvulsive therapy.

Estrogens. These hormones are important for sexual and reproductive development and are used to treat various sexual/reproductive disorders, primarily in women. Their role in treating depression is based on the observation that perimenopause is accompanied by an increased risk of new and recurrent depression coincident with declining ovarian function.

Evidence supports the antidepressant efficacy of transdermal estradiol plus progesterone for perimenopausal depression, but not for postmenopausal depression.^12-14 However, estrogens carry significant risks that must be carefully considered in relationship to their potential benefits. These risks include:

• vaginal bleeding, dysmenorrhea
• fibroid enlargement
• galactorrhea
• ovarian cancer, endometrial cancer, breast cancer
• deep vein thrombosis, pulmonary embolism
• hypertension, chest pain, myocardial infarction, stroke.

Continue to: The use of estrogens...

The use of estrogens as an adjunctive therapy for women with treatment-resistant perimenopausal depression should only be undertaken when standard strategies have failed, and in consultation with an endocrine specialist who can monitor for potentially serious AEs.

Opioid medications

Buprenorphine is used to treat opioid use disorder (OUD) as well as acute and chronic pain. The opioid system is involved in the regulation of mood and may be an appropriate target for novel antidepressants. The use of buprenorphine in combination with samidorphan (a preferential mu-opioid receptor antagonist) has shown initial promise for TRD while minimizing abuse potential.

Although earlier results were mixed, a pooled analysis of 2 recent large RCTs (N = 760) of patients with MDD who had not responded to antidepressants reported greater reduction in Montgomery-Åsberg Depression Rating Scale scores from baseline for active treatment (buprenorphine/samidorphan; 2 mg/2 mg) vs placebo at multiple timepoints, including end of treatment (-1.8; P < .010).¹⁵

The most common AEs included nausea, constipation, dizziness, vomiting, somnolence, fatigue, and sedation. There was minimal evidence of abuse, dependence, or opioid withdrawal. Due to the opioid crisis in the United States, the resulting relaxation of regulations regarding prescribing buprenorphine, and the high rates of depression among patients with OUD, buprenorphine/samidorphan, which is an investigational agent that is not FDA-approved, may be particularly helpful for patients with OUD who also experience comorbid TRD.

Antioxidant agents

N-acetylcysteine (NAC) is an amino acid that can treat acetaminophen toxicity and moderate hepatic damage by increasing glutathione levels. Glutathione is also the primary antioxidant in the CNS. NAC may protect against oxidative stress, chelate heavy metals, reduce inflammation, protect against mitochondrial dysfunction, inhibit apoptosis, and enhance neurogenesis, all potential pathophysiological processes that may contribute to depression.¹⁶

Continue to: A systematic review...

A systematic review and meta-analysis of 5 RCTs (N = 574) considered patients with various depression diagnoses who were randomized to adjunctive NAC, 1,000 mg twice a day, or placebo. Over 12 to 24 weeks, there was a significantly greater improvement in mood symptoms and functionality with NAC vs placebo.¹⁶

Overall, NAC was well-tolerated. The most common AEs were GI symptoms, musculoskeletal complaints, decreased energy, and headache. While NAC has been touted as a potential adjunct therapy for several psychiatric disorders, including TRD, the evidence for benefit remains limited. Given its favorable AE profile, however, and over-the-counter availability, it remains an option for select patients. It is important to ask patients if they are already taking NAC.

Options beyond off-label medications

There are a multitude of options available for addressing TRD. Many FDA-approved medications are repurposed and prescribed off-label for other indications when the risk/benefit balance is favorable. In Part 1 of this article, we reviewed several off-label medications that have supportive controlled data for treating TRD. In Part 2, we will review other nontraditional therapies for TRD, including herbal/nutraceuticals, anti-inflammatory/immune system therapies, device-based treatments, and other alternative approaches.

Bottom Line

Off-label medications that may offer benefit for patients with treatment-resistant depression (TRD) include pimavanserin, antihypertensive agents, ketamine, scopolamine, botulinum toxin, mifepristone, estrogens, buprenorphine, and N-acetylcysteine. Although some evidence supports use of these agents as adjuncts for TRD, an individualized risk/benefit analysis is required.

Related Resource

• Joshi KG, Frierson RL. Off-label prescribing: How to limit your liability. Current Psychiatry. 2020;19(9):12,39.

Drug Brand Names

Amlodipine • Katerzia, Norvasc
Atenolol • Tenormin
Bisoprolol • Zebeta
Buprenorphine • Sublocade, Subutex
Carvedilol • Coreg
Enalapril • Vasotec
Esketamine • Spravato
Estradiol transdermal • Estraderm
Ketamine • Ketalar
Mifepristone • Mifeprex
Pimavanserin • Nuplazid
Progesterone • Prometrium
Propranolol • Inderal
Ramipril • Altace
Verapamil • Calan, Verelan

Presently, FDA-approved first-line treatments and standard adjunctive strategies (eg, lithium, thyroid supplementation, stimulants, second-generation antipsychotics) for major depressive disorder (MDD) often produce less-than-desired outcomes while carrying a potentially substantial safety and tolerability burden. The lack of clinically useful and individual-based biomarkers (eg, genetic, neurophysiological, imaging) is a major obstacle to enhancing treatment efficacy and/or decreasing associated adverse effects (AEs). While the discovery of such tools is being aggressively pursued and ultimately will facilitate a more precision-based choice of therapy, empirical strategies remain our primary approach.

In controlled trials, several nontraditional treatments used primarily as adjuncts to standard antidepressants have shown promise. These include “repurposed” (off-label) medications, herbal/nutraceuticals, anti-inflammatory/immune system therapies, device-based treatments, and other alternative approaches.

Importantly, some nontraditional treatments also demonstrate AEs (Table^1-16). With a careful consideration of the risk/benefit balance, this article reviews some of the better-studied treatment options for patients with treatment-resistant depression (TRD). In Part 1, we will examine off-label medications. In Part 2, we will review other nontraditional approaches to TRD, including herbal/nutraceuticals, anti-inflammatory/immune system therapies, device-based treatments, and other alternative approaches.

We believe this review will help clinicians who need to formulate a different approach after their patient with depression is not helped by traditional first-, second-, and third-line treatments. The potential options discussed in Part 1 of this article are categorized based on their putative mechanism of action (MOA) for depression.

Serotonergic and noradrenergic strategies

Pimavanserin is FDA-approved for treatment of Parkinson’s psychosis. Its potential MOA as an adjunctive strategy for MDD may involve 5-HT2A antagonist and inverse agonist receptor activity, as well as lesser effects at the 5-HT2Creceptor.

A 2-stage, 5-week randomized controlled trial (RCT) (CLARITY; N = 207) found adjunctive pimavanserin (34 mg/d) produced a robust antidepressant effect vs placebo in patients whose depression did not respond to selective serotonin reuptake inhibitors (SSRIs) or serotonin-norepinephrine reuptake inhibitors (SNRIs).¹ Furthermore, a secondary analysis of the data suggested that pimavanserin also improved sleepiness (P < .0003) and daily functioning (P < .014) at Week 5.²

Unfortunately, two 6-week, Phase III RCTs (CLARITY-2 and -3; N = 298) did not find a statistically significant difference between active treatment and placebo. This was based on change in the primary outcome measure (Hamilton Depression Rating Scale-17 score) when adjunctive pimavanserin (34 mg/d) was added to an SSRI or SNRI in patients with TRD.³ There was, however, a significant difference favoring active treatment over placebo based on the Clinical Global Impression–Severity score.

Continue to: In these trials...

In these trials, pimavanserin was generally well-tolerated. The most common AEs were dry mouth, nausea, and headache. Pimavanserin has minimal activity at norepinephrine, dopamine, histamine, or acetylcholine receptors, thus avoiding AEs associated with these receptor interactions.

Given the mixed efficacy results of existing trials, further studies are needed to clarify this agent’s overall risk/benefit in the context of TRD.

Antihypertensive medications

Emerging data suggest that some beta-adrenergic blockers, angiotensin-inhibiting agents, and calcium antagonists are associated with a decreased incidence of depression. A large 2020 study (N = 3,747,190) used population-based Danish registries (2005 to 2015) to evaluate if any of the 41 most commonly prescribed antihypertensive medications were associated with the diagnosis of depressive disorder or use of antidepressants.⁴ These researchers found that enalapril, ramipril, amlodipine, propranolol, atenolol, bisoprolol, carvedilol (P < .001), and verapamil (P < .004) were strongly associated with a decreased risk of depression.⁴

Adverse effects across these different classes of antihypertensives are well characterized, can be substantial, and commonly are related to their impact on cardiovascular function (eg, hypotension). Clinically, these agents may be potential adjuncts for patients with TRD who need antihypertensive therapy. Their use and the choice of specific agent should only be determined in consultation with the patient’s primary care physician (PCP) or appropriate specialist.

Glutamatergic strategies

Ketamine is a dissociative anesthetic and analgesic. Its MOA for treating depression appears to occur primarily through antagonist activity at the N-methyl-D-aspartate ionotropic receptor of the glutamatergic system. There is preliminary evidence that its opioid receptor actions also may contribute to its antidepressant effect.⁵

Continue to: Many published studies...

Many published studies and reviews have described ketamine’s role for treating MDD. Several studies have reported that low-dose (0.5 mg/kg) IV ketamine infusions can rapidly attenuate severe episodes of MDD as well as associated suicidality. For example, a meta-analysis of 9 RCTs (N = 368) comparing ketamine to placebo for acute treatment of unipolar and bipolar depression reported superior therapeutic effects with active treatment at 24 hours, 72 hours, and 7 days.⁶ The response and remission rates for ketamine were 52% and 21% at 24 hours; 48% and 24% at 72 hours; and 40% and 26% at 7 days, respectively.⁶

The most commonly reported AEs during the 4 hours after ketamine infusion included⁷:

• drowsiness, dizziness, poor coordination
• blurred vision, feeling strange or unreal
• hemodynamic changes (approximately 33%)
• small but significant (P < .05) increases in psychotomimetic and dissociative symptoms.

Because some individuals use ketamine recreationally, this agent also carries the risk of abuse.

Research is ongoing on strategies for long-term maintenance ketamine treatment, and the results of both short- and long-term trials will require careful scrutiny to better assess this agent’s safety and tolerability. Clinicians should first consider esketamine—the S-enantiomer of ketamine—because an intranasal formulation of this agent is FDA-approved for treating patients with TRD or MDD with suicidality when administered in a Risk Evaluation and Mitigation Strategy–certified setting.

Cholinergic strategies

Scopolamine is a potent muscarinic receptor antagonist used to prevent nausea and vomiting caused by motion sickness or medications used during surgery. Its use for MDD is based on the theory that muscarinic receptors may be hypersensitive in mood disorders.

Continue to: Several double-blind RCTs...

Several double-blind RCTs of patients with unipolar or bipolar depression that used 3 pulsed IV infusions (4.0 mcg/kg) over 15 minutes found a rapid, robust antidepressant effect with scopolamine vs placebo.^8,9 The oral formulation might also be effective, but would not have a rapid onset.

Common adverse effects of scopolamine include agitation, dry mouth, urinary retention, and cognitive clouding. Given scopolamine’s substantial AE profile, it should be considered only for patients with TRD who could also benefit from the oral formulation for the medical indications noted above, should generally be avoided in older patients, and should be prescribed in consultation with the patient’s PCP.

Botulinum toxin. This neurotoxin inhibits acetylcholine release. It is used to treat disorders characterized by abnormal muscular contraction, such as strabismus, blepharospasm, and chronic pain syndromes. Its MOA for depression may involve its paralytic effects after injection into the glabella forehead muscle (based on the facial feedback hypothesis), as well as modulation of neurotransmitters implicated in the pathophysiology of depression.

In several small trials, injectable botulinum toxin type A (BTA) (29 units) demonstrated antidepressant effects. A recent review that considered 6 trials (N = 235; 4 of the 6 studies were RCTs, 3 of which were rated as high quality) concluded that BTA may be a promising treatment for MDD.¹⁰ Limitations of this review included lack of a priori hypotheses, small sample sizes, gender bias, and difficulty in blinding.

In clinical trials, the most common AEs included local irritation at the injection site and transient headache. This agent’s relatively mild AE profile and possible overlap when used for some of the medical indications noted above opens its potential use as an adjunct in patients with comorbid TRD.

Continue to: Endocrine strategies

Endocrine strategies

Mifepristone (RU486). This anti-glucocorticoid receptor antagonist is used as an abortifacient. Based on the theory that hyperactivity of the hypothalamic-pituitary-adrenal axis is implicated in the pathophysiology of MDD with psychotic features (psychotic depression), this agent has been studied as a treatment for this indication.

An analysis of 5 double-blind RCTs (N = 1,460) found that 7 days of mifepristone, 1,200 mg/d, was superior to placebo (P < .004) in reducing psychotic symptoms of depression.¹¹ Plasma concentrations ≥1,600 ng/mL may be required to maximize benefit.¹¹

Overall, this agent demonstrated a good safety profile in clinical trials, with treatment-emergent AEs reported in 556 (66.7%) patients who received mifepristone vs 386 (61.6%) patients who received placebo.¹¹ Common AEs included gastrointestinal (GI) symptoms, headache, and dizziness. However, 3 deaths occurred: 2 patients who received mifepristone and 1 patient who received placebo. Given this potential for a fatal outcome, clinicians should first consider prescribing an adjunctive antipsychotic agent or electroconvulsive therapy.

Estrogens. These hormones are important for sexual and reproductive development and are used to treat various sexual/reproductive disorders, primarily in women. Their role in treating depression is based on the observation that perimenopause is accompanied by an increased risk of new and recurrent depression coincident with declining ovarian function.

Evidence supports the antidepressant efficacy of transdermal estradiol plus progesterone for perimenopausal depression, but not for postmenopausal depression.^12-14 However, estrogens carry significant risks that must be carefully considered in relationship to their potential benefits. These risks include:

• vaginal bleeding, dysmenorrhea
• fibroid enlargement
• galactorrhea
• ovarian cancer, endometrial cancer, breast cancer
• deep vein thrombosis, pulmonary embolism
• hypertension, chest pain, myocardial infarction, stroke.

Continue to: The use of estrogens...

The use of estrogens as an adjunctive therapy for women with treatment-resistant perimenopausal depression should only be undertaken when standard strategies have failed, and in consultation with an endocrine specialist who can monitor for potentially serious AEs.

Opioid medications

Buprenorphine is used to treat opioid use disorder (OUD) as well as acute and chronic pain. The opioid system is involved in the regulation of mood and may be an appropriate target for novel antidepressants. The use of buprenorphine in combination with samidorphan (a preferential mu-opioid receptor antagonist) has shown initial promise for TRD while minimizing abuse potential.

Although earlier results were mixed, a pooled analysis of 2 recent large RCTs (N = 760) of patients with MDD who had not responded to antidepressants reported greater reduction in Montgomery-Åsberg Depression Rating Scale scores from baseline for active treatment (buprenorphine/samidorphan; 2 mg/2 mg) vs placebo at multiple timepoints, including end of treatment (-1.8; P < .010).¹⁵

The most common AEs included nausea, constipation, dizziness, vomiting, somnolence, fatigue, and sedation. There was minimal evidence of abuse, dependence, or opioid withdrawal. Due to the opioid crisis in the United States, the resulting relaxation of regulations regarding prescribing buprenorphine, and the high rates of depression among patients with OUD, buprenorphine/samidorphan, which is an investigational agent that is not FDA-approved, may be particularly helpful for patients with OUD who also experience comorbid TRD.

Antioxidant agents

N-acetylcysteine (NAC) is an amino acid that can treat acetaminophen toxicity and moderate hepatic damage by increasing glutathione levels. Glutathione is also the primary antioxidant in the CNS. NAC may protect against oxidative stress, chelate heavy metals, reduce inflammation, protect against mitochondrial dysfunction, inhibit apoptosis, and enhance neurogenesis, all potential pathophysiological processes that may contribute to depression.¹⁶

Continue to: A systematic review...

A systematic review and meta-analysis of 5 RCTs (N = 574) considered patients with various depression diagnoses who were randomized to adjunctive NAC, 1,000 mg twice a day, or placebo. Over 12 to 24 weeks, there was a significantly greater improvement in mood symptoms and functionality with NAC vs placebo.¹⁶

Overall, NAC was well-tolerated. The most common AEs were GI symptoms, musculoskeletal complaints, decreased energy, and headache. While NAC has been touted as a potential adjunct therapy for several psychiatric disorders, including TRD, the evidence for benefit remains limited. Given its favorable AE profile, however, and over-the-counter availability, it remains an option for select patients. It is important to ask patients if they are already taking NAC.

Options beyond off-label medications

There are a multitude of options available for addressing TRD. Many FDA-approved medications are repurposed and prescribed off-label for other indications when the risk/benefit balance is favorable. In Part 1 of this article, we reviewed several off-label medications that have supportive controlled data for treating TRD. In Part 2, we will review other nontraditional therapies for TRD, including herbal/nutraceuticals, anti-inflammatory/immune system therapies, device-based treatments, and other alternative approaches.

Bottom Line

Off-label medications that may offer benefit for patients with treatment-resistant depression (TRD) include pimavanserin, antihypertensive agents, ketamine, scopolamine, botulinum toxin, mifepristone, estrogens, buprenorphine, and N-acetylcysteine. Although some evidence supports use of these agents as adjuncts for TRD, an individualized risk/benefit analysis is required.

Related Resource

• Joshi KG, Frierson RL. Off-label prescribing: How to limit your liability. Current Psychiatry. 2020;19(9):12,39.

Drug Brand Names

Amlodipine • Katerzia, Norvasc
Atenolol • Tenormin
Bisoprolol • Zebeta
Buprenorphine • Sublocade, Subutex
Carvedilol • Coreg
Enalapril • Vasotec
Esketamine • Spravato
Estradiol transdermal • Estraderm
Ketamine • Ketalar
Mifepristone • Mifeprex
Pimavanserin • Nuplazid
Progesterone • Prometrium
Propranolol • Inderal
Ramipril • Altace
Verapamil • Calan, Verelan

Presently, FDA-approved first-line treatments and standard adjunctive strategies (eg, lithium, thyroid supplementation, stimulants, second-generation antipsychotics) for major depressive disorder (MDD) often produce less-than-desired outcomes while carrying a potentially substantial safety and tolerability burden. The lack of clinically useful and individual-based biomarkers (eg, genetic, neurophysiological, imaging) is a major obstacle to enhancing treatment efficacy and/or decreasing associated adverse effects (AEs). While the discovery of such tools is being aggressively pursued and ultimately will facilitate a more precision-based choice of therapy, empirical strategies remain our primary approach.

In controlled trials, several nontraditional treatments used primarily as adjuncts to standard antidepressants have shown promise. These include “repurposed” (off-label) medications, herbal/nutraceuticals, anti-inflammatory/immune system therapies, device-based treatments, and other alternative approaches.

Importantly, some nontraditional treatments also demonstrate AEs (Table^1-16). With a careful consideration of the risk/benefit balance, this article reviews some of the better-studied treatment options for patients with treatment-resistant depression (TRD). In Part 1, we will examine off-label medications. In Part 2, we will review other nontraditional approaches to TRD, including herbal/nutraceuticals, anti-inflammatory/immune system therapies, device-based treatments, and other alternative approaches.

We believe this review will help clinicians who need to formulate a different approach after their patient with depression is not helped by traditional first-, second-, and third-line treatments. The potential options discussed in Part 1 of this article are categorized based on their putative mechanism of action (MOA) for depression.

Serotonergic and noradrenergic strategies

Pimavanserin is FDA-approved for treatment of Parkinson’s psychosis. Its potential MOA as an adjunctive strategy for MDD may involve 5-HT2A antagonist and inverse agonist receptor activity, as well as lesser effects at the 5-HT2Creceptor.

A 2-stage, 5-week randomized controlled trial (RCT) (CLARITY; N = 207) found adjunctive pimavanserin (34 mg/d) produced a robust antidepressant effect vs placebo in patients whose depression did not respond to selective serotonin reuptake inhibitors (SSRIs) or serotonin-norepinephrine reuptake inhibitors (SNRIs).¹ Furthermore, a secondary analysis of the data suggested that pimavanserin also improved sleepiness (P < .0003) and daily functioning (P < .014) at Week 5.²

Unfortunately, two 6-week, Phase III RCTs (CLARITY-2 and -3; N = 298) did not find a statistically significant difference between active treatment and placebo. This was based on change in the primary outcome measure (Hamilton Depression Rating Scale-17 score) when adjunctive pimavanserin (34 mg/d) was added to an SSRI or SNRI in patients with TRD.³ There was, however, a significant difference favoring active treatment over placebo based on the Clinical Global Impression–Severity score.

Continue to: In these trials...

In these trials, pimavanserin was generally well-tolerated. The most common AEs were dry mouth, nausea, and headache. Pimavanserin has minimal activity at norepinephrine, dopamine, histamine, or acetylcholine receptors, thus avoiding AEs associated with these receptor interactions.

Given the mixed efficacy results of existing trials, further studies are needed to clarify this agent’s overall risk/benefit in the context of TRD.

Antihypertensive medications

Emerging data suggest that some beta-adrenergic blockers, angiotensin-inhibiting agents, and calcium antagonists are associated with a decreased incidence of depression. A large 2020 study (N = 3,747,190) used population-based Danish registries (2005 to 2015) to evaluate if any of the 41 most commonly prescribed antihypertensive medications were associated with the diagnosis of depressive disorder or use of antidepressants.⁴ These researchers found that enalapril, ramipril, amlodipine, propranolol, atenolol, bisoprolol, carvedilol (P < .001), and verapamil (P < .004) were strongly associated with a decreased risk of depression.⁴

Adverse effects across these different classes of antihypertensives are well characterized, can be substantial, and commonly are related to their impact on cardiovascular function (eg, hypotension). Clinically, these agents may be potential adjuncts for patients with TRD who need antihypertensive therapy. Their use and the choice of specific agent should only be determined in consultation with the patient’s primary care physician (PCP) or appropriate specialist.

Glutamatergic strategies

Ketamine is a dissociative anesthetic and analgesic. Its MOA for treating depression appears to occur primarily through antagonist activity at the N-methyl-D-aspartate ionotropic receptor of the glutamatergic system. There is preliminary evidence that its opioid receptor actions also may contribute to its antidepressant effect.⁵

Continue to: Many published studies...

Many published studies and reviews have described ketamine’s role for treating MDD. Several studies have reported that low-dose (0.5 mg/kg) IV ketamine infusions can rapidly attenuate severe episodes of MDD as well as associated suicidality. For example, a meta-analysis of 9 RCTs (N = 368) comparing ketamine to placebo for acute treatment of unipolar and bipolar depression reported superior therapeutic effects with active treatment at 24 hours, 72 hours, and 7 days.⁶ The response and remission rates for ketamine were 52% and 21% at 24 hours; 48% and 24% at 72 hours; and 40% and 26% at 7 days, respectively.⁶

The most commonly reported AEs during the 4 hours after ketamine infusion included⁷:

• drowsiness, dizziness, poor coordination
• blurred vision, feeling strange or unreal
• hemodynamic changes (approximately 33%)
• small but significant (P < .05) increases in psychotomimetic and dissociative symptoms.

Because some individuals use ketamine recreationally, this agent also carries the risk of abuse.

Research is ongoing on strategies for long-term maintenance ketamine treatment, and the results of both short- and long-term trials will require careful scrutiny to better assess this agent’s safety and tolerability. Clinicians should first consider esketamine—the S-enantiomer of ketamine—because an intranasal formulation of this agent is FDA-approved for treating patients with TRD or MDD with suicidality when administered in a Risk Evaluation and Mitigation Strategy–certified setting.

Cholinergic strategies

Scopolamine is a potent muscarinic receptor antagonist used to prevent nausea and vomiting caused by motion sickness or medications used during surgery. Its use for MDD is based on the theory that muscarinic receptors may be hypersensitive in mood disorders.

Continue to: Several double-blind RCTs...

Several double-blind RCTs of patients with unipolar or bipolar depression that used 3 pulsed IV infusions (4.0 mcg/kg) over 15 minutes found a rapid, robust antidepressant effect with scopolamine vs placebo.^8,9 The oral formulation might also be effective, but would not have a rapid onset.

Common adverse effects of scopolamine include agitation, dry mouth, urinary retention, and cognitive clouding. Given scopolamine’s substantial AE profile, it should be considered only for patients with TRD who could also benefit from the oral formulation for the medical indications noted above, should generally be avoided in older patients, and should be prescribed in consultation with the patient’s PCP.

Botulinum toxin. This neurotoxin inhibits acetylcholine release. It is used to treat disorders characterized by abnormal muscular contraction, such as strabismus, blepharospasm, and chronic pain syndromes. Its MOA for depression may involve its paralytic effects after injection into the glabella forehead muscle (based on the facial feedback hypothesis), as well as modulation of neurotransmitters implicated in the pathophysiology of depression.

In several small trials, injectable botulinum toxin type A (BTA) (29 units) demonstrated antidepressant effects. A recent review that considered 6 trials (N = 235; 4 of the 6 studies were RCTs, 3 of which were rated as high quality) concluded that BTA may be a promising treatment for MDD.¹⁰ Limitations of this review included lack of a priori hypotheses, small sample sizes, gender bias, and difficulty in blinding.

In clinical trials, the most common AEs included local irritation at the injection site and transient headache. This agent’s relatively mild AE profile and possible overlap when used for some of the medical indications noted above opens its potential use as an adjunct in patients with comorbid TRD.

Continue to: Endocrine strategies

Endocrine strategies

Mifepristone (RU486). This anti-glucocorticoid receptor antagonist is used as an abortifacient. Based on the theory that hyperactivity of the hypothalamic-pituitary-adrenal axis is implicated in the pathophysiology of MDD with psychotic features (psychotic depression), this agent has been studied as a treatment for this indication.

An analysis of 5 double-blind RCTs (N = 1,460) found that 7 days of mifepristone, 1,200 mg/d, was superior to placebo (P < .004) in reducing psychotic symptoms of depression.¹¹ Plasma concentrations ≥1,600 ng/mL may be required to maximize benefit.¹¹

Overall, this agent demonstrated a good safety profile in clinical trials, with treatment-emergent AEs reported in 556 (66.7%) patients who received mifepristone vs 386 (61.6%) patients who received placebo.¹¹ Common AEs included gastrointestinal (GI) symptoms, headache, and dizziness. However, 3 deaths occurred: 2 patients who received mifepristone and 1 patient who received placebo. Given this potential for a fatal outcome, clinicians should first consider prescribing an adjunctive antipsychotic agent or electroconvulsive therapy.

Estrogens. These hormones are important for sexual and reproductive development and are used to treat various sexual/reproductive disorders, primarily in women. Their role in treating depression is based on the observation that perimenopause is accompanied by an increased risk of new and recurrent depression coincident with declining ovarian function.

Evidence supports the antidepressant efficacy of transdermal estradiol plus progesterone for perimenopausal depression, but not for postmenopausal depression.^12-14 However, estrogens carry significant risks that must be carefully considered in relationship to their potential benefits. These risks include:

• vaginal bleeding, dysmenorrhea
• fibroid enlargement
• galactorrhea
• ovarian cancer, endometrial cancer, breast cancer
• deep vein thrombosis, pulmonary embolism
• hypertension, chest pain, myocardial infarction, stroke.

Continue to: The use of estrogens...

The use of estrogens as an adjunctive therapy for women with treatment-resistant perimenopausal depression should only be undertaken when standard strategies have failed, and in consultation with an endocrine specialist who can monitor for potentially serious AEs.

Opioid medications

Buprenorphine is used to treat opioid use disorder (OUD) as well as acute and chronic pain. The opioid system is involved in the regulation of mood and may be an appropriate target for novel antidepressants. The use of buprenorphine in combination with samidorphan (a preferential mu-opioid receptor antagonist) has shown initial promise for TRD while minimizing abuse potential.

Although earlier results were mixed, a pooled analysis of 2 recent large RCTs (N = 760) of patients with MDD who had not responded to antidepressants reported greater reduction in Montgomery-Åsberg Depression Rating Scale scores from baseline for active treatment (buprenorphine/samidorphan; 2 mg/2 mg) vs placebo at multiple timepoints, including end of treatment (-1.8; P < .010).¹⁵

The most common AEs included nausea, constipation, dizziness, vomiting, somnolence, fatigue, and sedation. There was minimal evidence of abuse, dependence, or opioid withdrawal. Due to the opioid crisis in the United States, the resulting relaxation of regulations regarding prescribing buprenorphine, and the high rates of depression among patients with OUD, buprenorphine/samidorphan, which is an investigational agent that is not FDA-approved, may be particularly helpful for patients with OUD who also experience comorbid TRD.

Antioxidant agents

N-acetylcysteine (NAC) is an amino acid that can treat acetaminophen toxicity and moderate hepatic damage by increasing glutathione levels. Glutathione is also the primary antioxidant in the CNS. NAC may protect against oxidative stress, chelate heavy metals, reduce inflammation, protect against mitochondrial dysfunction, inhibit apoptosis, and enhance neurogenesis, all potential pathophysiological processes that may contribute to depression.¹⁶

Continue to: A systematic review...

A systematic review and meta-analysis of 5 RCTs (N = 574) considered patients with various depression diagnoses who were randomized to adjunctive NAC, 1,000 mg twice a day, or placebo. Over 12 to 24 weeks, there was a significantly greater improvement in mood symptoms and functionality with NAC vs placebo.¹⁶

Overall, NAC was well-tolerated. The most common AEs were GI symptoms, musculoskeletal complaints, decreased energy, and headache. While NAC has been touted as a potential adjunct therapy for several psychiatric disorders, including TRD, the evidence for benefit remains limited. Given its favorable AE profile, however, and over-the-counter availability, it remains an option for select patients. It is important to ask patients if they are already taking NAC.

Options beyond off-label medications

There are a multitude of options available for addressing TRD. Many FDA-approved medications are repurposed and prescribed off-label for other indications when the risk/benefit balance is favorable. In Part 1 of this article, we reviewed several off-label medications that have supportive controlled data for treating TRD. In Part 2, we will review other nontraditional therapies for TRD, including herbal/nutraceuticals, anti-inflammatory/immune system therapies, device-based treatments, and other alternative approaches.

Bottom Line

Off-label medications that may offer benefit for patients with treatment-resistant depression (TRD) include pimavanserin, antihypertensive agents, ketamine, scopolamine, botulinum toxin, mifepristone, estrogens, buprenorphine, and N-acetylcysteine. Although some evidence supports use of these agents as adjuncts for TRD, an individualized risk/benefit analysis is required.

Related Resource

• Joshi KG, Frierson RL. Off-label prescribing: How to limit your liability. Current Psychiatry. 2020;19(9):12,39.

Drug Brand Names

Amlodipine • Katerzia, Norvasc
Atenolol • Tenormin
Bisoprolol • Zebeta
Buprenorphine • Sublocade, Subutex
Carvedilol • Coreg
Enalapril • Vasotec
Esketamine • Spravato
Estradiol transdermal • Estraderm
Ketamine • Ketalar
Mifepristone • Mifeprex
Pimavanserin • Nuplazid
Progesterone • Prometrium
Propranolol • Inderal
Ramipril • Altace
Verapamil • Calan, Verelan