Here are my thoughts on bipolar.
The contemporary concept of bipolar disorder provides fertile ground for exploring the interfaces among mind, brain, body, relationship, environment, spirit, and culture. But what is bipolar disorder? Is it a thing, like a rock or a liver? Is it a convenient word? Is it a firm biological reality or is it a cultural construct riding upon the waves of an uncertain biological sea? Is it more plot than condition; strategy, than a thing; verb, than a noun?
The conventional story about bipolar disorder. The conventional story about bipolar disorder is a consensual one, forged by psychiatrists and other mental health professionals in collaboration with patients, the media, and the pharmaceutical companies. It developed through communication and dialogue. People learned to recognize the signifiers of bipolar disorder within themselves and to present them to professionals. Whatever the underlying biology of bipolar disorder is, the communication about it and the development of its signifiers and the learning by the population to assume that label and present themselves to professionals, and the response by professionals – all of that is highly cultural laden. Given the predominant cultural concept that genetic or biological illnesses require pharmaceuticals, then that is what is offered. But there are alternate strategies for all conditions and modes of suffering – traditional Chinese medicine, for example, uses needles, herbs, massage, diet, and counseling. Traditional North American healing uses ceremony, prayer, massage, herbs, and more, whether for arthritis or for excessive and prolonged sadness.
In the conventional story bipolar disorder is a chronic disease affecting over 2 million Americans at some point in their lives. The American Psychiatric Association's "Diagnostic and Statistical Manual of Mental Disorders" describes two forms of bipolar disorder, type I and type II.  In type I, there has been at least one full manic [*] episode.* In type II, periods of "hypomania" involve less severe manic symptoms that alternate with at least one major depressive episode. People who suffer from bipolar disorder are thought to have pathological mood swings from mania to depression, with cyclic patterns of exacerbation and remission.
The Social Construction of Mood. Culture enters when we encounter mood. Mood is not an object like a table. We must learn how to describe the emotions that characterize mood. Cultures recognize and enact emotions differently. Mood varies from family to family and from culture to culture.
I suspect that the youngest children feel without words. They live an unbroken stream of experience. When they exhibit behavior that significant adults in their lives can interpret, a pointing or labelling process begins. A significant adult says, “Oh, so you’re sad.” Now a category appears and the unbroken experience becomes organized into a concept called sadness. As children learn words and language, they learn to label their internal states in some correspondence with the way “they appear to be feeling” to the adults who care for them. Families define sadness in different ways as do cultures. Some languages even lack words for sadness. Similarly, behavior can become labelled as excitement or as irritability in a sophisticated exercise in pointing. The adult points at the child and says a word to label what the child is doing and experiencing. The child learns to equate the word with his internal experiences associated with the word and begins to say that he feels sad or excited or irritable. A major project in negotiating love relationships is the collaborative mapping of what words for emotions mean in one family with what the same words mean in another family. My sadness may bear little resemblance to yours.
All the thoughts we "can think and the mental operations [we] can perform have their source in some … interpretive community."  The range, complexity, and subtlety of our thought, its power, the practical and conceptual uses to which we can put it, and the issues we can address result from the degree to which we have been initiated into the knowledge communities to which we belong. "Human thought is consummately social: social in its origins, social in its functions, social in its form, social in its applications.  The thoughts and feelings that come to be labelled as bipolar disorder are initially social.
Values, habits, emotions, manners of behaving at the table, and spitting are transmitted through social interaction. Erasmus  wrote manuals of good behavior to codify social interaction. Conversely, social interaction produces patterns of behavior.
Throughout our childhood, and even before birth, we are a lifelong process of negotiation. Even infants engage in conversations with their mothers and other caretaking adults through crying, smiling, and through their eyes. Because their well-being depends on understanding their mother's language, both verbal and gestural, infants are interpreting these conversations with their mothers (and other caretaking adults) as soon as they can register and distinguish changes in physical attitude and gesture, tone of voice, and facial expression. And because a mother's well-being depends in part on understanding and adapting to her infant's needs, infant and mother are, to that extent, knowledgeable peers. Together they compose a unique but culturally crucial knowledge community whose members are learning from each other as they go.
Vygotsky  wrote a classic description of this process of community composition and collaboration that involved a six-month-old infant. The infant saw an attractive object—a shiny spoon—and extended his hand to grasp it. The spoon was out of reach. For a moment, Vygotsky said, the infant's "hands stretched toward that object, remaining poised in the air. His fingers made grasping movements." The infant appeared to be trying, at the most elemental level, to establish contact with a bit of physical reality. Shoved around by the physical world, he shoved back. He wanted a response from the object or a relationship with it that corresponded to his reaching out for it. But the object did not cooperate in the effort to be known. Objects never do. For a moment, then, the infant reached and nothing happened.
“Then something did happen. The object still didn't cooperate, but mother did. The infant's mother moved the object closer, so that the infant could feel it, look at it, and put it into his mouth.”
This brief, mundane scene provides a key to understanding knowledge and collaborative learning. When infants reach for an object, they do not merely reach. They send a message. When a caretaker gets the message and responds, infants learn indelibly the importance of this seemingly irrelevant side effect. Our first effort to grasp an object, Vygotsky tells us, is the first step we take in learning to point. Pointing, Vygotsky argues, “is an unsuccessful attempt to grasp something, a movement aimed at a certain object which designates forthcoming activity. . . . When the mother comes to the child's aid and realizes that his movement indicates something, the situation changes fundamentally. Pointing becomes a gesture for others. The child's unsuccessful attempt engenders a reaction not from the object he sought but from another person  .
Vygotsky tells us that knowing is not an unmediated, direct relationship between us and an object. We need other people’s involvement in order to know something. Other people are always involved in our learning processes. The infant in Vygotsky's illustration eventually learns to know and master the shiny spoon through learning how to make an adult respond to give it the spoon. Infants begin to "understand [their grasping] movement as pointing," Vygotsky says, when they understand that their "object-oriented movement" has really become "a movement aimed at another person, a means of establishing relations."  Learning always involves relationships with other people. The experience is collaborative because, when they finally get the message and respond, the caretakers have understood the infant. They have learned a gestural word or phrase with which the infant is now able to converse. They have learned to expect forthcoming activity from the infant. From the infant's point of view, they have learned to obey orders.
These moments in the lives of six-month-old infants contend seriously for the attention of college and university teachers, because the process implied can be traced from infancy through childhood to the learning of adults. Infant and mother learn what they need to know about each other by internalizing the language that constitutes their community, encapsulating the results of their ongoing conversations in conventions and routines. As infants grow and learn, becoming children and then adolescents and adults, they incrementally nest membership in that first, small, closed knowledge community of mother and child, expanding toward communities with which to pledge allegiance. 
Vygotsky described the actions of a four or five-year-old child trying to take possession of a piece of candy by figuring out how to use some basic tools to advantage, in this case a stick and a stool. As the child worked, she talked through her solution to the problem. But she did not talk in a state of fantasy involvement with the objects that concerned her. She talked about them, and about herself, to someone. Sometimes she talked to another person at hand. Most of the time, she talked to herself as if she were another person.
Vygotsky said that the child was using social speech instrumentally, to get something done. By the time she was four or five, much of her "socialized speech (which had previously been used to address an adult) had turned inward.” Rather than appealing to the adult, she appealed to herself. 
Vygotsky observed children talking to themselves as if they were talking to someone else. Eventually, they did talk to themselves silently and private thoughts emerged. He wrote that "every function in [our] cultural development appears twice: first, on the social level, and later, on the individual level; first, between people . . ., and then inside."
Children and adults interact to shape and change each others’ responses. Learning and understanding emerge as individuals create and accomplish interactive tasks in everyday conversations. Bamberg calls this talk-in-interaction. We learn as we go. In this same spirit, people learn to recognize their emotions and to talk about their emotions through interactive dialogue with each other. Knowledge about sadness cannot be separated from the conversation going on between the people involved. "Knowledge" is not separate from "society." Instead, we see "trials of strength"1 in which knowledge, conversation, emotional involvement, and social relationships are inseparable. By exercising her native talent for linguistic improvisation, a young child translates and retranslates until she gets it "correct."
Intuitively we know this. In a classic Gunsmoke episode, a mother has run away from her gunslinger husband when she discovers that she is pregnant. This is so her child will not grow up to be like his father. The drama unfolds when the gunslinger rides into town and recognizes his wife and realizes he has a child. He leaves when he realizes that she is right – the child will grow up to be just like him if he stays or if he takes his family with him. Vygotsky’s insights show us how we learn our emotional strategies as we grow up within our families. This argument suggests that the emotional strategies and negotiations that come to be labeled as bipolar are socially learned through ongoing interactive processes in families and shape the brain and its connections as they go. Genetic susceptibility may also play a role that remains to be determined (twins separated at birth help us to tease this out), but the observation that the bipolar strategy runs in families is not evidence for its genetic basis but for its existence in social learning. (We will shortly consider biology and genetics.)
A correct response or emotion is what is acceptable to the community in which the child lives. Membership in a community means that everything we do is unhesitatingly correct or incorrect according to specific criteria within that local community.
The child interacts with the important adults in her community to learn how to correctly point to sad, angry, irritable, mad, and all the other labels available in that community. Eventually she internalizes these conversations. At first, she talks to herself. Soon she thinks to herself without the need for talking. She learns about emotions and their proper expression.
As we mature, we internalize conversations about emotions as thoughts. The fact that we tend to re-externalize thought under stress as direct or indirect conversation (talk and writing) demonstrates the continuing relationship of thought and conversation in adult learning, even when that relationship ceases to be readily apparent. It is stress that occasions our talking to ourselves ("Don’t let yourself feel so sad.”). It is stress that occasions the rap sessions and endless dorm-room talk typical of adolescence and early adulthood." 
Having made that digression into how children learn to categorize emotional experience, to carve it up from the unbroken whole, and how they learn to be emotional and express emotions, returns us to the possibility that being manic (which is defined as elevated mood, hyperactivity, over-involvement in activities, inflated self-esteem, a tendency to be easily distracted, and little need for sleep) may be learned. We may learn how to enact mania as a strategy for communication. Perhaps it is a partially successful strategy that gets carried away with itself? Once learned, perhaps we cannot stop. Depression typically follows mania, with its accompanying loss of self-esteem, withdrawal, sadness, and even risk of suicide. I suspect we also learn how to do depression – when to label ourselves as depressed. We learn what useful functions depressive actions play in social life, and how to be depressed. Mania and depression may also be like light and darkness, the opposites that must co-exist.
Modern psychiatry's recognition of bipolar disorder comes from Kraepelin's 1921 application of the term “manic-depressant insanity” to cyclic episodes of mania alternating with depression, a syndrome which has been recognized in various forms for over 2000 years [Barclay RM (trans), Robertson GM (ed). (1921). Manic-Depressive Insanity and Paranoia. Edinburgh, E & S Livingstone.]The profession introduced the term “bipolar disorder” in the mid-1970s in a largely unsuccessful attempt to lessen confusion between this condition and schizophrenia [Kupfer DJ. Epidemiology and clinical course of bipolar disorder. In Kupfer DJ (ed.) Bipolar Disorder: The Clinician’s Reference Guide. Montvale, NJ: Clinical Psychiatry LLC, 2004.]
In the conventional story, bipolar disorder appears between the ages of 15 and 25, affecting men and women equally. From 1.2 percent13 to 1.6 percent [Kessler RC, McGonagle KA, Zhao S, et al: Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States. Arch Gen Psychiatry 51:8-19, 1994.]14 of the US population is diagnosed as bipolar with prevalence increasing. The recognized incidence is thought to be an underestimate because of underreporting and under-recognition of manic and hypomanic episodes. The dominant discourse of modern psychiatry recognizes pharmacological treatment as the mainstay for bipolar disorder.
The symptoms consist of alternating episodes of mania and depression. In the manic phase, we see an increase in goal-directed activities (either socially or at work), increased energy, distractibility, flight of ideas or subjective experience that thoughts are racing, an inflated self-esteem or grandiosity, an increased involvement in activities that may be pleasurable, but may have dire consequences (e.g., spending sprees), a decreased need for sleep (person feels rested after 3 hours of sleep). The patient may be more talkative than usual or may feel pressured to speak. He may be easily agitated or irritated and may lack self-control
Hypomanic episodes are similar but less intense. Delusions, if present, are congruent with mood (such as delusions of grandeur, or a sense of special powers and abilities).
In the depressive phase people experience persistent sadness and depressed mood; feelings of hopelessness, worthlessness, pessimism, and "emptiness"; loss of interest or pleasure in activities that were once enjoyed, including sex; sleep disturbances; motor slowing or agitation; withdrawal; feelings of guilt and worthlessness; fatigue; overwhelming sluggishness; difficulty concentrating, remembering, or making decisions; loss of appetite and/or weight loss, or overeating and weight gain; and thoughts of death or suicide.
If delusions are present, they are typically congruent with mood (such as delusions of worthlessness or accusing voices). In "atypical depression," people sleep more than usual and have increased appetite.
Conventional medicine usually hospitalizes people who are having acute symptoms so that medications may be started to control the symptoms. These medications include neuroleptics (antipsychotics), antianxiety agents (such as benzodiazepines), anticonvulsants, and antidepressant agents. The conventional story involves other people only as support group members or in educational sessions about the illness.
Biological models for bipolar depression have focused largely on the effects of uncontrollable stressors [Swann AC. (2006). Neurobiology of Bipolar Depression in El-Mallack RS, Ghaemi SN.(eds.) Bipolar Depression: A comprehensive guide. Washington, DC: American Psychiatric Publishing, Inc., pp. 37 – 68.] These models have moderate pharmacological validity but lack any specificity for distinguishing types of depression (routine or unipolar depression from bipolar depression, for example) [Machado-Vieira R, Kapczinski F, Soares JC. (2004). Perspectives for the development of animal models of bipolar disorder. Prog Neuropsychopharmacol Biol Psychiatry 28: 209-224; Nestler EJ, Gould E, Manji J. (2002). Preclinical models: status of basic research in depression. Biol Psychiatry 52: 503-528.] The core depressive symptoms are indistinguishable between bipolar depression and unipolar depression [Mitchell P, Parker G, Jamieson K. (1992). Are there any differences between bipolar and unipolar melancholia? J Affective Disorders 25:97-105.] On average people with bipolar depression are more slowed down and experience more lack of energy than people with unipolar depression [Katz et al 1982; Kupfer et al 1974] though some studies contradict this [Mitchell et al, 1992]. These differences are not diagnostic, and, in fact, the two depressions cannot be distinguished by symptoms [Benazzi 2003b; Katz et al, 1982]
Another emerging type of depression is called a mixed depression in which the symptoms of depression are accompanied by two or more symptoms of mania [Benazzi 2003a]. The presence of manic symptoms does help to distinguish somewhat, with two manic symptoms present in 78.1% of people with bipolar depression and 41.5% of people with unipolar depression. Three manic symptoms were present in 46.6% of bipolar depressed people compared to 7.6% of unipolar depressed people [Benazzi 2001]. Contrary to popular belief, the presence of anxiety and inner tension is ubiquitous in all depressions [Benazzi et al 2004; Katz et al 1982; Wolff et al 1985]. What does suggest bipolar depression is the presence of the other manic symptoms, such as increased goal-directed activity, grandiosity, hypersexuality, or true racing thoughts [First et al 1996; Frank et al 2002; Swann et al 1993], Increased norepinephrine function has been found in predominantly manic mixed states compared with pure depressive episodes [Swann et al 1994].
All studies considered, it has been impossible to determine if depression, bipolar or otherwise, stems from too little or too much of any neurotransmitter [Maas et al 1991]. A second generation of unsuccessful hypotheses implicated imbalances between neurotransmitters, such as norepinephrine versus serotonin [Prange et al 1974] or norepinephrine versus acetylcholine [Janowsky et al 1972]. A third generation of unsuccessful hypotheses held that second messenger function associated with neurotransmitter receptors with increased activity during mania [Lachman and Papolos, 1995; Stewart et al 2001].
There is a state-dependent elevation of norepinephrine in manic and mixed states, but no reliable changes in norepinephrine or it metabolites during depression [Swann et al 1987; Koslow et al 1983]. Norepinephrine is apparently metabolized differently during depression with lower relative concentrations of its intracellular metabolites, consistent with increased pulsatile releases [Maas et al 1987; Swann et al 1987] in mania. A mathematical calculation (the D-score) of different amine metabolite levels does discriminate between bipolar I depression, bipolar II depression, and other depressions [Grossman and Potter, 1999; Schatzberg et al 1989]
People with bipolar depression appear to have increased reactivity to norephinephrine. Moreso in bipolar than in unipolar depression, norepinephrine is related to mood and “slowing down” [Swann et al 1990], treatment response [Maas et al 1984] and relationship to stressful events [Swann et al 1990]. People with bipolar depression have increased sensitivity to the subjective effects of stimulants [Anand et al 2000]. Pharmacologically increased norepinephrine precipitates mania in people with bipolar depression [Price et al 1984] and may selectively improve bipolar depression [Osman et al 1989]. People with bipolar disorder have a greater noradrenergic response to orthostasis [Rudorfer et al 1985], having more noradrenergic neurons in the locus coeruleus [Baumann and Bogerts, 2001].
3-met04 hoxy-4-hydroxyphenylene glycol (MHPG), the major metabolite of norepinephrine, is low among patients with bipolar depression, suggesting a role for central norepinephrine in this disorder. [iv] . One-fourth to one-half of MHPG is derived from the central nervous system and the remainder from the adrenal medulla and the sympathetic nervous system. The most common endocrine finding is excess cortisol, resulting from excess secretion of corticotrophin releasing hormone, which is stimulated by norepinephrine and acetylcholine, and inhibited by GABA. Hypercritical, exploitative and emotionally unresponsive environments can feature as important precursors. Chronic stress early in life in vulnerable persons is thought to predispose them to both bipolar and ordinary depression.
Studies of serotonergic function are consistent with reduced functional capacity but are not specific [Price et al 1991; Sher et al 2003; Sobczak et al 2002]. Indepependent of all else, lower serotinergic function may be related to potential suicidality [Goodwin and Post, 1983; Mann, 1999], though the relationship may be stronger in unipolar than in bipolar depression [Stokes et al 1984]. Corticospinal fluid cortisol concentrations and degree of dexamethasone suppression test nonsuppression are related to depressed mood, especially in mixed states, among people with bipolar depression [Swann et al 1992]. Sensitivity to norepinephrine does appear to differentiate bipolar from unipolar depression along with blunted responses to 5-hydroxy-tryptophan in non-depressed people and abnormal behavioral responses to tryptophan depletion in relatives of people with bipolar depression.
Recent studies suggest that the disturbances in bipolar depression may relate to systems involved in neuronal adaptations to changes in activity or second messenger systems, including the nitric oxide system [Akyol et al 2004]. One indirect study showed lower blood arginine levels and higher nitrite levels among people with bipolar depression [Van Calker and Belmaker 2000; Yanik et al 2004]. Cell-signaling systems, particularly involving inositol and protein kinase C may be involved in the action of so-called mood stabilizing drugs [Harwood and Agam 2003]. Systems involved membrane lipids, such as the arachidonic acid cascade, may be important [Rapoport 2001].
Subtle abnormalities in arousal, lateralization, and susceptibility to impulsivity have been found among people with bipolar depression[Buchsbaum et al 1977; Brocke et al 2000; Hegerl et al 2001; Dubal et al 2000]. Abnormalities in arousal or in sensitivity to neurotransmitters may be related to abnormal regulation of ion distribution [Whybrow and Mendels, 1969]. Active transport is reduced per sodium pump site in cultured, lymphoblastoid cells from Old Order Amish people with bipolar depression compared with nonaffected relatives or controls [Cherry and swann, 1994]. The response of the active transport of sodium to increased sodium influx maintains membrane potential over time in excitable cells, provides the cation gradient that drives uptake processes for neurotransmitters and other compounds and is the major cause of activity-dependent energy utilization [Stahl 1986]. This process is diminished in cells from people with bipolar disorder [Li and El-Mallakh 2004]. Inhibition of the active transport of sodium by ouabain leads to abnormal hippocampal cell excitability [El-Mallakh et al 2000] and increased motor activity in ras [El-Mallakh et al 1995, 2003].
Several studies have found alterations in glial density in people with bipolar depression when compared with normal controls, including reductions in the subgenual part of the anterior cingulate cortex [Chana et al 2003] and the subgenual part of the prefrontal cortex [Ongur et al 1998]. Glial cells are responsible for glutamate clearance. A build-up in glutamate from reduced clearance could result in overexcitation of neurons [Schurr 2002; Schurr et al 1997a] with consequent excitotoxicity [Lipton, 2004]. Neuronal loss has been found in layer III of the dorsolateral prefrontal cortex [Rajkowska et al 2001]. Glial cell loss in the amygdala was only evident in people who had not been treated [Bowley et al 2002]. A glial specific protein, the alpha2 subunit of the sodium potassium ATPase pump, has been found to be reduced in the temporal cortex of people with bipolar depression [Rose et al 1998].
Impairments in fine motor function are found in all depressions, but are more closely linked to noradrenergic function and to severity of the depression among people with bipolar depression [Swann et al 1999]. Treatment response [Maas et al 1984] and sensitivity to stressful life events [Swann et al 1990] are more strongly related to norepinephrine among people with bipolar depression.
In the story of neuroimaging, we are seeking replicable findings associated with specific behavioral states. To date few neuroimaging findings have been replicable from laboratory to laboratory. [Neuroimaging Studies of Mood Disorder Effects on the Brain. Yvette I. Sheline. Biol Psychiatry 2003;54:338–352.]
Associated with Size of Temporal Lobe Other
Depression Ventricles changes Changes
Reduced size of prefrontal cortex
(Coffman et al 1990; Schlaepfer et al 1994; Strakowski et al 1993)
Reduced prefrontal grey matter independent of treatment of mood state, bipolar or unipolar
(Drevets et al 1998 in Swann)
Reduction in total cortical volume (DelBello et al 2004)
Increased size of amydgala (Altshuler et al 1998); Decreased size of amydgala (Pearlson et al 1977)
No cortical grey matter loss
Dupont et al 1995; Harvey et al 1994; Pearlson et al 1997; Schlaepfer et al 1994; Zipursky et al 1997.
No change in amygdala size (Swayze et al 1992)
Decrease in cortical grey matter, intermediate betweel control and schizophrenics
(Lim et al 1999)
Increased lateral ventricle size
Swayze et al 1990; Figiel et al 1991; Strakowski et al 1993
Mixed results for size changes in thalamus
(Dupont et al 1995; Strakowski et al 1993)
Reduced volume of amygdala in adolescents (DelBello et al, 2004 in Swann)
Reduced signal intensity in the corpus callosum consistent with neuropsychological reports of switching (Pettigrew and Miller, 1998) (Branbilla et al 2004 in Swann)
Reduced temporal lobe size
(Altshuler et al 1991)
Decreased Hippocampal size (Altshuler et al, 1991)
Harvey et al 1994; McDonald et al 1991
Mixed results for size changes for hippocampus
(Altshuler et al 1998; Hauser et al 1989; Swayze et al 1992)
Increased right hippocampal volume correlated with poor cognitive functions (Ali et al 2000 in Swann)
Loss of normal asymmetry
Swayze et al 1992
Increased size on left and in size of amygdale and striatumHarvey et al 1994
No difference in size
Johnstone et al 1989). Strakowski et al (1999)
Chronic lithium treatment prevents volume loss
(Manji et al 2000)
Enlarged cortical sulci (fissures on the surface of the brain) found in middle aged (Lim et al 1999)
Decreased numbers of glial cells in the prefrontal cortex
(Ongur et al 1998)
Larger caudate nucleus size in males
(Aylward et al 1994)
Larger globus pallidum volume with no change in striatal volume
(Strakowski et al 1999)
No differences in caudate, putamen, lenticular nuclei (Dupont et al 1995; Strakowski et al 1993; Swayze et al 1992)
Increased white matter hyperintensities, decreased cerebellar size, and increased sulcal and third ventricular volumes (Stoll et al 2000 in Swann)
No relationship with hyperintensities (Brown et al 1992; Sassi et al 2003 in Swann)
Bipolar with multiple episodes of mania
Lateral ventricular enlargement
Strakowski et al 2002
Decreased size of Hippocampus, amygdala, basal ganglia, frontal cortex associated with depression
(Starkstein and Robinson 1989)
Decreased size of Hippocampus, amygdala, basal ganglia, frontal cortex associated with depression
Decreased size of Hippocampus, amygdala, basal ganglia, frontal cortex associated with depression
(Sawrie et al 2001)
Decreased size of Hippocampus, amygdala, basal ganglia, frontal cortex associated with depression
(Burns et al 1990)
Decreased size of parietal cortex
(Kanne et al 1998)
Decreased size of Hippocampus, amygdala, basal ganglia, frontal cortex associated with depression
(Folstein et al 1983)
Major unipolar depressoin
7% overall reduction in frontal lobe volume (Coffey et al 1992)
48% reduction in volume in the subgenual prefrontal cortex (Drevets et al 1997)
Subgenual prefrontal cortex glial cell loss (Ongur et al 1998)
Rostral orbitofrontal cortex decreases in cortical thickness, neuronal size decrease, and loss of glial cells in layers II to IV (Rajkowska et al 1999)
Reductions in glial and neuronal cells throughout all layers, as well as reduction in cell size, were reported in dorsolateral prefrontal cortex
(Rajkowska et al 1999)
Increased volume in the right amygdala (Bremner et al 2000)
Increased volume in bilateral amygdala in first episode subjects (Frodl et al 2002b)
Loss of normal amygdalar asymmetry (Mervaala et al 2000)
Reduction in the bilateral core nuclei of the amygdale (Sheline et al 1998)
Decreased volumes of basal ganglia structures in major depression, especially in late-onset depression (Greenwald et al 1997; Husain et al 1991; Krishnan et al 1992; Steffens and Krishnan 1998)
No changes in caudate and putamen volume in depressed subjects who were otherwise physically healthy (Lenze and Sheline 1999), a criterion not clearly present in other studies.
Hippocampal volume loss
(Bell-McGinty et al 2002; Bremner et al 2000; MacQueen et al 2003; Shah et al 1998; Sheline et al 1996, 1999), but not in bipolar depression (Geuze et al 2004)
No change in hippocampal volume
(Ashtari et al 1999; Axelson et al 1993; Mervaala et al 2000; Swayze et al 1992)
Reductions in hippocampal volumes
(Vakili et al 2000)
Hippocampus, amygdale, basal ganglia, frontal cortex
cortical and subcortical atrophy
(Pantel et al 1997; Rabins et al 1991; Rothschild et al 1989; Soares and Mann 1997)
Diffuse and ventricular enlargement
(Pantel et al 1997; Rabins et al 1991; Rothschild et al 1989; Soares and Mann 1997)
Higher rates of neuropsychological impairment and greater treatment refractoriness
(Alexopoulos et al 2002; Simpson et al 1998)
Associated with brain atrophy
(Kobayashi et al 1991)
Associated with brain atrophy
(Starkman et al 1992)
Associated with brain atrophy
Associated with brain atrophy
Overall the structural imaging studies described above suggest that there are anatomical abnormalities that may exist relatively early in bipolar depression, but their functional and diagnostic significance is uncertain (Kanner 2004 in Swann).
The prefrontal cortex plays a key role in thinking, also modulating emotional centers of the basal ganglia and limbic regions. A two-armed circuit of brain areas has been proposed as being involved with depression. One arm consists of a limbic-thalamic-cortical branch and is composed of the amygdala, the hippocampus, the dorsomedial nucleus of the thalamus, and the medial and ventrolateral prefrontal cortex. A second limbic-striatal-pallido-thalamic branch is proposed as the other arm of this circuit. The caudate and putamen (striatum) brain areas along with the globus pallidus are organized in parallel to connect with limbic and cortical regions.
One hypothesis (Swerdlow and Koob 1987) about the depressive symptoms of bipolar disorder says that it results from disinhibition of the limbic striatum from underactive forebrain dopamine activity, thereby producing overinhibition of the ventral pallidum with decreased inhibitory connection with the dorsomedical thalamus, which, in turn, results in disinhibition of the excitatory loop involving the mediodorsal thalamus, prefrontal cortex, and amygdala. This is thought to underlie the guilty ruminations, motor slowing, and recurrent thoughts of death found in depression. It does not relate to decreased attention and impairment in executive functioning (Degl’Innocenti et al 1998). The brain story is fascinating and can be seen parallel to the cultural story. What we quickly forget is how biological understanding does not restrict us to biological causality. Biology is inseparable from culture – the major point of this work.
The neurological diseases associated with depression involve damage to brain structures critical in emotional functioning -- namely the frontal cortex, hippocampus, thalamus, amygdala, and basal ganglia. These same brain structures are involved in classical and early-onset major depression (Jellinger 1999). Deoxyglucose metabolism studies using positron-emission tomography (PET) have demonstrated selectively decreased activity in the caudate and orbital-inferior frontal lobe (Mayberg et al 1990). Ischemic lesions located in the anterior frontal cortex were associated with more severe depression (Robinson et al, 1983 and Lipsey et al, 1983). Subsequently, inconsistent results have been reported on the relationship between infarct site and depression after stroke, with systematic review of the numerous studies not supporting the hypothesis that stroke lesion location predicts depression (Carson et al 2000). A strong correlation has been found between lesions affecting the prefrontosubcortical circuits, particularly on the left, and with subsequent depression (Vataja et al 2001).. Among individuals with cognitive impairment, baseline depression was associated with a threefold increased risk of dementia. In vivo MRI studies (Steffens et al 2002) have shown that small left hippocampal size predicts later dementia.
Late-age depression is characterized by a lower prevalence of affective disorders in other family members (Baron et al 1981), greater medical morbidity and mortality (Jacoby et al 1981), and higher rates of neuroradiological abnormalities, particularly white-matter hyperintensities (Coffey et al 1988; Figiel et al 1991).
Any condition which produces neuronal ischemia or neurotoxicity can potentially contribute to brain atrophy.
Some of the MRI volumetric findings in frontal cortex could be accounted for by neuropathological changes such as these. The prefrontal cortex is particularly important as a target of monoamine projections and abnormalities in monoamine receptors, transporters, and second messenger systems (Arango et al 1995; Biver et al 1997; Duman 1998; Mintun et al 2000; Price 1999) are reported to occur in major depression. Another possibility is that overactivation in one part of the interconnected LCSPT neuroanatomical circuit may lead to overexcitation in the other components, resulting in excitotoxic damage. The orbitomedial prefrontal cortex has high concentrations of glucocorticoid receptors, potentially rendering it vulnerable to stress-mediated damage
Hippocampal volume loss appears to have functional significance with an association between acute depression and abnormalities of declarative memory (Burt et al 1995) and recollection memory (Mac-Queen et al 2003), as well as an association between depression in remission and lower scores on tests of verbal memory (Sheline et al 1999). In one study (Shah et al 1998), hippocampal atrophy was found in patients with chronic depression but not in patients with remitted depression. Vakili et al (2000) also observed correlations between depression severity and hippocampal volumes, although no group differences between depressed and control subjects. In one study (Frodl et al 2002a), white matter changes were noted but no overall differences in hippocampal volume. In most of these studies that assessed depression severity in unipolar subjects and used high-resolution MRI techniques, depression was associated with hippocampal volume loss, ranging from 8% to 19%. Studies which only measured the hippocampus/ amygdala complex found no differences. A recent postmortem study (Bowley et al 2002) has found glial cell loss in the dentate gyrus of the hippocampus as well as in the amygdala in major depression. In addition, a recent study has found increased neuronal and glial cell packing density (Stockmeier et al, unpublished data), suggesting a decrease in the hippocampal neuropil in MDD.
Potential Mechanisms for Volume Loss in Recurrent Depression. Approximately half of depressive episodes are associated with elevated cortisol levels. Hypothalamic-pituitary-adrenal (HPA) axis dysfunction can produce repeated episodes of hypercortisolemia in depression. Volume studies do not routinely include measures of cortisol and cannot determine past episodes of hypercortisolemia. In addition to elevated cortisol levels, several different mechanisms could potentially explain volume loss, including neuronal loss through exposure to repeated episodes of hypercortisolemia, stress-induced reduction in neurotrophic factors, stress-induced reduction in neurogenesis, and glial cell loss, resulting in increased vulnerability to glutamate neurotoxicity. Glucocorticoid (GC)-mediated neurotoxicity (Sapolsky 2000) with repeated hypercortisolemic episodes of depression giving rise to atrophy of affected structures is a mechanism that could potentially account for hippocampal, amygdala, and prefrontal cortex volume loss, all areas which have high concentrations of GC receptors; however, it is also well known that the hippocampus has structural plasticity, driven by excitatory amino acids and facilitated by glucocorticoids. In animal studies (Watanabe et al 1992), hippocampal apical dendrites shortened by a single GC exposure or restraint stress returned to normal after 3 weeks. In Cushing’s disease, following successful surgery and a return to normal for GC levels, previously smaller hippocampal volumes returned to normal (Starkman et al 1992; Bourdeau et al 2002). Thus, up to a point, plasticity may be at least partially reversible. Early life stress may produce a permanent hypersensitivity to stress, with the production of ongoing HPA axis dysregulation, particularly in subjects who develop depression (Heim et al 2000). With repeated episodes, plasticity may give way to permanent damage. Inverse correlations between the total amount of time patients have been depressed and hippocampal volume found in some studies (Bell-McGinty et al 2002; Mac-Queen et al 2003; Sheline et al 1996, 1999) but not all (Bremner et al 2000) support recurrent depressive episodes having an antecedent or causal relationship. In addition, a study by Lupien et al (1998) demonstrated a correlation between higher cortisol levels measured longitudinally and greater hippocampal volume loss in normal human aging. A study of first episode patients identified memory impairment on neuropsychological testing but no hippocampal volume loss, whereas multiple episode patients in the same study had both memory impairment and volume loss (MacQueen et al 2003). Thus, while neurotoxic damage may occur, plasticity would permit return of function if the right intervention were used in time.
Excitatory connections between the amygdala and hippocampus (White and Price 1993) raise the possibility that damage in one structure could produce damage in the connected structure. Also, interconnections between prefrontal cortex and hippocampus (Carmichael and Price 1995) could produce excitotoxic damage. Glial cells sequester glutamate, maintain metabolic and ionic homeostasis, and produce trophic factors, including brain derived neurotrophic factor (BDNF) (Ransom and Sontheimer 1992; Szatkowski and Attwell 1994). Thus, loss of glial cells could increase vulnerability to neurotoxic damage, supporting the idea that glutamate neurotoxicity may be involved in the volume loss in the limbic-cortical-striatal-pallidal circuit.
Either directly or indirectly, glial cell loss is another potential mechanism for producing volume loss. Gray matter atrophy has been reported in the prefrontal cortex in an area ventral to the genu of the corpus callosum (Drevets et al 1997), an area associated in postmortem studies with glial cell loss (Ongur et al 1998). Glial cell loss has been found in two different areas of prefrontal cortex (Rajkowska et al 1999), as well as in the amygdala and the hippocampus (Bowley et al 2002) in postmortem studies of major depression.
Stress-induced inhibition of neurogenesis (Gould et al 1997) may also explain depression-related volume loss. Psychosocial stress has been shown to suppress neurogenesis in the tree shrew (Gould et al 1997). Corticosterone treatment in adult rats also produced suppression of neurogenesis, which was reversed by removal of the adrenal gland (Cameron and Gould 1994). It is also possible (Gould et al 1999) that neurogenesis may occur in the frontal cortex in addition to the hippocampus and subventricular zone.
PET Scan studies in bipolar depression
Mood Induction Studies
Dorsolateral pre-frontal cortex
Reduced glucose metabolism (also unipolar depression)
Ketter et al 2001; Strakowski et al 2000, in Swann)
Negative correlations (Dunn et al 2002)
Reduced N-acetylaspartate levels (Winsberg, et al 2000) (correlated with reduced neuronal integrity.
Correlates negatively with 2-deoxyglucose uptake in unipolar depression (Dunn et al 2002)
Increased responses to both positive and negative stimuli compared to unipolar and controls (Lawrence et al 2004 in Swann)
Increases responses to both positive and negative stimuli (Lawrence et al, 2094
(Ketter above in Swann)
Ventral cingulated-cortical-limbic activity
Changes found similar to those of controls, especially those with depressive temperaments (Keightley et al 2003 in Swann)
Mobilizing additional subcorticla and limbic areas (Malhi et al 2004).
Insula and claustrum
Negatively correlated with 2-deoxyglucose uptake (Dunn et al 2002 in Swann)
Anterior cingulated cortex
Positively correlated with 2-deoxyglucose uptake (Dunn et al 2002 in Swann)
Supracallosal cingulated gyrus
Negative correlation with 2-deoxyglucose uptake (Dunn et al. 2002)
Abnormal choline metabolism (Strakowski et al 2000)
Anterior cingulate gyrus
Abnormal choline metabolism (Moore et al 2000 in Swann)
PET studies suggest three levels of specificity: subjects with negative affective states regardless of diagnosis (Keightley et al 2003 in Swann), depressive subjects regardless of polarity (Dunn et al, 2002; Ketter et al 2001 in Swann), and bipolar subjects regardless of affective state (Ketter et al. 2001 in Swann). The studies suggest poorly regulated affective responses, possibly resulting from the failure of the prefrontal cortex to modulate subcortical and temporal signals (Strakowski 2004, 2005 in Swann). While some abnormalities can be found early (Delbello et al 2004), others develop later (Strakowski et al 2004).
While physiological and receptor imaging abnormalities in bipolar disorder are less well established, noteworthy findings that await replication include elevated resting amygdala activity which correlates with stress related plasma cortisol levels (Drevets, Ann Rev Med 49:331-361; 1998); abnormal blood flow responses to viewing facial expressions of emotion in the amygdala, basal forebrain and inferotemporal cortex; reduced dopamine D1 receptor radioligand binding in the frontal cortex (Suhara et al. 1992); increased striatal uptake of [11C]- N-methylspiperone, a dopamine D2 receptor ligand among psychotic bipolar people relative to controls and non-psychotic bipolar subjects (Pearlson et al. 1995).
Bipolar depression has a more recurrent course (Angst et al 2003; Kessing and Andersen, 1999) with earlier onset (Akiskal et al 1994; Benazzi 2002, 2004; Kessing, 1999), and more frequent episodes (Angst et al 2003; Goldberg and Harrow, 2004; Kessing, 1999; Kessing and Andersen, 1999; Winokur and Wesner, 1987). Rapid fluctuations between high and low can occur in either form of depression, but is more common in bipolar depression (Wolpert et al 1999).
Kindling and sensitization models have been proposed to account for the recurrent nature of bipolar depression (Antelman et al 1998), implying that early episodes would be more likely to be associated with environmental stressors and later episodes becoming progressively more autonomous. This turns out to be true for both unipolar and bipolar depression (Swann et al 1990).
Top of Form
The The rates of concordance in monozygotic twins ranges from 47% to 70% (Craddock N, Jones I. 1999. Genetics of bipolar disorder. J Med Genet. Aug;36(8):585-94) compared to only 14 percent of dyzygotic twins.11 [NEJM Aug 11 1988; 319(6 part 1)348-53, Aug 18 1988; 319(7 part 2):413-20.] The approximate lifetime risk of bipolar disorder in relatives of a person with bipolar disorder are: monozygotic co-twin, 40-70 percent; first degree relative, 5-10 percent (Craddock and Jones, 1999); and unrelated person , 0.5-1.5 percent.12 [Psychiatric Ann Jul 1989; 19(7):354-59]. Among offspring of a parent with bipolar disorder, 51% were found to have a psychiatric disorder and the risk for bipolar was increased with earlier onset in the parent (Chang et al 2000). The age of onset appears to get earlier in successive generations ( Rice et al 1987). Unipolar depression is increased in families with bipolar depression (Gershon et al 1982). An epidemiological study found an association of mania and major depression in twins, and only a small effect on heritability of major depression if a history of mania was removed, consistent with a continuum model for inheritability for unipolar and bipolar depression (Karkowski and Kendler, 1997). In a study of 67 bipolar (30 monozygotic) and 176 unipolar (68 monozygotic) twin pairs, heritability of bipolar depression was reported to be 85%, with 71% of the genetic risk for mania not shared for depression (McGuffin et al 2003).
Genetic Studies. No clear area has emerged for heritability for bipolar depression, and multiple alleles have been proposed – in fact, several on every chromosome (Hayden and Nurnberger, 2006) The search for genetic anomalies in the alleles regulating neurotransmitter systems has larger been negative (Swann, 2006). A form of the serotonin transporter gene appears to be related to response to SSRIs in both unipolar and bipolar depression (Lerer and Macciardi, 2002; Serretti et al 2004). Neither tryptophan hydroxylase alleles nor serotonin receptor 1A, 2A, and 2C were related to lithium responsivity (Serretti et al 1999, 2000). The incidence of a form of the %5-HT2C receptor was reported elevated in both unipolar and bipolar depression (Lerer et al 2001). No differences were found between controls, people with unipolar depression, and people with bipolar depression in alleles of the 5-HT1B receptor (Huang et al 2003), 5-HT2A (Massat et al 2000; Ni et al 2002), 5-HT5A (Arias et al 2001), serotonin transporter (Cusin et al 2001; Mansour et al 2005) and tryptophan hydroxylase (Cusin et al 2001).
No genetic differences have been found between unipolar and bipolar depressed people (or normal controls) in any of the catecholamine enzyme or receptor systems genes (Swann, 2006). One study associated the LL allele of the COMT (catechol-O-methyl transferase) gene (which has low activity, leading to reduced extracellular breakdown of catecholamines) with rapid cycling (Papolos et al 1998). The A1 allele of the D2 dopamine receptor gene was associated with increased risk for substance abuse (Noble 2000). A form of the D4 dopamine receptor gene was associated with delusions in either unipolar or bipolar depression (Serretti et al 1998b).
Genetic investigations of GABA-A receptors (Coon et al 1994; Serretti et al 1998a), corticotrophin releasing hormone synthesis (Stratakis et al 1997) and proneurotensin synthesis (Austin et al 2000) have all yielded no results.
A dinucleotide repeat of the brain-derived neurotrophic factor (BDFN) is associated with increased risk for childhood-onset mood disorders (Wood et al 2003). BDFN is important in neural adaptations to stress and also has antidepressant properties in animal models (Hashimoto et al 2004). In family based studies the val66met allele was associated with bipolar disorder (Neves-Perreira et al 2002; Sklar et al 2002) and childhood onset bipolar disorder (Geller et al 2004), but it did not distinguish bipolar, unipolar, or normal controls in case-control studies (Nakata et al 2003; Neves-Pereira et al 2002; Oswald et al 2004)or identify children with childhood-onset mood disorders in case control studies (Wood et al 2003). This form of BDNF was also associated with childhood obsessive compulsive disorder (Hall et al 2003) which may be related to risk for bipolar disorder (Chen and Dilsaver, 1995; Thomsen, 1992).
Clock genes have been associated with increased recurrence in bipolar depression (Benedetti et al 2003) and with age at onset (Benedetti et al 2004). A form of GSK-3-beta may be protective, but has a low frequency (Benedetti et al 2004).
Brain tissue studies show evidence of abnormal regulation of receptor second messenger signaling, but not in the receptor binding sites themselves for thalamic glutamatergic systems (Clinton et al 2004). Reductions have been found in a group of synaptic proteins called complexins in schizophrenia and bipolar depression, but not in unipolar depression (Eastwood and Harrison, 2000). Expression of the CREB gene was increased in suicide victims, regardless of diagnosis (Young et al 2004).
Conventional Treatments. Various texts have been written about conventional therapies [El-Mallakh RS, Ghaemi SN. (2006) Bipolar Depression: A comprehensive guide. Washington, DC: American Psychiatric Association Press.] so that it is not necessary to repeat that information here. For conventional therapies, I am partial to the Texas Medication Algorithm [ref], which has been shown effective in a number of studies and superior to psychiatrists working without an algorithm. The only addition I personally make to that Algorithm is to use quietapine (Seroquel) also as monotherapy for bipolar depression related to recent studies showing its efficacy [ref]. I prefer lamotrigine due to its fewer side effects, but recognize that quietapine is thought to have a more rapid onset of actions. For the purposes of this book, I would prefer to focus on the uncertainties which remain in treatment and the potential role that alternatvec could play as either adjuncts to conventional treatment or replacements. \
What must be emphasized about conventional treatments is that success is often partial, non-responders are common, treatments wear off, and side effects can be serious and debilitating. Depression and depressive cycling remains a substantial problem for about two-thirds of intensively treated bipolar outpatients (Post [Post RM, Leverich GS, Nolan WA. (2003). A re-evaluation of the role of antidepressants in the treatment of bipolar depression: data from the Stanley Foundation bipolar network. Bipolar Disorder 5: 396-406.]
For example, in a study on the use of aripiprazole for acute bipolar mania [Keck PE, Marcus R, Tourkodimitris S, Ali M, Liebeskind A, Saha A, Ingenito G, Aripiprazole Study Group (2003). A Placebo-Controlled, Double-Blind Study of the Efficacy and Safety of Aripiprazole in Patients with Acute Bipolar Mania. Am J Psychiatry 160: 1651-1658], an agent which I do use in conventional settings when the need arises, the response rate was only 40% (compared to a 19% response rate for the placebo). The trial was only three weeks, which leaves open the question of medication wearing off. The reduction in symptom severity on the Young Mania Scale when from 8.2 to 3.4, which means that people were, on average, still symptomatic, albeit less so. Only 42% of patients studied completed the three week trial (compared to 21% with placebo), meaning that the majority of the patients did not or could not remain on this medication. Of course, the exclusion criteria eliminated the more severe patients (as drug studies usually do). For example, anyone with mania lasting more than 4 weeks was excluded. Anyone who might need another medication was excluded (how this could be known in advance puzzles me). Anyone who was contemplating suicide or had taken illicit drugs or drank too much alcohol was excluded. As usual, the majority of the patients I see would be ineligible for inclusion in this study. Only 31% of total patients completed the trial, meaning that over two-thirds of patients discontinued double blind treatment. Twenty-one percent of patients were switched to open label treatment (13% of those receiving drug and 28% of those receiving placebo), 10% discontinued because of an adverse reaction (11% for drug, and 10% for placebo). Eleven percent discontinued for lack of efficacy (10% with drug; 12% with placebo), and 27% were lost for “other” reasons, including withdrawing consent, disappearing, or being deemed unreliable.
The adverse events in the drug group consisted of three people becoming more manic, one decompensating, one overdosing on sedatives, and one becoming hypertensive. People receiving placebo reported agitation, accidental injury, chest discomfort, syncope, and urticaria. The list of adverve events appears below, and is typical for these types of studiesAdverse Event
We can also conclude from the above study that placeboes are certainly powerful, which speaks to the power of the mind.
A second placebo-controlled study addressed aripiprazole in the treatment of acute manic or mixed episodes in patients with bipolar type I disorder with similar results [Sachs G, Sanchez R, Marcus R, Stock E, McQuade R, Carson W, Abou-Gharbia N, Impellizzeri C, Kaplita S, Rollin L, Iwamoto T, The Aripiprazole Study Group. (2005). Aripiprazole in the treatment of acute manic or mixed episodes in patients with bipolar I disorder: a 3-week placebo-controlled study. J. Psychopharmacology (Pre-print)]. This study excluded the same types of patients and started with 272 hospitalized patients. This time, 53% of subjects completed the three week study. Reasons for discontinuation were similar. The response rate to the drug was 39% by the end of the first week (compared to 27% with placebo) and 53% by the end of the third week (compared to 32% with placebo). Apparently most placebo responders, respond within the first week, while medication responders continue to accumulate after the first week. Drug treated patients did improve more than placebo treated patients on a number of measures, including the Clinical Global Inventory (CGI), the Young Mania Scale, and the Positive and Negative Symptom Scale, but not the Montgomery-Asburg Depression Scale.
Divalproex15 [Sachs G, Collins M. A placebo-controlled trial of divalproex sodium in acute bipolar depression. Paper presented at: 40th annual meeting of the American College of Neuropsychopharmacology.; December 9-13, 2001.] and other anticonvulsants are used as single agents in the treatment of mania, but have bot been shown as effective for bipolar depression with the exception of lamotrigine.1 An 8-week, multi-center study assigned 22 people to divalproex and 22 to placebo. At the study’s end, drug and placebo were similar in reducing depressive symptoms.
While the data for lamotrigine17 as a single agent for bipolar depression is convincing, many people are still left out in the cold. studying one study comparing low dose (50 mg/day), high dose (200 mg/day)and placebo with over 60 people in each group,18 [Calabrese JR. Bowden CI, Sachs GS. Ascher JA, Monaghan E, Rudd GD. A double-blind, placebo controlled study of lamotrigine monotherapy in outpatients with bipolar I depression. Lamictal 602 Study Group. J Clin. Psychiatry 1999; 60(2):79-88.] 44 to 50% of people taking lamotrigine improved compared to 36% of people taking placebo on the Hamilton Rating Scale for Depresson. On another rating scale – the Montgomery Asburg Depression Rating Scale, 50 to 54 % of people taking lamotrigine improved compared to 28 percent of people taking placebo, similar to what was seen on the Clinical Global Impressions Scale, on which 42-50 percent of people taking lamotrigine improved compared to 28% (again) taking placebo. The high percentage of non-responders and the high percentage of placebo responders again, suggests to me, that more alternatives are needed for treatment and that internal, environmental, and other poorly understand factors are subsumed under the heading of placebo. The presence of a placebo response shows that people improve despite medical treatment, and we should study these people to understand why.
A second study found lamotrigine to be equivalent to placebo in treating a mixed group of people with both bipolar I and II depressions. It was more effective than placebo only for people with bipolar I depression.19 [Bowden CL. Novel treatments for bipolar disorder. Expert Opinion Invetig Drugs 2001;10(4): 661-671.] In a third study, people with “treatment resistant, rapid-cycling” bipolar depression improved more with lamotrigine than with placebo.
Combining an antidepressant with an anticonvulsant is common for bipolar depression. A study of 27 people compared two anticonvulsants with an anticonvulsant plus an antidepressant (most commonly, an SSRI like Prozac). People tolerated the second combination better than the first, with a 0 percent drop-out rate compared to a 38 percent drop out rate for anticonvulsants alone. The effectiveness was similar with half the people unresponsive.
Lithium alone or with an anticonvulsant or with a third drug, an SSRI or a tricyclic antidepressant is another common combination.. [Nemeroff CB, Evans DL, Gyulai L. Double-blind, placebo-controlled comparisons of imipramine and paroxetine in the treatment of bipolar depression. Am J Psychiatry 2001; 158(6): 906-12.] In a study of paroxetine (Paxil)20, imipramine, or placebo with lithium, the highest response rate barely topped 50 percent.
Psychotherapy. Psychotherapy is effective with bipolar depression [Colom F, Vieta E. Psychological interventions in bipolar depression. In El Mallack RS, Ghaemi SN. (2006). Bipolar Depression: A comprehensive guide. Washington, DC: American Psychiatric Association Press, pp. 215-226.] though I would draw a distinction between psychotherapy and healing, which will be discussed under alternative therapies. My particular approach to healing with bipolar depression involves (in addition to medications or nutrients, either of which addresses the underlying mood instability) involves the use of narrative therapy, energy medicine, and spiritual healing. I find this much more effective than psychotherapy alone, which we all do, since, as Colom and Vieta point out, much of cognitive-behavior therapy is just plain common sense.
A number of conventional psychotherapeutic approaches have achieved statistical significance in randomized, controlled trials, including family focused interventions,[Miklowitz DJ, George EL, Richards AJ. (2003). A randomized study of family focused psychoeducation and pharmacotherapy in the outpatient management of bipolar disorder. Arch Gen Psychiatry 60: 904-912.], training in prodromal identification [Perry A, Tarrier N, Morris R (1999). Randomised controlled trial of efficacy of teaching patients with bipolar disorder to identify early symptoms of relapse and obtain treatment. Br Med J (418: 149-153.], cognitive-behavioral therapy [Lam DH, Watkins ER, Hayward P. (2003). A randomized, controlled study of cognitive therapy for relapse prevention for bipolar affective disorder. Outcome of the first year. Arch Gen Psychiatry 60: 145-152.], and psychoeducation [Colom F, Vieta E, Martinez-Aran A. (2003a). A randomized trial on the efficacy of group psychoeducation in the prophylaxis of recurrences in bipolar patients whose disease is in remission. Arch Gen Psychiatry 60: 402-407; Colom F, Vieta E, Reinares M. (2003b). Psychoeducation efficacy in bipolar disorders beyond compliance enhancement. J Clin Psychiatry 4: 1101-1105.].In Canada, 85% of psychiatrists include psychotherapy in their management of bipolar disorder in addition to medications (Sharma V, Masmanian DS, Persad E. (1997). Treatment of bipolar depression: A survey of Canadian psychiatrists. Can J Psychiatry 42: 298-302.].
Nutrient therapies. Several studies have demonstrated that psychiatric symptoms such as depression, mood swings, and aggression may be ameliorated by supplementation with broad-based nutrient formulas containing vitamins, minerals, and sometimes essential fatty acids. [ Kaplan BJ, Fisher JE, Crawford SG, Field CJ, Kolb B. Improved mood and behavior during treatment with a mineral-vitamin supplement: an open-label case series of children. J Child Adolesc Psychopharmacol. 2004 Spring;14(1):115-22. ]
Eleven patients were studied to determine the therapeutic benefit of a nutritional supplement (Empower Plus) for bipolar depression. The study consisted of 11 people with DSM-IV-diagnosed bipolar disorder who were aged 19 to 46 years and taking a mean of 2.7 psychotropic medications each at entry into the study. The supplement was a broad-based combination of chelated trace minerals and vitamins administered in high doses. At study entry and periodically thereafter, people’s symptoms were assessed with the Hamilton Rating Scale for Depression, the Brief Psychiatric Rating Scale (BPRS), and the Young Mania Rating Scale (YMRS). For those who completed the minimum 6-month open trial, symptom reduction ranged from 55% to 66% on the outcome measures. The need for psychotropic medications decreased by more than 50%. Benefit was shown on all the outcome measures with a reduction in the mean HAM-D score at entry from 19.0 to 5.4, in the BPRS mean score from 35.3 to 7.4, and in the YMRS mean score at entry from 15.1 to 6.0, all statistically significant. The effect size for the intervention was large (> .80) for each measure. The number of psychotropic medications decreased significantly. In some cases, the supplement replaced psychotropic medications and the patients remained well. The only reported side effect (i.e., nausea) was infrequent, minor, and transitory. [ Kaplan BJ , Simpson JS , Ferre RC , Gorman CP , McMullen DM , Crawford SG . (2001). Effective mood stabilization with a chelated mineral supplement: an open-label trial in bipolar disorder. J Clin Psychiatry. Dec;62(12):936-44.]
A study of 11 children (7 boys and 4 girls) with mood and behavioral problems participated in an open-label trial of the role of a nutritional supplement (Empower Plus) in treating their symptoms. [ Kaplan BJ, Fisher JE, Crawford SG, Field CJ, Kolb B. Improved mood and behavior during treatment with a mineral-vitamin supplement: an open-label case series of children. J Child Adolesc Psychopharmacol. 2004 Spring;14(1):115-22]. Parents completed the Child Behavior Checklist (CBCL), Youth Outcome Questionnaire (YOQ), and theYoung Mania Rating Scale (YMRS) at entry and following at least 8 weeks of treatment. The YOQ and the YMRS improved significantly from the baseline to the final visit. Improvement was significant on seven of the eight CBCL scales, the YOQ, and the YMRS. The effect sizes for all the outcome measures were relatively large.
The biological narrative has even infiltrated the field of literary criticism as one would expect since we are all embedded in the same larger culture and share a common history, whether psychiatrists or English professors. In writing about Virginia Woolf's mood swings (which we are now calling bipolar disorder), Thomas Caramagno21 [Caramagno, Thomas. The flight of the mind: Virginia Woolf's art and manic-depressive illness. Berkeley: University of California Press, 1992] demonstrated the effect of the psychiatric narrative on literary criticism. He says, “[We must reconsider Virginia Woolf’s] fiction in light of recent medical discoveries about the genetic and biological nature of manic-depression—findings allied with drug therapies that today help nearly one million American manic-depressives live happier and more productive lives. In the real world of the clinic, treatments using lithium, anti-depressants, and anti-psychotics have revolutionized psychiatric care for mood swings and produced miracle remissions for cases that thirty years ago would have been considered hopeless. But in the rarefied atmosphere of literary academia, many critics still cling to the Freudian model of this disorder as a neurotic conflict that the patient is unwilling (either consciously or unconsciously) to resolve.”
Proponents of the biological story argue superiority over the previous Freudian story in that it describes people as unable to change instead of unwilling to change. What hasn’t changed is the assignation of the “problem” to the individual with an implicit judgment of inferiority. Freud’s movement stole “mental problems” from the Church as moral inferiority to create psychological inferiority. Biological psychiatry transforms this to genetic inferiority. Caramagno has characteristically little to say about non-responders and placebo responders.
The costs of conventional psychiatric treatment of bipolar disorder are large. A 1991 report from the National Institutes of Mental Health estimated total U.S. costs for bipolar illness at $45 billion annually. It is growing exponentially.
The construction of bipolar disorder as a biological-genetic disorder allows psychiatry to flourish. The accompanying idea that bipolar disorder can only be treated with medication allows the pharmacology industry to flourish. People diagnosed with bipolar disorder who subscribe to the biological story can relinquish all need to participate in any conversations except about which medications to take. Larger communities are excluded from dialogue except to educate family about “how to live with a mentally ill member,” as one class is described at the hospital where I work. If bipolar is entirely biological and is entirely treated biologically, then no value comes from conversations among affected parties except to educate them from the expert paradigm.
Gergen writes that “enormous problems inhere in distinctly psychological modes of explanation.”22 [Gergen KJ. Realities and Relationships: Soundings in Social Construction. Cambridge, MA: Harvard University Press, 1994, p. 276.] The same can be said for distinctly biological modes of explanation. We could say that privileged explanations are usually inadequate.
What do non-responders do within the biological narrative. Their lives are rarely studied. I have observed that they pursue a variety of courses, common ones included chronic alcohol and/or drug abuse as a self-medication strategy. Some use hidden talents and resources to excel in communities were their mood swings become attractive eccentricities. The stereotype of the leading lady of the 1940’s had this kind of moodiness. Some pursue other types of healing, spontaneously loose their bipolarity, die, become soldiers, have spiritual transformations, and otherwise remain hidden from the dominant discourse. I want to tell several of their stories in the next section.
I have tried to find references to what could be considered bipolar like symptoms in writings about post-Columbus indigenous people, only to have failed. The closest I have come is to the visionary healer or spiritual leader and the fabled warrior. Descriptions exist of those who have had days of ecstatic visions only to collapse and sleep for days. This behavior was honored and supported. If it was “manic” or “psychotic”, the person was protected during the ecstasy since the visions were expected to be portentous, of great importance to the tribal group. When spirit left the person, as these descriptions commonly report, and the person collapsed, the community was present to nurse them back to health. If this was bipolar disorder, it had a completely different meaning and context in tribal society.
Perhaps the tendency of indigenous cultures to see problems in terms of whole communities and not individuals renders the bipolar label unintelligible. Perhaps bipolar requires reduction to individualism in order to appear or to exist. Within traditional communities, I suspect that affect is regulated differently than among members of modern Euro-American culture. When people live tribally, women menstruate at the same time, and probably other biological cycles synchronize. I suspect that the community modulates affect and provides regulation when the individual is unable to do so. Biological psychiatry’s genetic drift hypothesis states that the more severe bipolar people would be isolated and alone, since they are too bizarre to maintain normal social relations. Of course, they would say this from their individualistic, biological story. An indigenous explanation might reverse this and say that isolation and loneliness without community is harmful.
People’s Alternative Stories
Mary, like many people who have acquired the bipolar label, was fascinating. She was articulate, talkative, intelligent, and suffering. I begin by asking people to tell their story. Inevitably the story revolves around what they think is wrong with them. Mary thought bipolar was wrong with her. Almost always the story focuses on a defect within the person.
Mary constructed lives and destroyed them on a regular basis. Each life lasted about 3 years. A life would entail a new job, new relationship, new location—everything new. As time passed and frustration mounted, a threshold was reached in which Mary would quit her job, end her relationship, move—change everything. Often these tumultuous periods were associated with hospitalizations. During these periods of change, she would forget to sleep, eat, or rest. She would receive divine inspiration that would launch her new life. In the hospital, she would be calmed with medication, which she would gradually eliminate.
Mary wanted to focus upon her incredible visions and insights. I was more interested in the moments before she lost control. She described these moments as resembling Alice standing before the mirror, deciding whether or not to walk through. Mary recounted that walking through meant that she was gone. It meant complete commitment to upheaval. I suspected that we had a chance to address that moment of decision making, the moment before she decided to walk through the mirror.
Mary taught me how repetition can bring forth transformation. I asked her to watch The Wizard of Oz everyday for one month. She lived some distance away and could only come once monthly. She wondered why. I answered that I suspected that she and Dorothy had the same dilemma, and that she could learn from Dorothy’s choices and mistakes. Mary accepted this explanation and resolved to watch the movie.
When Mary returned, she reported a transformation. She had realized how wasteful and destructive her actions had been, that she hadn’t really needed to overthrow her life so frequently. She had realized this from watching Dorothy night after night. I could have walked away from the wicked witch, she said, clicked my heels, and stayed in Kansas after all. I didn’t need to go find the wicked witch of the West or struggle with the wizard on his terms. I could have just done something else.
Then we focused on compassionate self-acceptance—the idea that we are here today as a historical creation of all that has come before, that Mary’s past struggles, even her hospitalizations, had contributed to making her the awesome person that she was today.
Mary has stabilized without medication. She has broken her three-year habit and is in her fifth year, now, of work within the same field, with the same boyfriend, without upheaval. She watches Dorothy periodically and continues her daily practice of aikido. She credits this practice and The Wizard of Oz, with changing her life.
Betty Running Bear came to see me in the heights of medications despair. She had been prescribed so many different medications for her bipolar disorder, that it was confusing. She took Thorazine, Lithium, Haldol, Depakote, Prozac, and Klonopin. I wondered how anyone could get so many drugs. Betty was half-Cherokee; half-African in origin. That weekend a yuwipi ceremony was scheduled with a local Native healer from the nearby reservation. Yuwipi means “they tie him up” in Lakota. The ceremony was given to Horn Chips in a vision on the top of Bear Butte (South Dakota) in 1868. It’s purpose was to cure “white people disease.”
In the yuwipi ceremony, a ceremonial space is created in the center of an empty room. A rug is laid upon the floor, covered by a star quilt. Number 10 tin cans are placed at each corner holding enough earth to support a stick carrying a flag for each of the four directions (yellow, red, white, and black). Strung like the ropes of a boxing ring are 405 prayer ties – small pieces of fabric in which tobacco is placed, tied upon a cotton string. The windows of the room and the doors are covered with black plastic until no light can be seen. The ceremony begins with the healer’s hands being tied behind his back, then together, then his arms are tied, and then he is wrapped in a star quilt which is tied around him. When he is completely tied up, the lights go out, the singing and drumming begins, and somehow, sometime during the darkness, the spirits untie him. When the lights go on and he is revealed to be untied, the healing of the sick commences. Lakota people joke that the ceremony must have worked because smallpox has vanished.
Betty presented herself to be healed in the yuwipi ceremony, and the medicine man took her home. He recognized her suffering and had her move in with his sister whose husband had just died. Perhaps he recognized that Betty’s pressured loquaciousness would offset the silent loneliness of a bereaving widow. Being a client of the mental health system, Betty had no where better to go. She lived in a group home in which the major activity was television watching. Connectedness among residents was minimal. She stayed.
The moment Betty moved in, she became part of a large extended kinship network in which life could be completely contained. Between healing ceremonies, family obligations, and social activities, Betty needed no planning for any aspect of her life. She did help the sister with her household activities and was an avid cleaner, which was appreciated in that family. Like a schizophrenic who was taken in by Melvin Grey Fox in Coyote Healing, Betty became another family member. Her life was regulated. One year later, she was on half the medication she had used previously. Two years later it had been reduced to about a fourth. Four years later she was off medication.
An alternate story for healing had been substituted for psychiatry’s biological narrative. In this story, embeddedness in community and participation in ceremony regulates mood quite effectively. The participants in this story would not even be able to enunciate it, for it is their life. Now a member of the medicine man’s family, Betty’s life revolved around ceremony, social obligations, reciprocal family relationships, and service to others. She had learned skills for self-soothing – notably ceremony and ritual, all of which serve to induce the kind of trance state found in hypnosis or meditation. The physical work of housecleaning, cooking for post-ceremony feasts, and the comraderie of Native American women, transformed Betty’s life experience. When life experience changes, I argue, so does genetic expression and physiology. What a different story from the medication for life narrative.
Lauren came with a diagnosis of bipolar II disorder, currently depressed. Her mood alternated between being depressed and feeling mildly euphoric (what some people would call happy). Lauren had been participating in the medication story for more than 15 of her 45 years, but was becoming increasingly unhappy with the side effects of the drugs. She was taking Prozac, Zyprexa, and Depakote, though she had already started to reduce doses on her own.
I worked with Lauren, along with Will (an osteopathic physician) and Amy (a yoga teacher). My perspective with Laura was that other daily practices need to replace the daily practice of taking pills. Yoga is an effective daily practice. My colleague, Amy Weintraub23 has written about yoga for depression and has reviewed its positive benefits. I offered Laura the Chinese point of view that you couldn’t very well evaluate something’s benefit to you until you have done it 100 consecutive days.
Lauren came to us from Louisiana with plans to stay for 10 days. Each day would focus on skill-building, her learning practices (tools) to take home with her and use on a daily basis. We had collectively decided that her tools would be yoga, ceremony, and visualization. Will would help this process along through facilitating structural change in the body (craniosacral therapy, neuro-muscular realignment, and whatever else emerged as potentially useful). I would teach visualization and ceremony.
I began by suggesting we banish the “bipolar word,” as I often do. I suggested we enter into an altered state of consciousness and let “what we were dealing with” define itself. Laura agreed and I began to lead her through (and teach her) meditation/visualization techniques. I begin with focusing on breathing and mindfulness of the present moment and lead into relaxing more and more and letting go of thoughts and preconceptions. After about 15 minutes of this, I suggested that the things that had brought her to me could organize themselves into their own entity. In the dreamtime (trance, altered state, etc.) they could even take form and shape. That form and shape could have a voice and the voice could communicate.
The ensemble of everything tied to the bipolar label, and probably other things, too, began to coalesce into a shape of a character who called himself “Take It Easy.” I double checked for gender, and he was definitely male.
I gave suggestions to go backward on the river of time, using Lauren’s desired mode of river travel—steam, raft, canoe, powerboat, whatever—until the driver of her craft landed her at the place in the stream of time in which “Take It Easy” had been conceived. We journeyed backward by seven generations to Lauren’s ancestral home in Scotland. Images arose of children being beaten to conform. The Scoth Protestant ethic was to “beat it out of them.” A story emerged of the many generations for whom “taking it easy” meant conforming, stuffing originality, protest, opposition, and other undesirable traits. Overthrowing “Take It Easy” meant rebelling against self-criticism for the perspective from which the criticism could be made was a family culture of conformity which didn’t match the story Laura wanted to tell about herself. She wanted to be more outlandish, more outspoken, even flamboyant. Being hypomanic allowed her to flirt with these qualities, but always in a way in which she could feel shame later and be criticized.
We developed a series of practices to use to oppose self-deprecation. These included the daily yoga, daily ceremony, and daily meditation. My theme for her was, “It’s all good. It’s all fine. You don’t have to change. In fact, don’t do anything.” This was a recipe against the self-critical perspective.
Mary had found a healing temple to join for community. Laura was encouraged to gather together everyone she knew who was interested in healing and to meet on a weekly basis to form a healing community. She has done this and it has been an important part of her maintaining her harmony and balance.
This is in stark contrast to the SMI (seriously mentally ill) clients who populate the mental health care system and who remain isolated without support, despite their plethora of medications.
In these three stories, people have broken free from the biological psychiatric narrative. They spend no money on drugs. Their daily “meds” include yoga, meditation, prayer, all embedded within community. They have alternate stories which would be unnoticed by the dominant paradigm. If noticed these stories would be dismissed as “quacky,” so rare as to be unimportant and ungeneralizable to any significant number of people, or evidence of misdiagnosis (since bipolar disorder is, by definition, life-long).
These stories represent an “underbelly” of American psychiatry, but perhaps also a repository of potential transformation, and help for the 50 percent of non-responders – an alternative that has other solutions in addition to medication or besides medication, as the case warrants. This solution would perplex modern psychiatry, for it is post-modern. Each individual has his or her own path to less suffering. The solution comes develops from the affected community and not the professional expert.
[*] Mania is operationally defined in the DSM-IV-TR in terms such as grandiosity, hyper-religiosity, lack of need for sleep, boundless energy, excessive spending, suspension of good judgment over the consequences of one’s actions, loquaciousness, inability to stop a behavior once started, and an inability to modify one’s plans based on feedback from the environment. Hypomania is defined more as persistent irritability or mild euphoria—in other words, a lessened and harder to recognize form of mania that oscillates with periods of depression. Both types of bipolar patients spend more time being depressed than high or irritable.
 American Psychiatric Association. Practice Guidelines for the treatment of patients with bipolar disorder (revision), Available at www.psych.org/psych_pract/treatg/pg/prac_guide.cfm . Accessed August 12, 2004.
 Literature in the Reader: Affective Stylistics, Stanley Fish, New Literary History, Vol. 2, No. 1, A Symposium on Literary History (Autumn, 1970) , pp. 123-162
 Interpretation of Cultures, Clifford Geertz, New York: Basic Books, 2000.
 Fanthum E, Rummel E. The Collected Works of Erasmus. Toronto: University of Toronto Press, 1989
 Hall M, Ratner C, Riebner R. The Collected Works of L.S. Vygotsky, Volume 5. Child Psychology. Boston: Kluwer Academic/Plenum Publishing, 1999
 Vygotsky, op. cit., pp 169-174
 Zrehen S, Kitano H, Fujito M. Learning in Psychologically Plausible Conditions: The Case of the Pet Robot. In Pfeiffer, Blumberg, Meyer, Wilson. From Animals to Animats 5: Proceedings of the Fifth International Conference on Simulation of Adaptive Behavior. Cambridge, MA: MIT Press, 1998
 Champandard A. AI Game Development: Synthetic Creatures with Learning and Reactive Behavior. New York: New Riders, 2003.
 Williams M. Wittgenstein, Mind, and Meaning: Towards a Social Conception of Mind. New York: Routledge, 2002, pp. 265-274.
 Burns WD, Klawunn W. The Web of Caring: An approach to accountability in alcohol policy. http://phoenix.edc.org/hec/pubs/theorybook/burns.pdf , Last accessed 19 July 2006.