Michael J. Prince, B.A., Psychology

Psychophysiological Articles


Recurrent Depression

Charles L. Bowden MD


This year's Annual Meeting of the American Psychiatric Association provides an unparalleled opportunity for updating clinical knowledge in psychiatric disorders and related mental health issues. This is, in part, due to the size of the program -- with nearly 19,000 registrants in attendance -- as well as to the careful attention devoted to the range of fields to be studied and the provision of multiple venues to facilitate learning. One important program held during the meeting is the Review of Psychiatry. The American Psychiatric Press, Inc, commissions and publishes texts that assimilate the latest information in a particular field. These texts then form the basis for symposia presented by leading psychiatrists. One of 4 Review sessions focused on the treatment of recurrent depression.

Recurrent Depression and Mania

Reviewing the evidence on the severe morbidity and the problems of inadequate recognition points out the need for early intervention of recurrent depression and mania according to John Greden, MD,[1] of the University of Michigan, Ann Arbor. A recent survey of members of the National Depressive and Manic-Depressive Association illustrates the gravity of the problem of accurate identification. The average patient with bipolar disorder had seen 4 physicians before the correct diagnosis of bipolar disorder was made. The proportion of persons who sought help within the first year of symptoms increased from 30% to 36% from 1992 to 2000. However, misdiagnosis, or no diagnosis at first evaluation, improved minimally over the 8-year period. Diagnoses such as depression accounted for more than half of all misdiagnoses, with an incorrect diagnosis being significantly more common among women than men (68% vs 43%). By contrast, schizophrenia was more often misdiagnosed in men than women (28% vs 14%). The patients surveyed viewed a lack of understanding of bipolar disorder among mental health professionals as the primary reason for the misdiagnosis (60%).

Dr. Greden indicated that improved recognition of persons at high risk for recurrent depression could be accomplished by keeping in mind several relatively simple concepts. Persons who have had 3 or more depressive episodes are at increased risk of relapse. Also, those who are at high risk for recurrent depression are:

Sheila Marcus, MD, also of the University of Michigan, reviewed the impact of recurrent depression on women across their lifespan.[2] Women have approximately twice the rate of recurrent unipolar depression as do men. Women treated in the University of Michigan clinical programs who had depressive disorders had overall healthcare costs that were approximately triple the costs incurred by nondepressed women. Berndt and colleagues,[3] found that depressed women had 12% to 18% lower incomes than their nondepressed counterparts. There are also some differences in the common features of depression between adolescent girls and boys. Boredom is more common than depressed mood, and girls have more physical complaints than boys. Some components of depression in women are likely related to hormonal factors. The follicular phase of the menstrual cycle, characterized by relatively high levels of estrogen, is associated with increased well-being and optimism.

Early treatment of depression is needed. Women who became depressed while pregnant, but were untreated, gave birth to lower-weight infants who had more neonatal complications than control newborns. Postpartum depressive states often start within hours of delivery. Attention to history can be helpful in anticipating postpartum disorders, as prior postpartum episodes are common in women at higher risk.[4]

Despite the importance of early treatment, little systematic treatment research has been conducted. Selective serotonin reuptake inhibitors (SSRIs) have advantages of relative safety. Earlier work by Rubinow and colleagues[5]

suggests that estrogen supplementation may help some symptoms of depression in women. Herbal remedies are now being used more than ever for the treatment of depressive disorders. However, a recent well-designed study of St. John's wort in major depression found it no more effective than placebo.[6]Psychotherapy is extensively employed to treat depression.[7] Keller and colleagues[8] recently reported that the combination of cognitive behavioral psychotherapy and the antidepressant nefazodone resulted in 20% higher rates of remission vs those achieved with either treatment alone.

Novel Nondrug Treatments

Mark George, MD, of the Medical University of South Carolina,[9,10] reported on 2 novel nondrug treatments for depression that are currently under investigation. Transcranial magnetic stimulation induces an electrical current in the brain, but, unlike electroconvulsive therapy, causes no seizure and requires no anesthesia. Early studies have yielded promising reports of improvement, including studies that have endeavored to compare results with a sham-treatment group.[9] A second technique is vagus nerve stimulation.[10] This technique takes advantage of the fact that 80% of fibers in this important nerve are afferent. First employed in treatment-resistant epilepsy, it has been tested in patients with resistant depression since 1998. The episodic stimulation of the vagus using an implanted electrode results in changes in blood flow in the medial temporal cortex and hippocampus. Early studies indicate that initial improvement is retained for up to 1 year following a 12- week period of treatment.[11]

Prevention of Depression Recurrence in Patients With Bipolar Disorder

Charles Bowden, MD,[12] of the University of Texas Health Science Center at San Antonio, discussed the prevention of recurrences in persons with bipolar disorder. Treatment-resistant depression in women is often a bipolar II disorder, whereas full depressive episodes are accompanied by milder, sometimes not functionally impairing partial manic states. Bipolar II disorder occurs mostly in women, despite overall equal rates of bipolar disorder in women and men. Also, rapid cycling bipolar disorder occurs predominantly in women. Practically all persons with bipolar disorder need preventive treatment. Mood stabilizers, principally divalproex and/or lithium, are the foundations of most preventive treatment.

In a recent study of divalproex and lithium vs placebo, both drugs delayed recurrence of mania, with approximately a doubling of time to a recurrent manic episode with divalproex. Divalproex was better tolerated than lithium. Bowden reported on several studies that indicate that somewhat lower maintenance doses of both drugs than the dose ranges used to treat acute mania are both better tolerated and as likely to be effective in prevention.[13,14] These 2 drugs, as well as carbamazepine and antipsychotic drugs, which are also often used for prophylaxis, are generally more effective on the manic side of the illness than the depressive side.

Two recent studies of lamotrigine in prevention of recurrences both found it superior to placebo in long-term treatment of bipolar patients, including the generally difficult-to-treat rapid cycling group.[15,16] The second study,[16] found both lamotrigine and lithium to be superior to placebo in delaying time to a recurrence. The lamotrigine benefit was delaying time to recurrence of depression, but not of mania. In contrast, the lithium benefit was delaying time to recurrence of mania, but not of depression. Both studies indicate an effectiveness of lamotrigine principally on the depressive component of bipolar disorder, which, if confirmed, could be complementary to the relatively good manic preventive properties but weak depression prevention effects of other mood stabilizing agents.


Session I of the Review of Psychiatry made a compelling case both for the health burden caused by recurrent depression and bipolar disorder as well as for the remarkable treatment advances for these conditions. Earlier recognition and intervention, therefore, is an achievable and urgent public health issue. Earlier recognition requires attention to more subtle cues of the illness. If we wait until a person has been hospitalized for depression or a full manic episode, we are likely to have missed several years of depression, or mild dysfunctional impulsivity. Also, the disease may become more difficult to treat. Research by Dr. Greden[17] and others indicates that the stress glucocorticoid cascade may result in permanent structural changes in brain. Stressful events from the environment can reduce serotonergic activity that, in turn, increases cortisol-releasing hormone and cortisol, which, over time depress the expression of neurotrophins thought to be important in aiding neuronal integrity. The reduced protective effect of neurotrophins appears to contribute to development of atrophic regions in the hippocampus.[18]


Greden JF. Recurrent depression and mania: their overwhelming burden. Program and abstracts of the 154th Annual Meeting of the American Psychiatric Association; May 5-10, 2001; New Orleans, Louisiana. Review of Psychiatry Session I-1

Marcus S, Flynn HA. Recurrent depression in women through the lifespan. Program and abstracts of the 154th Annual Meeting of the American Psychiatric Association; May 5-10, 2001; New Orleans, Louisiana. Review of Psychiatry Session I-2

Berndt ER, Koran LM, Finkelstein SN, et al. Lost human capital from early-onset chronic depression. Am J Psychiatry. 2000;157:940-947.

Nielsen Forman D, Videbech P, Hedegaard M, Dalby Salvig J, Secher NJ. Postpartum depression: identification of women at risk. BJOG. 2000;107:1210-1217.

Schmidt PJ, Nieman L, Danaceau MA, et al. Estrogen replacement in perimenopause-related depression: a preliminary report. Am J Obstet Gynecol. 2000;183:414-420.

Shelton RC, Keller MB, Gelenberg A, et al. Effectiveness of St John's wort in major depression: a randomized controlled trial. JAMA. 2001;285:1978-1986.

Boland RJ. Chronic and recurrent depression: pharmacotherapy and psychotherapy combinations. Program and abstracts of the 154th Annual Meeting of the American Psychiatric Association; May 5-10, 2001; New Orleans, Louisiana. Review of Psychiatry Session I-3

Keller MB, McCullough JP, Klein DN, et al. A comparison of nefazodone, the cognitive behavioral-analysis system of psychotherapy, and their combination for the treatment of chronic depression. New Engl J Med. 2000; 342:1462-1470.

George MS, Nahas Z, Molloy M, et al. A controlled trial of daily left prefrontal cortex TMS for treating depression. Biol Psychiat. 2000;48:962-970.

Mahas Z, Sackeim HA, Rush AJ, et al. Vagus nerve stimulation (VNS) in outpatients with treatment-resistant depression. Program and abstracts of the 154th Annual Meeting of the American Psychiatric Association; May 5-10, 2001; New Orleans, Louisiana. Symposium NR 394.

Prater JF. Recurrent depression with vagus nerve stimulation. Am J Psychiatry. 2001;158:816-817.

Bowden CL. Prevention of recurrences in bipolar patients: the best of the old and the new. Program and abstracts of the 154th Annual Meeting of the American Psychiatric Association; May 5-10, 2001; New Orleans, Louisiana. Review of Psychiatry Session I-4

Bowden CL, Calabrese JR, McElroy SL, et al. A randomized, placebo-controlled 12-month trial of divalproex and lithium in treatment of outpatients with bipolar I disorder. Arch Gen Psychiatry. 2000;57:481-489.

Bowden CL, Gonzales CL. Prevention of recurrences in patients with bipolar disorder In: Greden JF, ed. Treatment of Recurrent Depression. Review of Psychiatry, Vol 20, (No. 5). Washington DC: American Psychiatric Publishing; 2001;81-101.

Calabrese JR, Suppes T, Bowden CL, et al. A double-blind, placebo controlled, prophylaxis study of lamotrigine in rapid cycling bipolar disorder. J Clin Psychiatry.

Calabrese JR, Bowden CL, McElroy SL, et al. Spectrum of activity of lamotrigine in treatment-refractory bipolar disorder. Am J Psychiatry. 1999;156:1019-1023. Greden JF. Clinical prevention of recurrent depression, In: Greden JF, ed. Treatment of Recurrent Depression. Review of Psychiatry, Vol 20, (No. 5). Washington DC: American Psychiatric Publishing; 2001;143-170.

Sheline YI, Wang PW, Gado MH, Csernansky JG, Vannier MW. Hippocampal atrophy in recurrent major depression. Proc Natl Acad Sci U S A. 1996;93:3908-3913.

Antidepressant Treatment and the Biology of Depression

Jerrold F. Rosenbaum MD Disclosures


A state-of-the-art review of the biology of depression needs to include an understanding of antidepressant mechanisms of action and offer new data as well as current concepts on the biology of depression, imaging studies of the depressed brain before and after treatment, receptor effects of the new generation antidepressants, and stress and neurogenesis. One such review was presented during this year's Annual Meeting of the American Psychiatric Association.

Brain Imaging and Depression

Over the years, depression has been viewed in many ways. Helen S. Mayberg, MD, of the University of Toronto,[1] stated that it is important to consider the historical shifts from thinking of depression existentially, to understanding the phenomenon as a syndrome with a categorical classification (ie, Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition [DSM-IV]), to taking a more brain-based view. When viewed in terms of dysregulated brain function, symptoms cluster into categories of mood, somatic, motor, and cognitive regulation. This view allows for identification of different regions which make up the disorder.

Dr. Mayberg described the importance of the study of brain function and described functional imaging strategies including lesion mapping, pharmacologic challenge, and behavioral provocation. All of these strategies provide a window on the brain's resting state, blood flow and metabolism, disease markers, and clinical feature correlates.

In the 1930s, it was realized that stimulating the surgically exposed brain in specific areas could evoke mood and emotional sensations. At the same time, surgical destruction of these same brain regions could alter negative mood. The results focused attention on the orbitofrontal cortex and the frontal, temporal, and basal ganglia as likely components of a depression circuit.

Dr. Mayberg looked at individual positron emission tomography (PET) scans of brain-glucose metabolism in depressed and nondepressed patients with Parkinson's disease, Huntington's disease, and primary unipolar depression.[2] The data show that depressed patients have striking bilateral and frontal hypometabolism in the inferior frontal lobe when compared with matched nondepressed patients. Imaging studies of the anterior insula, a deep limbic structure, also had shown hypometabolism in unipolar depressed patients but hypermetabolism in bipolar patients.

A study that imaged baseline glucose metabolism in depressed male inpatients then compared them with images taken after blind randomization to either fluoxetine or placebo treatment at 1 and 6 weeks.[3] At 6 weeks, dramatic increases in the previously hypometabolic frontal lobe relative to baseline were evident. There was also an increase in metabolism in the dorsal brainstem at both 1 and 6 weeks, whereas in the posterior cingulate, there was suppression of metabolic activity at week 1 that switched to become hyperactive at week 6. Additionally, fluoxetine increased metabolism in the hippocampus in the short term, but suppressed it in the long term by week 6. Hypometabolism seen in the frontal, parietal, and cingulate lobes and insula differentiated patients from controls.

When looking at brain images of treatment responders vs nonresponders, the data fit this idea of impact on both neocortical and limbic regions, according to Dr. Mayberg. Improved frontal hypometabolism is seen in responders. Increased posterior cingulate metabolism was not, however, seen in nonresponders. She also suggested that the critical changes for clinical improvement are not only normalization of the frontal activities but also suppression of the limbic areas, especially area 25.

Major depression can be understood as a dysfunctional distributed network or circuit within the brain, not based in 1 specific region, and treatment may require a resetting or balancing of the circuit. The latter could in theory be accomplished by interventions at different sites within the circuit.

Understanding What Antidepressants Do

Pierre Blier, MD, PhD, of the University of Florida at Gainesville discussed the biological effects of antidepressant molecules on neurotransmitters and receptors. He began by looking closely at selective serotonin reuptake inhibitors (SSRIs) and how these drugs impact the serotonin system.[4] Dr. Blier reviewed the regulation of the brain's serotonin system. The cell bodies of the serotonin neurons are located in the brainstem, mostly in the median and dorsal raphe nuclei. Projections from these two nuclei reach multiple brain structures. He illustrated the activities of the serotonin (5-HT)-1A autoreceptor and the 5-HT uptake transporter. He explained that when the 5-HT-1A receptor is stimulated by an increase in the amount of serotonin, it will decrease the firing activity of the serotonin neurons.

When an SSRI is introduced, initially there is an increase in serotonin levels at the cell body due to blockade of the 5-HT uptake transporter. The 5-HT-1A autoreceptors are stimulated and decrease the firing frequency of the neuron. Yet, at the level of the synaptic bouton, there is no marked increase in serotonin availability, even though the transporters are inactive, because the firing activity reaching the terminals is decreased. However, after a few weeks of treatment with the SSRI, the autoreceptors are desensitized and allow a recovery of firing activity. At the synaptic bouton, reuptake is still blocked, firing activity is restored, and there is more serotonin because the terminal autoreceptors are desensitized. This allows 5-HT neurons to release their transmitters without negative feedback action in the presence of reuptake inhibition. The time necessary to obtain these adaptive changes is consistent with the delayed onset of action of SSRIs in major depression.

What is the influence of serotonin neurons on norepinephrine (NE) neuron firing activity? It is possible to answer this question by lesioning the serotonin neurons and then observing the norepinephrine neuron firing activity. Results show that serotonin has an inhibitory action in the locus coeruleus, as the experiment results in a 75% increase in NE neuron firing. If one then attempts to potentiate 5-HT activity by giving an SSRI, NE activity decreases. Dr. Blier suggested that the observation of sustained SSRI treatment leading to a gradual decrease of the spontaneous firing rate of NE neurons may in part explain the beneficial action of SSRIs in panic disorder, a condition associated with a hyperadrenergic state.

Dr. Blier also investigated dopamine, norepinephrine, and serotonin neuron activities during sustained administration of bupropion. According to his hypothesis, if bupropion blocks NE, dopamine, and 5-HT reuptake transporters, firing activity should decrease due to an excess activation of the cell body autoreceptors. He found that subacute bupropion treatment inhibits NE neuron firing by enhancing the synaptic availability of NE, which in turn enhances 5-HT firing by activating excitatory alpha-1-adrenoreceptors on 5-HT neurons. Further study has led Dr. Blier to believe that bupropion acts as an NE releaser (as opposed to an NE reuptake blocker) because the NE reuptake blockers, desipramine and reboxetine, do not alter the firing of 5-HT neurons.[5]

Mirtazapine is selective for the NE alpha-2-autoreceptor but causes marked changes in serotonin action. Mirtazapine is an alpha-2-adrenergic antagonist. Systemic administration of mirtazapine blocks alpha-2-receptors at the cell body of the NE neuron and at the axon terminals, which will release more NE into the synapse. NE neurons project to the raphe nucleus and 5-HT availability in the raphe increases as a result. Long-term administration of mirtazapine results in increased NE firing activity in the locus coeruleus and increased firing of 5-HT neurons in the raphe.[6]

Depression and the Brain at the Neuronal Level

According to Eric Nestler, MD, PhD, of the University of Texas Southwestern Medical Center, Dallas,[7] it is important to review intraneuronal signal transduction. Over the past 7 to 8 years, understanding of synaptic transmission has advanced significantly. In particular, neuroscientists have discovered that multiple types of protein phosphorylation cascades inside nerve cells ultimately influence gene transcription. Neuronal signals are communicated through not only neurotransmitters but also other types of signaling molecules called neurotrophic factors (also called nerve growth factors), of which there are many types. Dr. Nestler focused on the effect of antidepressants on the second messenger cyclic adenosine monophosphate (cAMP) and the pathway by which it leads to changes in gene expression. One product expressed after antidepressant treatment is brain-derived neurotrophic factor (BDNF).

Neurotrophic factors are peptides that support the growth, differentiation, and survival of neurons. In contrast to peptide neurotransmitters, neurotrophic factors produce their effects by activation of tyrosine kinases. Research in this area suggests that the sustained perturbation of monoamine systems leads to cAMP pathway adaptations that, through the regulation of target genes that regulate neural plasticity, underlie antidepressant therapeutic action.

Dr. Nestler proposed gene regulation as one mechanism that might explain the delayed effects of antidepressant drugs. According to this view, repeated exposure to a drug such as an antidepressant would repeatedly perturb the intracellular messenger cascades and regulate the expression of other genes. For example, these cascades regulate transcription factors and proteins that bind to the regulatory region of genes and control the rate and time at which target genes are expressed.

As a drug enters the brain, it immediately binds to and inhibits the transporters for serotonin and/or norepinephrine, thereby increasing the availability of these neurotransmitters at certain synapses. According to this model, repeated antidepressant administration leads to increased production of the second messenger, cAMP, increased activation of protein kinase A, and activation of the transcription factor cAMP response element binding protein (CREB); this activation in turn triggers changes in the expression of CREB-activated target genes such as BDNF.

Dr. Nestler and his colleagues investigated whether these hypothesized changes in cAMP, protein kinase A, and CREB levels were actually occurring. With the use of in situ hybridization to measure levels of CREB mRNA in the rodent hippocampus, they found dramatic induction of CREB mRNA after courses of chronic electroconvulsive seizures, fluoxetine, and tranylcypromine as compared with controls.[8] These data support the idea that repeated exposure to antidepressants leads to an upregulation in the cAMP cascade in certain nerve cell areas of the hippocampus.

The investigators then hypothesized that independent activation of the cAMP pathway (without affecting 5-HT or NE levels) should also be antidepressant. They tested this idea in animals by administering two phosphodiesterase inhibitors (enzymes that block the breakdown of cAMP)[9] and found that prolonged administration of each phosphodiesterase inhibitor alone increased CREB levels. These findings support the theory that enhancement of the cAMP pathway in the hippocampus is correlated with an antidepressant response. To draw direct causal relationships between CREB activity and the cAMP pathway to a real depression-like phenotype, Dr. Nestler and his colleagues used the method of viral gene transfer. In this method, a virus coupled to CREB is injected into a rodent hippocampus and then the animal is observed while it performs a forced swim test. (In a forced swim test, clinically effective antidepressants increase the period that a rodent will struggle before giving up.) When levels of CREB were increased, the results of the behavioral test mimicked antidepressant action.

Through what target genes could CREB be acting to cause these effects? Clinical evidence suggests a role for neurotrophic factors in depression. Some depressed patients are hypocortisolemic. Neuroimaging data showed small decreases in the total volume of the hippocampus in depression and posttraumatic stress disorder.

When an animal is acutely or chronically stressed, the hippocampal level of the growth factor BDNF dramatically decreases; antidepressants (but not other psychotropic medications) have the opposite effect. Stress by itself decreased BDNF but prior exposure to antidepressant completely blocked the ability of stress to produce that effect.

In sum, Dr. Nestler and his colleagues proposed a model that considered both genetic and nongenetic vulnerabilities, where stress reduces levels of BDNF in the hippocampus and can lead to increased vulnerability to a depressive episode. A chronic course of antidepressant treatment, however, through effects on the 5-HT and NE stems, leads to the regulation of target genes that may change the numbers of neurotransmitter receptors and the levels of CREB and BDNF. Thus, enhanced trophic support of the hippocampus may help restore the nerve cells to a healthier state.

Biology of Depression: A New Look

Over the next decade, psychiatric researchers expect to discover new vulnerability genes that influence the development of depression and other neuropsychiatric disorders and then use this information to develop novel treatments. Dennis S. Charney, MD, of the National Institute of Mental Health, Bethesda, Maryland, believes that new science ultimately will allow practitioners to look at the interaction between genes that confer vulnerability and resilience to depression and the environmental risk and protection factors that combine to generate the observed phenotype.[10] The current diagnostic criteria are not informative for this task area; they have not been useful in understanding the role of genes or interactions between genes and the environment, and there are no biological markers that map onto the diagnoses of the DSM-IV.

He emphasized the important contribution that both early life and recent stresses make to the risk factors for a major depressive episode. Data indicate that early stress can exert major and long-lasting effects on the neuropeptide system, including corticotropin-releasing hormone (CRH). In 1study, when mice were separated from their mothers for a few hours daily during their first 14 days of life, researchers observed a persistent elevation in CRH not seen in control animals.[11] In another study, monkeys that had been stressed as infants with the unpredictable availability of food to their mothers also showed elevated CRH levels.[12] Neuroimaging studies found that stress-related disorders are associated with reduced hippocampal volume, which may be related to elevation in plasma cortisol levels and the duration or severity of the stress or depressive episode. Reduced hippocampal volume in patients suffering from major depression are replicated findings.[13] Dr. Charney mentioned that the replication is especially important, given the heterogeneity of major depressive disorder. These data have generated interest in the pharmaceutical industry and it is speculated that the next years will bring increased study of CRH antagonists for the treatment of depression and other stress-related disorders.

Other depression studies suggest that there may be loss of neurons and glial cells in patients who suffer from major depression.[14] Three patterns of cellular changes have been noted in major depression: cell loss in the subgenual prefrontal cortex, cell atrophy and loss in the dorsolateral prefrontal and orbitofrontal cortexes, and increased cell numbers in the hypothalamus and dorsal raphe nucleus. Dr. Charney noted that these postmortem data are consistent with concepts that Dr. Mayberg[1] had spoken of earlier in the symposium -- ie, there are important reciprocal connections between subcortical structures such as the hippocampus and amygdala and the prefrontal cortex. Through a combination of imaging, structural, and postmortem studies, it might be possible to define a more robust functional neuroanatomy of depression.

Dr. Charney revisited the study findings of hippocampal neurogenesis that Dr. Nestler had spoken of and tied those to Dr. Blier's explanation that antidepressant drugs enhance the function of the 5-HT system and the particularly important 5-HT-1A receptor. Serotonin itself stimulates the production of hippocampal granular cells, and its mechanism occurs with the 5-HT-1A receptor, thereby bringing together the concepts of Drs. Blier and Nestler to form a more unifying hypothesis involving both monoamines and neurotrophic effects. Enhancement of 5-HT-1A function may have profound treatment effects.

Dr. Charney admitted that the mechanisms of these structural abnormalities are not understood. Some future targets of research that reach beyond monoamines are the CRH system and the mechanisms of mood stabilizers, phosphokinase C, phosphodiesterase inhibitors, and the n-methyl-D-aspartic acid receptor. He predicted that several new classes of antidepressants will emerge over the next decade.


Mayberg HS. The depressed brain image. Program and abstracts of the 154th Annual Meeting of the American Psychiatric Association; May 5-10, 2001; New Orleans, Louisiana. Industry Symposium 5A.

Mayberg HS, Liotti M, Brannan SK, et al. Reciprocal limbic-cortical function and negative mood: converging PET findings in depression and normal sadness. Am J Psychiatry. 1999;156:675-682.

Mayberg HS, Brannan SK, Tekell JL, et al. Regional metabolic effects of fluoxetine in major depression: serial changes and relationship to clinical response. Biol Psychiatry. 2000;48:830-843.

Blier P. What do antidepressants do? Understanding antidepressant molecules, neurotransmitters, and receptors. Program and abstracts of the 154th Annual Meeting of the American Psychiatric Association; May 5-10, 2001; New Orleans, Louisiana, Industry Symposium 5C.

Frazer A. Norepinephrine involvement in antidepressant action. J Clin Psychiatry. 2000;61(suppl 10):S25-S30.

Haddjeri N, Blier P, de Montigny C. Effects of long-term treatment with the alpha 2-adrenoceptor antagonist mirtazapine on 5-HT neurotransmission. Naunyn Schmiedebergs Arch Pharmacol. 1997;355:20-29.

Nestler EJ, Duman RS. Healing the depressed brain: signal transduction and neural plasticity. Program and abstracts of the 154th Annual Meeting of the American Psychiatric Association; May 5-10, 2001; New Orleans, Louisiana, Industry Symposium 5D.

Chen AC, Eisch AJ, Sakai N, et al. Regulation of GFRalpha-1 and GFRalpha-2 mRNAs in rat brain by electroconvulsive seizure. Synapse. 2001;39:42-50. Nibuya M, Nestler EJ, Duman RS. Chronic antidepressant administration increases the expression of cAMP response element binding protein (CREB) in rat hippocampus. J Neurosci. 1996;16:2365-2372.

Charney DS. The neurobiology of mood disorders. Program and abstracts of the 154th Annual Meeting of the American Psychiatric Association; May 5-10, 2001; New Orleans, Louisiana, Industry Symposium 5B.

Anisman H, Zaharia MD, Meaney MJ, Merali Z. Do early-life events permanently alter behavioral and hormonal responses to stressors? Int J Dev Neurosci. 1998;16:149-164.

Lyons DM, Yang C, Mobley BW, Nickerson JT, Schatzberg AF. Early environmental regulation of glucocorticoid feedback sensitivity in young adult monkeys. J Neuroendocrinol. 2000;12:723-728.

Sheline YI, Gado MH, Price JL. Amygdala core nuclei volumes are decreased in recurrent major depression. Neuroreport. 1998;9:2023-2028.

Rajkowska G. Postmortem studies in mood disorders indicate altered numbers of neurons and glial cells. Biol Psychiatry. 2000;48:766-777.

Emerging Trends in Understanding Posttraumatic Stress Disorder

Martin L. Korn MD


Posttraumatic stress disorder (PTSD) is characterized by a traumatic event that results in long-standing psychological, social, and biological sequelae. Because of the occurrence of a defined external event(s) that initiated the disorder, PTSD allows us to study the process by which the environment influences the individual in dramatic ways. PTSD is therefore an ideal vehicle to study the integrative process taking place between external and internal processes. Several symposia at the 154th Annual Meeting of the American Psychiatric Association were devoted to the integration of these perspectives.

PTSD: Definitions and Epidemiology

A cluster of symptoms resulting from exposure to a traumatic stressor characterizes the syndrome of PTSD. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM IV-TR) added a new component to the "stressor" component. In addition to characterization of the types of appropriate stressors (eg, experienced or witnessed a life-threatening event or serious injury criterion A1), a second criterion was added, which required that the trauma be viewed with "intense fear, helplessness, or horror" (criterion A2). The associated symptoms include intrusive thoughts, physiologic arousal, physical or psychological avoidance, and cognitive or memory disturbances. Naomi Breslau, PhD,[1] of the Henry Ford Health System , Detroit, Michigan, discussed the prevalence of PTSD, indicating a wide variance depending on the study population. In an earlier study, Breslau and colleagues[2] reported on the lifetime incidence of PTSD, derived from the 1996 Detroit Area Survey of Trauma, and found that the conditional risk of PTSD following a traumatic event was 9.2%. Assault carried the highest risk of subsequent PTSD (20.9%), although the most common event that precipitated the disorder was the sudden unexpected death of a loved one (31%). This event was experienced by 60% of the sample and resulted in a 14.3% risk of PTSD. Approximately 90% of people have experienced a significant trauma in their lives and males experience more traumatic events than women. Women, however, are more likely to develop PTSD.

Individual Variation in PTSD

A basic premise underlying the DSM criteria is the "dose-response" model, which was reviewed by Marilyn Bowman, PhD,[3] of Simon Fraser University, Burnaby, British Columbia, Canada. The dose-response model considers the most important factors relating to the development of subsequent a posttraumatic syndrome to be the extent of the trauma and the proximity of the event to the individual. However, only a minority of exposed individuals actually go on to develop the syndrome of PTSD. Thus, individual differences and predispositions clearly play a major role.[4,5]

One of the most powerful psychological predictors influencing development of PTSD is "neuroticism." This character trait is best described as the tendency to see the world in a negative manner. Neuroticism has been shown to have approximately a 50% rate of heritability and to be relatively stable over time. It appears to act as a filter, biasing the individual toward the selection of negative events from the environment.

Several other predisposing factors have been identified. These include beliefs about one's vulnerability or resilience, relationships to family and society, perceived meaning and consequences of the trauma, the sense of the locus of control, religiosity, intelligence, as well as other factors. In addition to coloring one's perception of the meaning of event, these characteristics alter emotional arousal, regulate the interpretation of danger signals, and alter the ability to regulate the stress response. Comorbid psychiatric disorders also increase the probability of the development of PTSD. Dr. Joel Paris, MD,[5] of Sir Mortimer B. Davis Jewish General Hospital, Montreal, Canada, echoed the importance of predisposing factors and social support in his presentation. Veterans suffering from the Gulf War Syndrome had high rates of neuroticism, however, caution must be exercised in the interpretation of these data, considering the potential for real organic factors.

Unfortunately, many treatment initiatives in general use today are based on the dose-response model and tend to ignore individual variation. Intensive immediate posttraumatic counseling is based on the mistaken assumption that PTSD is a normative response to the traumatic event. However, talking through the problem and reliving the traumatic event may increase anxiety in some individuals and thereby enhance the probability that PTSD will occur.

Sociocultural Aspects of PTSD

In addition to the psychological differences affecting the response to traumatic events, sociocultural factors are also an important determining factor. The cultural milieu provides a context of dynamic meanings that filters the experience, analysis, and interpretation. According to Eric Jarvis, MD,[6] of McGill University, Montreal, Canada, psychological illnesses, although having biological bases, are also strongly influenced by the cultural filter and therefore should be considered sociobiological phenomena. This is especially true of psychiatric disorders such as PTSD.[7] Cross-cultural studies of psychiatric disorders are difficult to conduct and estimated rates of PTSD vary widely. Reported rates of PTSD in Asians immigrating into the United States, for example, vary from 3% to 85%. Cambodian refugees have a collective view of self and downplay individual reactions, therefore, the individually oriented DSM definitions cannot be reliably applied to this variant culture. Some researchers, however, have contended that the PTSD diagnosis can be reliably applied to this population.[8] Others have provided evidence for cultural differences in physiologic responses.[9] Cultural factors may modify the response to trauma. Many cultures believe strongly in fate or accept the role of divine intervention. These beliefs may lead to an acceptance of the traumatic experience. Some Western techniques emphasize the need for cathartic verbalization as a means of resolution of a trauma. This may be contraindicated in a culture that views discussion of a traumatic event as inappropriate. For example, acknowledgement of an incident of rape may be socially inappropriate in some cultures as it makes the victim unmarriageable.

Roberto Lewis-Fernandez, MD,[10] reported that Latinos had higher rates of PTSD after the Vietnam war compared to both blacks and whites.[11] Although this may be associated with factors such as amount of exposure to combat or social isolation due to minority status, part of the difference has been attributed to different rates of symptom reports among subgroups of Latinos. For example, Puerto Ricans tend to report more symptoms in response to trauma than do Mexican Americans.[11] Thus, variance in expressive styles may lead to artificially elevate rates of PTSD. Alternatively, they may actually have higher posttraumatic effects. Further cultural analyses must be performed to assess the validity of the various hypotheses.

Therapeutic Formulations

Cognitive-behavioral interventions have become the 1 of the primary means of treatment of PTSD and are central to our theoretical understanding of this disorder. Part of the reason for this is the stress on experimental verification that has lead to widespread acceptance of the model and the technique. Results of clinical studies indicate that therapeutic effects can last several years.[12] Cognitive-behavioral approaches emphasize the learning aspects of the disorder and use exposure and/or desensitization procedures as the mainstay of treatment.[13] Dysfunctional cognitions supporting the traumatic memories should be addressed.[14]

Alternatively, according to Arieh Shalev, MD,[15] of Hadassah University, Jerusalem, Israel, psychodynamic formulations have lost their appeal for the scientific and academic community. This is due to difficulties in operationalizing the psychotherapeutic techniques and placing them into manual-driven algorithms. There has also been a traditional resistance among practitioners toward reducing the discipline to simplified formulations. Despite this, psychodynamic treatments still are widely practiced in the treatment of victims of trauma.[16]

Some psychodynamic formulations appear to be supported, in part, by the trauma literature. Freud's original conceptions were based on the psychic trauma of the Oedipal conflict, a dynamic that worked to shape much of the individual's personality but remained largely unconscious. Similarly, the traumatic experience of PTSD often operates at an unconscious level. Somatic symptoms often predominate and dissociative processes are common. Both these mechanisms operate as unconscious means of distancing oneself from the overwhelming experience of the event. Grief and loss often play a significant part in the traumatic event and may be psychodynamically related to real or imagined losses during childhood.

Therapeutic interventions for PTSD often emphasize the importance of verbalization of circumstances concerning the event as well as the associated affects. This technique is similar to those advocated in psychodynamic formulations. Initial memories of the events may take the form of dissociated mental imprints consisting of isolated affective and sensory elements. With time, the individual is able to form a more coherent verbal narrative that allows for more effective integration of the experience.[17] The process of memory may be subdivided into implicit memory, which is unconscious, and conscious explicit memory. The task of therapy may be viewed as converting implicit forms of recall to the explicit form.

Integrating Medication and Psychosocial Therapeutic Models in PTSD

Randall Marshall, MD,[18] of the NY State Psychiatric Institute-Columbia University, New York City, indicated that several drugs have been shown to be effective in treating PTSD, including tricyclic antidepressants, monoamine oxidase (MAO) inhibitors and, more recently, selective serotonin reuptake inhibitors (SSRIs). Sertraline was the first medication specifically approved for PTSD in the United States. Other SSRIs also appear to be useful including citalopram[19] and paroxetine.[20]

Drug therapy yields only a partial response, providing approximately a 50% reduction in symptoms. Therapeutic interventions are clearly indicated and there are several models for intervention. Although these techniques differ in theoretical and practical ways, they have some elements in common. They often attempt to expose of the patient to traumatic memories, help with articulation of the experience, and advocate the use of relaxation techniques to diminish the impact. Medications work through the modulation of affective responses including anxiety, depression, and fear. Therapeutic interventions are oriented to re-exposure and subsequent desensitization. Both drug therapy and psychotherapy help to decrease avoidance behaviors. The combined effects of these interventions should work in concert to maximize the response rate.

Neurobiological Bases of PTSD

The use of increasingly sophisticated imaging techniques including the positron emisssion tomography (PET) scan, magnetic resonance imaging (MRI), and functional MRI (fMRI) have allowed researchers to identify brain regions involved in the traumatic response. Several studies have shown reduced hippocampal volume, consistent with the memory disturbances characteristic of PTSD. Stein and colleagues,[21] for example, found that women who were victims of childhood sexual abuse had bilateral reduction of the hippocampus as seen on MRI. Only the left-sided reduction reached significance. The percent of volume reduction on the left correlated strongly with dissociative symptoms (r = 0.73) but not with explicit memories. These reductions may be a consequence or a pre-existing abnormality among PTSD sufferers.

Van der Kolk[22] has suggested that the deficiencies in explicit or declarative memory may result in a reorganization of the trauma on a somatosensory level. The somatosensory memories include visual and physical representations, and they are processed in subcortical regions outside of the hippocampal system and difficult to extinguish. Cortical activity may able to modify these subcortical responses; however, the extent of physiologic arousal and hormonal responsiveness may limit the ability of the cortex to carry out this inhibitory process. Activation of the amygdala, a limbic structure involved in emotional memories, such as fear responses, has been demonstrated by Liberson and colleagues.[23] Using fMRI, Rauch and coworkers[24] were able to link activation of the amygdala response to masked presentation of trauma-related stimuli.

A Primate Model of Enhanced Stress Responses

Jeremy Coplan, MD,[25,26] of Columbia University, New York City, presented on a primate model that was developed to evaluate the effect of maternal stress on offspring. Bonnet macaques from India show a high level of attachment to maternal figures. Three foraging conditions were established, with the first 2 being consistent reward systems. The "high foraging demand" condition required much searching for food in a scarce food environment. The "low foraging demand" condition required little searching in a food surplus environment. The last condition was a "variable foraging demand" (VFD), which presented the animal with an unpredictable food supply.

The offspring of the VFD mothers manifested increased anxiety, timidity, avoidance, sadness, and slumped posture. Grown VFD animals demonstrated a variety of neurobiological abnormalities including increased levels of cerebrospinal fluid (CSF) corticotropin-releasing factor (CRF), reduced CSF cortisol levels, elevated levels of CSF serotonin and dopamine metabolites, as well as increased CSF somatostatin levels. CRF concentrations remained high over a 30-month period, suggesting that the rearing condition produced a trait-like condition. If the VFD was introduced later in the developmental process, during the late weaning phase, the neurobiological profile differed. These animals demonstrated low CRF, high cortisol, and high serotonin metabolite levels. Thus, timing of the intervention is a significant factor. In PTSD, early experiences may similarly result in a sensitization of the neurobiological system of the individual and predispose one to a traumatic stress response.

The Cognitive Science Approach To Trauma

Dan Stein, MD,[27] University of Stellenbauch Medical School, Tygerberg, South Africa, attempted to integrate some of the diverse findings in the PTSD literature. He suggested that a cognitive science perspective might be helpful. Based on computational models of the mind, cognitive science attempts to use nonlinear conceptual models, such as parallel-processing models and neural networks.

Cognitive science suggests that people use cognitive "schemas" to view and interpret the external information. It allows the individual to sort and simplify the world by categorizing information. Although helpful in the processing of information, it also results in bias and distortions of one's perceptions of the world. Neural networks are the brain's way of forming schemas. These networks process data in a specific reproducible manner. There is redundancy in the system thus 1 part of the network may drop out and the overall schema preserved. Examples of theoreticians who have used the concept of schemas include developmental psychologists, such as Piaget, and cognitive therapists, such as Aaron Beck.

Experimental techniques have offered support for the concept of PTSD-related information-processing schemas. Studies have indicated that PTSD sufferers tend to selectively process trauma-related material, demonstrate enhanced memories for events related to the traumatic experience, have greater difficulty in forgetting words related to the trauma, and tend to use more global "overgeneral" recall patterns.[28]

Attentional bias to trauma-relevant stimuli and deficits in memory for trauma relevant stimuli appear to be related to structural findings, including smaller hippocampal volume as well as enhanced responsiveness of the amygdala. Alternatively, higher cortical processes generated in the medial prefrontal areas of the brain may suppress the fear response. Individuals with low IQ are more susceptible to PTSD, which may be secondary to decreased ability to use these higher cortical processes to mediate the traumatic memories and affective responses.[29] The "sensorimotor" processing of memories is used in PTSD rather than verbal strategies. Evidence for this nonverbal preference is supported by studies demonstrating decreased blood flow to Broca's area during recall of traumatic events.[30]


The study of PTSD offers fertile ground for understanding the means by which the environment exerts its influence on the organism. In PTSD, a specific trauma or series of traumas trigger a prolonged dysfunctional psychological and physiologic response. This response must be understood in the context of the psychological, sociocultural, and neurobiological determinants that influence the clinical presentation. New findings from imaging studies offer exciting insights into the neural events mediating the symptoms. New theoretical viewpoints are helping to integrate this information and develop enhanced treatment strategies.


Breslau N. The stressor criterion in DSM-IV. Program and abstracts of the 154th Annual Meeting of the American Psychiatric Association; May 5-10, 2001; New Orleans, Louisiana. Symposium 4A.

Breslau N, Kessler RC, Chilcoat HD, Schultz LR, Davis GC, Andreski P. Trauma and posttraumatic stress disorder in the community: the 1996 Detroit Area Survey of Trauma. Arch Gen Psychiatry. 1998;55:626-632.

Bowman ML. Individual difference in post-traumatic response. Program and abstracts of the 154th Annual Meeting of the American Psychiatric Association; May 5-10, 2001; New Orleans, Louisiana. Symposium 4B.

Bowman ML. Individual differences in posttraumatic distress: problems with the DSM-IV model. Can J Psychiatry. 1999;44:21-33.

Paris JF. Personality factors in susceptibility to PTSD. Program and abstracts of the 154th Annual Meeting of the American Psychiatric Association; May 5-10, 2001; New Orleans, Louisiana. Symposium 4E.

Jarvis GE. Coercive treatment of immigrants with psychosis. Program and abstracts of the 154th Annual Meeting of the American Psychiatric Association; May 5-10, 2001; New Orleans, Louisiana. Symposium NR16.

Mezzich JE, Kirmayer LJ, Kleinman A, et al. The place of culture in DSM-IV. J Nerv Ment Dis. 1999;187:457-464.

Kinzie JD, Denney D, Riley C, Boehnlein J, McFarland B, Leung P. A cross-cultural study of reactivation of posttraumatic stress disorder symptoms: American and Cambodian psychophysiological response to viewing traumatic video scenes. J Nerv Ment Dis. 1998;186:670-676.

Sack WH, Seeley JR, Clarke GN. Does PTSD transcend cultural barriers? A study from the Khmer Adolescent Refugee Project. J Am Acad Child Adolesc Psychiatry. 1997;36:49-54.

Lewis- Fernandez R. Ethnocultural factors in PTSD. Program and abstracts of the 154th Annual Meeting of the American Psychiatric Association; May 5-10, 2001; New Orleans, Louisiana. Symposium 14D.

Ruef AM, Litz BT, Schlenger WE. Hispanic ethnicity and risk for combat-related posttraumatic stress disorder. Cultural Divers Ethnic Minority Psychology. 2000;6:235-251.

Deblinger E, Steer RA, Lippmann J. Two-year follow-up study of cognitive behavioral therapy for sexually abused children suffering post-traumatic stress symptoms. Child Abuse Negl. 1999;23:1371-1378.

Paunovic N. Exposure counterconditioning (EC) as a treatment for severe PTSD and depression with an illustrative case. J Behav Ther Exp Psychiatry. 1999;30:105-117

Hembree EA, Foa EB. Posttraumatic stress disorder: psychological factors and psychosocial interventions. J Clin Psychiatry. 2000;61(suppl 7):33-99.

Shalev, AY. Contemporary psychodynamic perspectives on trauma development. program and abstracts of the 154th Annual Meeting of the American Psychiatric Association; May 5-10, 2001; New Orleans, Louisiana. Symposium 14A

Cohen JA, Mannarino AP, Rogal S. Treatment practices for childhood posttraumatic stress disorder. Child Abuse Negl. 2001;25:123-135.

van der Kolk BA, Fisler R. Dissociation and the fragmentary nature of traumatic memories: overview and exploratory study. J Trauma Stress. 1995;8:505-525.

Marshall RD. Integrating medication and psychosocial therapeutic models in PTSD. Program and abstracts of the 154th Annual Meeting of the American Psychiatric Association; May 5-10, 2001; New Orleans, Louisiana. Symposium 14B. Seedat S, Stein DJ, Emsley RA. Open trial of citalopram in adults with post-traumatic stress disorder. Int J Neuropsychopharmacol. 2000;3:135-140.

Marshall RD, Schneier FR, Fallon BA, et al.. An open trial of paroxetine in patients with noncombat-related, chronic posttraumatic stress disorder. J Clin Psychopharmacol. 1998;18:10-18.

Stein MB, Koverola C, Hanna C, Torchia MG, McClarty B. Hippocampal volume in women victimized by childhood sexual abuse. Psychol Med. 1997;27:951-959.

van der Kolk BA. The body keeps the score: memory and the evolving psychobiology of posttraumatic stress. Harv Rev Psychiatry. 1994;1:253-265.

Liberzon I, Taylor SF, Amdur R, et al. Brain activation in PTSD in response to trauma-related stimuli. Biol Psychiatry. 1999;45:817-826.

Rauch SL, Whalen PJ, Shin LM, et al. Exaggerated amygdala response to masked facial stimuli in posttraumatic stress disorder: a functional MRI study. Biol Psychiatry. 2000;47:769-776.

Coplan, JD, Smith EL, Scharf LA, Baptiste S, Altamash I, et al. Variable foraging demand rearing in primates: biobehavioral sequelae. Program and abstracts of the 154th Annual Meeting of the American Psychiatric Association; May 5-10, 2001; New Orleans, Louisiana. Symposium 14C.

Coplan JD, Andrews MW, Rosenblum LA, et al. Persistent elevations of cerebrospinal fluid concentrations of corticotropin-releasing factor in adult nonhuman primates exposed to early-life stressors: implications for the pathophysiology of mood and anxiety disorders. Proc Natl Acad Sci U S A. 1996;20;93:1619-1623.

Stein DJ. The cognitive-affective science of trauma and development. Program and abstracts of the 154th Annual Meeting of the American Psychiatric Association; May 5-10, 2001; New Orleans, Louisiana. Symposium 14E.

McNally RJ. Experimental approaches to cognitive abnormality in posttraumatic stress disorder. Clin Psychol Rev. 1998;18:971-982.

Buckley TC, Blanchard EB, Neill WT. Information processing and PTSD: a review of the empirical literature. Clin Psychol Rev. 2000;20:1041-1065.

Fredrikson M, Fischer H, Wik G. Cerebral blood flow during anxiety provocation. J Clin Psychiatry. 1997;58 Suppl 16:16-21.

The Depressive Phase of Bipolar Disorder: Focus on Bipolar II

Hagop S. Akiskal MD


There is considerable confusion about the depressive phase of bipolar disorder. In particular, it is uncertain whether depression manifests with psychomotor retardation or agitation. Indeed both have been reported. Table 1 summarizes what we know from the literature about the differential features of bipolar and unipolar disorders in their depressive phase. It is a differential diagnostic question with major therapeutic implications. Bipolar II depression is the most common expression of bipolar disorder, and recent developments on the clinical features of the depressive phase of bipolar disorder were presented during this year's Annual Meeting of the American Psychiatric Association.[1] This review focuses on clinical characteristics of bipolar II depression; the optimum way by which it could be prospectively predicted before the occurrence of hypomania; and new data that amplify or modify past findings.

Prevalence of Bipolar II in Major Depression

The diagnosis of bipolar II is crucial, not only because of its therapeutic implications, but also for prognostic reasons (eg, high suicidality).[2] Accordingly, it is gratifying that a great deal of recent research has been conducted on the clinical prevalence of bipolar II among patients presenting with major depressive disorder in public and private psychiatric clinics worldwide.[3] The data show that from 30% to 55% of all major depressions conform to bipolar II or its variants. According to Simpson and colleagues,[4] bipolar II represents the most common phenotype of bipolar disorder. The French EPIDEP study,[5] which was based on a representative national clinical sample, has provided the most compelling data on the high prevalence of bipolar II among major depressive patients. The overarching purpose of this study was to assist practicing psychiatrists to recognize bipolarity in all of its varieties. The major finding was that at index interview 22% of major depressive patients could be diagnosed as bipolar II based on past history of hypomania. Upon interview a month later 40% of patients were diagnosed as bipolar II on the basis of a more in-depth evaluation and collateral information from significant others, as well as observed hypomania by the clinician. These data indicate that clinicians in diverse practice settings can be trained to recognize bipolar II and its variants, leading to changes in diagnostic practice at the national level.

Diagnostic Issues

Often patients with bipolar II disorder present with a major depressive episode, and upon further inquiry, a history for hypomanic episodes is elicited. Accurate diagnostic subtyping then depends on the vagaries of the patient's memory and how systematically the clinician pursues lead questions about hypomania, and, crucially, whether relatives are interviewed. Rice and colleagues[6] in the context of the National Institute of Mental Health (NIMH) collaborative study of depression, reported low reliability of the bipolar II diagnosis. However, all such diagnoses occurred strictly according to the criteria for bipolar disorder, suggesting that once history for hypomania was obtained, it did carry significant diagnostic specificity. In practice, then, the diagnosis of bipolar II requires repeated evaluation at different time points to obtain more reliable history for hypomania. Dunner and Kai[7] found that clinicians specifically trained to recognize bipolar II, far outperformed those using structured instruments such as the Schedule for Affective Disorders and Schizophrenia (SADS) or the Structured Clinical Interview for DSM-IV (SCID) in the diagnosis of bipolar II disorder. Although the foregoing consideration runs against the usual tenets in the literature on structured diagnostic instruments, they are consistent in suggesting that the proper identification of bipolar II requires a sophisticated approach in interviewing.

Among the major reasons for diagnostic inconsistency in bipolar II are the underlying temperamental dysregulation that results in clinical presentations of a labile-variable nature as well as confusing cross-sectional features. That is, these patients could present with cross-sectional features of atypical depression and a lifelong history of anxiety states, bulimia, substance abuse, or cluster B personality disorder (eg, "borderline").[8] A prospective study has also demonstrated that atypical depressions more often than not progresses to bipolar spectrum disorders.[9] Hypersomnia, weight gain, and related atypical features then represent an important clinical pointer to bipolar II that is also observed in cross-sectional studies; such features may actually serve as a clinical marker for bipolar II.[10]

Another interesting development in the complexity of bipolar II is "panic attacks." There have been stronger load score estimates in molecular linkage in bipolar II patients with familial panic disorder vs those without.[11] The affective dysregulation of bipolar disorder obviously extends beyond elation and depression to include, among others, negative affective arousal states such as panic, irritability, and mood lability.

The Specificity of Mood Lability and Related Life Disruptions Analyses from the NIMH Collaborative Depression Study on 559 "unipolar" patients showed that 48 converted to bipolar II during a prospective observation period of 11 years. [12] What characterized these bipolar II converters at entry was early age at onset of first depression, recurrent depression, high rates of divorce or separation, high rates of scholastic and/or job maladjustment, isolated "antisocial acts," and drug abuse, in other words, a more tempestuous affective and life history. In addition, the index depressive episode was further characterized by features such as

Phobic anxiety

Interpersonal sensitivity

Obsessive-impulsive symptoms

Somatization (often with subpanic symptoms)

Worsening in the evening



Subjective or overt anger, jealousy, suspiciousness, and ideas of reference All of which testify to a broad melange of "atypical" depressive symptoms with "borderline" features. Temperamental attributes obtained at index interview proved decisive (sensitivity, 91%) in identifying those who switched from depression to hypomania: these attributes consisted of trait "mood lability," "energy activity," and "daydreaming," all characteristic of Kretschmer's description of the cyclothymic temperament. Mood lability was the most specific predictor (specificity, 86%) of which depressions will prospectively change to bipolar. This study testifies to the fact that bipolar II disorder is a complex affective disorder with biographical instability, deriving more often than not from an intense temperamental dysregulation. Mood lability -- with rapid shifts, often in a depressive polarity -- was the hallmark of "unipolar" patients who switched to bipolar II. Unfortunately, our formal diagnostic systems (eg, International Classification of Diseases 10 [ICD-10] and DSM-IV) are symptom-oriented and do not consider extreme temperamental dispositions in clinical evaluation; also regrettably such patients often get labeled "borderline."

Bipolar II Depressive Mixed States

Because history for hypomania is often difficult to obtain from a depressive patient, one may examine hypomanic features during a depressive phase. Hypomanic symptoms, such as racing and grandiose thoughts, sexual arousal, and psychomotor acceleration, have been described in major depressive episodes in contemporary psychiatry (thereby testifying to Kraepelin's diagnostic acumen) yet the number of studies reporting on "bipolar depressive mixed states" are too few.[13] A recent study by Benazzi[14] demonstrated that irritability, distractibility, and racing thoughts were the most common hypomanic features during depression, mostly among bipolar II depressives. Unfortunately, such studies have not commanded sufficient interest in official nosologic systems, nor in the clinical literature. This is a clinical tragedy because these are the very "unipolar" depressive patients who are likely to do poorly on antidepressants and require mood stabilizers, antipsychotics, or electroconvulsive therapy. Koukopoulos and Koukopoulos[15] have recently written a superb clinical article on agitated depression as a mixed state. This is an instance where clinical acumen has outpaced the conventional scientific literature.

New Data From EPIDEP In the French national collaborative study on depression (EPIDEP),[5] interim analyses demonstrated that hypersomnia and suicidal ideation were more common in bipolar II vs unipolar disorder (which had more insomnia). In new analyses based on the entire sample of 493 patients, the differential features of the 2 disorders emerged more clearly (unpublished data from EPIDEP):

Uniformity of global intensity of depression (HAM-D)

Bipolar II patients scored higher on suicidal thoughts, guilt, depersonalization, derealization, hypersomnia, and weight gain

Strictly unipolar major depressives had more psychic anxiety and insomnia

Unipolars on various self-assessment (eg, Multiple Analog Visual Scale for bipolar in the Ahearn-Carroll scale) scored higher in terms of slow thinking, no energy, feeling worst, avoiding risks, perception of life as dull and dreary. Indeed, psychomotor retardation was highest in unipolars, and this difference appeared most significant when compared with bipolar II depression.


Depression appears to be the more common expression of the bipolar spectrum. Indeed bipolar II may actually be the most common phenotype of this disorder. Although much has been published about the clinical features of mania, there has been a comparative paucity of research on the depressive phase of bipolar disorder.

This presentation has summarized the differences between the depressive phase of bipolar and unipolar disorder. Apart from demographic, familial, and disorder course differences, the literature has emphasized differences in psychomotor retardation. Most studies have reported higher psychomotor retardation in bipolar than unipolar disorder. However, this depends on whether mixed features are permitted in bipolar depression, and whether or not the comparison with unipolar is between bipolar I or bipolar II. The overall data seem to suggest that in bipolar II depression, irritability, distractibility, racing thoughts, and mood lability are quite common; this is despite the presence of hypersomnia.

Once bipolar II is properly diagnosed and excluded from loosely defined unipolar major depression, not only are atypical features less common in strictly defined unipolar patients, but the mixed features documented above represent the clinical signature of bipolar II. It is therefore not surprising that psychomotor retardation per se might be more severe in unipolar patients. On the other hand, psychic anxiety and insomnia, long known for their association with unipolar disorder, continue to accompany this subtype of depression. The situation is actually more intricate because of increasing reports about high prevalence of comorbid anxiety states in bipolar II. This means that the presence of anxiousness does not automatically suggest unipolarity.

The foregoing conclusions are complex and suggest that heterogeneity of depressive disorders, in particular bipolar II, should be taken into consideration in examining the differentiating characteristics of unipolar and bipolar disorders.


Akiskal HS. Bipolar Depression: Phenomenology and Diagnosis. Program and abstracts of the 154th Annual Meeting of the American Psychiatric Association; May 5-10, 2001; New Orleans, Louisiana. Symposium 96A.

Dunner DL, Gershon ES, Goodwin FK. Heritable factors in the severity of affective illness. Biol Psychiatry. 1976;11:31-42.

Akiskal HS, Bourgeois ML, Angst J, Post R, Möller HJ, Hirschfeld R. Re-evaluating the prevalence of and diagnostic composition within the broad clinical spectrum of bipolar disorders. J Affect Disord. 2000;59 (suppl): S5-S30.

Simpson SG, Folstein SE, Meyers DA et al. Bipolar II: the most common bipolar phenotype? Am J Psychiatry. 1993.150:901-903.

Hantouche EG, Akiskal HS, Lancrenon S, et al. Systematic clinical methodology for validating bipolar-II disorder: data in mid-stream from a French national multisite study (EPIDEP). J Affect Disord. 1998;50: 163-173.

Rice JP, McDonald-Scott P, Endicott J, et al. The stability of diagnosis with an application to bipolar II disorder. Psychiatry Res. 1986;19:285-296.

Dunner DL, Kai Tay L. Diagnostic reliability of the history of hypomania in bipolar II patients with major depression. Compr Psychiatry. 1993;34:303-307.

Perugi G, Toni C, Akiskal HS. Anxious-bipolar comorbidity: diagnostic and treatment challenges. Psychiatr Clin North Am. 1999;22: 565-583.

Ebert D, Barocka A, Kalb R, Ott G. Atypical depression as a bipolar spectrum disease: evidence from a longitudinal study. The early course of atypical depression. Psychiatria Danubina. 1993;5:133-136.

Benazzi F. Depression with DSM-IV atypical features: a marker for bipolar II disorder. Eur Arch Psychiatry Clin Neurosci. 2000;250:53-55.

MacKinnon D, Xu J, McMahon FJ, et al. Bipolar disorder and panic disorder in families: an analysis of chromosome 18 data. Am J Psychiatry. 1998;55:829-831.

Akiskal HS, Maser JD, Zeller P, Endicott J, Coryell W, Keller M. Switching from "unipolar" to "bipolar II": an 11-year prospective study of clinical and temperamental predictors in 559 patients. Arch Gen Psychiatry. 1995;52:114-123.

Akiskal HS. Delineating depressive mixed states in the course of Bipolar I disorder.

Benazzi F. Depressive mixed states: unipolar and bipolar II. Eur Arch Psychiatry. 2000;Clin Neurosci. 250:249-253.

Koukopoulos A, Koukopoulos A. Agitated depression as a mixed state and the problem of melancholia. Psych Clin North Am. 1999;22:547-564.

Psychiatry, February 2002

Author: Irving Kuo, MD

Journal Scan is the clinician's guide to the latest clinical research findings in the American Journal of Psychiatry, The Lancet, Archives of General Psychiatry, Journal of Clinical Psychiatry, and Journal of Clinical Psychopharmacology. Short summaries of feature articles include links to the article abstracts and full text, when available. (Access to full text usually requires registration at the journal's Web site.)

Clozapine, Olanzapine, Risperidone, and Haloperidol in the Treatment of Patients With Chronic Schizophrenia and Schizoaffective Disorder

The American Journal of Psychiatry

February 2002 (Volume 159, Number 2)

Volavka J, Czobor P, Sheitman B, et al.

The American Journal of Psychiatry. 2002;159(2):255-262

Clozapine has, since its introduction, been the gold standard for the treatment of treatment-resistant psychoses. There have been few head-to-head studies examining the efficacy of the atypical antipsychotic medications in treatment-resistant patients. The authors attempted to compare the efficacy of 4 antipsychotic agents (clozapine, olanzapine, risperidone, and haloperidol) in a population of 157 patients with a diagnosis of schizophrenia or schizoaffective disorders who had a suboptimal response to previous medications. Patients who were previous nonresponders to clozapine, olanzapine, or risperidone were excluded from the study. Study duration was for 14 weeks, and the patients were blindly assigned to the medication group. The 3 atypical agents (but not haloperidol) were shown to significantly improve patients' total scores on the Positive and Negative Syndrome Scale (PANSS), although the improvements were relatively modest. Negative symptoms showed improvement with both clozapine and olanzapine. Although clozapine appeared to be the most effective in treating negative symptoms and risperidone appeared not to be quite as effective as clozapine or olanzapine, the differences were relatively small.


Functional Impairment in Patients With Schizotypal, Borderline, Avoidant, or Obsessive-Compulsive Personality Disorder

Skodol AE, Gunderson JG, McGlashan TH, et al.

The American Journal of Psychiatry. 2002;159(2):276-283

Functional impairment is a prerequisite in making the diagnosis of a personality disorder. The authors recruited 668 patients with one of the following disorders -- schizotypal personality disorder (n = 86), borderline personality disorder (n=175), avoidant personality disorder (n = 157), obsessive-compulsive personality disorder (n = 153), and major depressive disorder (n=97) -- and examined them in their psychosocial functioning by using both the Longitudinal Interval Follow-up Evaluation--Baseline Version and the Social Adjustment Scale. The authors found that patients with schizotypal and borderline personality disorders had the most severe impairment in psychosocial functioning in areas of work, social relationships, and leisure activities. Those with obsessive-compulsive personality disorder or major depression had the least impairment, and those with avoidant personality disorder fell in between. The authors concluded that personality disorders were associated with significant impairment and more functional impairment than individuals with major depressive disorder alone.

Personality Disorders Can Be More Disabling Than Depression

Clinical & Research News

Joan Arehart-Treichel

Researchers compared the functioning of persons with a major depressive disorder with that of persons with various personality disorders. They found that while having a major depression may not be easy, having a borderline, schizotypal, or avoidant personality disorder is probably more difficult.

Which mental illnesses are most severe? A new study may surprise some people with its finding that certain personality disorders are even more impairing than major depression.

The study was conducted by Andrew Skodol, M.D., a psychiatrist-researcher at the New York State Psychiatric Institute, and colleagues and was reported in the February American Journal of Psychiatry.

More than 600 subjects with borderline personality disorder, schizotypal personality disorder, avoidant personality disorder, obsessive-compulsive personality disorder, or major depressive disorder were included in the investigation. They were recruited from sites of the long-term Collaborative Longitudinal Personality Disorders Study, which is being conducted with the participation of investigators from Brown University, Columbia University, Harvard University, Texas A&M University, and Yale University.

The psychological and sociological functioning of these subjects was then assessed with the Longitudinal Interval Follow-Up Evaluation. It included questions to evaluate functioning in employment, household duties, student work, and recreation; interpersonal relationships with parents, siblings, partners, and friends; and global functioning. Subjects also rated their own functioning using the Social Adjustment Scale. The researchers then tallied the results of these assessments for each of the five groups and compared results.

Results from the Longitudinal Interval Follow-Up Evaluation indicated that persons with borderline personality disorder or schizotypal personality disorder were most impaired functionally; that individuals with avoidant-personality disorder were next; and that persons with obsessive-compulsive disorder or major depressive disorder were least so.

The subjects’ own functional ratings using the Social Adjustment Scale gave similar results. Persons with schizotypal personality disorder or borderline personality disorder rated themselves as significantly more impaired in all individual domains of functioning and overall than did persons with obsessive-compulsive personality disorder or major depressive disorder. Individuals with avoidant personality disorder remained intermediate.

In fact, persons with schizotypal personality disorder and borderline personality disorder had greater impairment on virtually every measure of impairment than did persons with obsessive-compulsive disorder or major depressive disorder regardless of whether the evaluation was interview based or by patient self-report.

What’s more, these results remained statistically significant even when the investigators took into consideration possibly confounding factors such as age, gender, or minority status.

"Our findings are especially noteworthy," Skodol and his team wrote in their study report, "given the growing appreciation for the degree and persistence of limitations in functioning of patients with major depressive disorder. Impairment due to major depressive disorder has been found to be comparable with that of patients with chronic medical illnesses such as diabetes and arthritis."

In other words, although major depressive disorder is quite debilitating, borderline personality disorder, schizotypal personality disorder, and avoidant personality disorder appear to be all the more so.

Paul Links, M.D., is a professor of psychiatry at the University of Toronto in Canada. Since he is especially interested in personality disorders, Psychiatric News asked him whether he would comment on the study by Skodol and his team.

The study, Links said, provides "an unparalleled opportunity to compare the functioning across subjects with severe, less-severe, and no personality disorders. As our clinical experience would suggest, patients with severe personality disorders—schizotypal and borderline—have greater impairments in all areas measured compared with subjects with no or less-severe personality disorders."

This report, however, "is just the first step," he continued. "We need to have better methods of assessing functioning, including observational data. We need to understand the relationship between personality psychopathology and functioning. Finally, we need to examine how interventions affect functioning."

The investigation was financed by grants from the National Institute of Mental Health and from the Borderline Personality Disorder Research Foundation.

The study report, "Functional Impairment in Patients With Schizotypal, Borderline, Avoidant, or Obsessive-Compulsive Personality Disorder," is posted on the Web at http://ajp.psychiatryonline.org under the February issue.

Eating Disorders: New Features and New Treatments

Martin L. Korn MD


Despite the advances in treatment modalities and scientific investigations, patients with eating disorders still present one of the most difficult challenges faced by clinicians. In order to better understand and intervene in these patients, the participants in this symposium on eating disorders presented on many perspectives related to the understanding and treatment of these patients.

Cognitive Functioning in Anorexia Nervosa

It is well known that individuals with anorexia nervosa often focus on perceived excesses in selected areas of the body, such as the hips and stomach. This is often done to the neglect of the overall picture of diminished weight or even emaciation. An outside observer often feels that the anorexic individual "just doesn't get it" and therapeutic interventions are often ineffective in correcting these perceptual difficulties. David Herzog, MD,[1] of the Massachusetts General Hospital, Boston, discussed the possible cognitive factors relating to these observations.

There are some similarities between the symptoms of obsessive-compulsive disorder (OCD) and anorexia. The focus on weight and defective body areas may be viewed as obsessional in nature, whereas the ritualistic concern about eating habits is analogous to compulsive symptoms. There are also family studies linking the eating disorders and OCD. Bellodi and colleagues[2] found that that there is increased risk of OCD in relatives of patients with eating disorders compared with controls (9.69% vs 0%). Patients with eating disorders have been shown to have a higher incidence of OCD.[3]

Cognitive problems have been observed in patients with OCD, and several studies have shown difficulties such as visuospatial problems, impaired organizational strategies, distractibility by irrelevant stimuli, impaired set-shifting capabilities, as well as difficulties in selective attention.[4] Similarly, there are some studies that have shown cognitive deficits in eating disorders. Lauer and colleagues[5] demonstrated impairment in attention and problem solving, although mnemonic processes were normal. The cognitive function tended to remit with clinical improvement.

One ongoing study[1] was designed to test organizational strategies and the relationship to cognitive dysfunction in patients with eating disorders via the Rey-Osterrieth Complex Figure Test. This neuropsychological test involves presentation of a complex figure which is then drawn by the subject during the presentation. The figure is then removed and the subject is then asked to draw the figure again. This is a test of immediate recall. Thirty minutes later, the subject is asked to draw the figure for the last time. This is a measure of delayed recall. The subjects are scored on constructional ability and organizational ability in addition to the memory tasks.

Patients with anorexia were shown to have deficits in both immediate and delayed recall compared with controls. The study also suggests that there were difficulties in organizational strategy that mediated these memory deficits. Differences in body mass index did not explain the differences. Comorbid conditions or treatment with psychotropic medications also did not account for the differences. These findings support the idea that patients with anorexia have cognitive deficits similar to patients with OCD. This may be a result of orbitofrontal-striatal dysfunction and imaging studies will be needed to further elucidate these findings.

Eating Behavior While Driving in a Car

In order to examine eating patterns of bulemic patients, James Mitchell, MD,[6] of the University of Bath, Bath, England, has been studying eating habits while driving. Referred to as the "drive and eat" study, patients with eating disorders who reported eating in their cars at least once a week were compared with community controls who also engaged in this behavior. A subgroup of bulemic patients indicated that they found the car to be an isolated place in which bingeing could safely take place. Vomiting was also sometimes practiced in the car. The ease of obtaining food at fast food outlets and convenience stores was thought to contribute to this behavior.

Cognitive-Behavior Therapy Relapse Predictors for Bulimia Nervosa

It is often difficult to determine when therapeutic interventions will be effective for a particular individual. Katherine Halmi, MD,[7] of Cornell University Medical College, White Plains, NY, conducted a study examining the determinants of sustained improvement after a course of cognitive-behavioral therapy (CBT).

Twenty sessions of CBT were administered over a period of 16 weeks to 194 patients. Fifty-eight of these patients had a complete remission by the end of the treatment course; 34 weeks after treatment completion, 48 of the 58 remitted patients were evaluated. Twenty-one of the 48 patients had relapsed and 27 remained in remission.

When these 2 groups were compared, there were several unexpected findings. The patients who sustained remission after 34 weeks had been ill for a longer period of time compared with those who did not sustain remission (14.7 months vs 9.2 months). The presence of a comorbid personality disorder, usually is indicative of poor prognosis, did not prove to be a significant factor in this study. Comorbid disorders including substance abuse had no effect.

In order to assess the extent of eating disorder difficulties, the Yale-Brown-Cornell Eating Disorder Scale (YBC-EDS) was used.[8] This instrument consists of 8 items that assess the severity of food preoccupations and rituals as well as 6 items that assess motivational factors. Individuals who had higher motivational scores or readiness to change scores after the CBT sessions were more likely to sustain the remission at 34 weeks. The same was true of patients with lower preoccupation and food ritual scores. Other scores that were positive predictors of sustaining remission were decreased thoughts of bulemia (bingeing-purging) as well as lower hunger scores at the end of the CBT sessions. Sustained remitters also had lower "restraint" scores. Restraint refers to the tendency to be preoccupied with restriction of intake. For example, individuals who stop eating before they are satiated and those who closely monitor the amount of caloric and nutritional intake have high levels of restraint. Lower scores on this measure both at the initiation of treatment as well as at the end of the CBT sessions were associated with 34-week remission. Patients with higher levels of self-esteem at the end of treatment were also more likely to sustain their improvement.

The greater the number of continuous weeks of complete abstinence during treatment also correlated with sustaining improvement. Patients tended to show a relatively steep decline of binging in the first several weeks after the initiation of treatment. Those who remained binge-free after treatment had a steeper decline in binging during treatment. Most of the long-term remitters were completely abstinent at 10 weeks.

To help predict outcome, a technique called signal detection statistical analysis was used. Patients with a YBC-EDS score of 7 or greater at the end of treatment demonstrated a relapse rate of 87%. Alternatively, if the score was less than or equal to 6, the relapse rate fell to 24%. Remission at week 10 of the treatment course was also helpful in predicting the risk of relapse. Using these techniques, a clinician will be better able to determine individuals most at risk and target clinical interventions as needed.

New Treatment Strategies for Anorexia Nervosa

Both anorexia and bulemia nervosa have many features in common. They both show a strong inheritance pattern, often appear in the same individual or in the same families, and share a cluster of psychological traits. These traits include negative affect, behavioral inhibition, tendencies toward perfectionism, high harm avoidance, as well as an obsessive concern with symmetry. Walter Kaye, MD,[9] of the University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania, suggested that these commonalities may be related to increased serotonergic function.

There are several indications that patients with eating disorders have serotonergic abnormalities.[10] Anorexics have increased levels of CSF 5-HIAA, the main metabolite of serotonin after recovery. Since CSF 5-HIAA is present after weight is restored, it may be an indication of a biochemical trait factor. Low levels of 5-HIAA have been associated with impulsive and aggressive behavior, traits that are opposite to those seen in many patients with anorexia nervosa. Serotonin is also involved in the regulation of eating behaviors.

Although several studies have demonstrated that bulemia is responsive to treatment with selective serotonin reuptake inhibitors (SSRIs), there have been negative results with SSRIs in anorexia nervosa.[11] Dr. Kaye suggested that this may be related to nutritional deficiencies in the anorectic individual. Kaye and colleagues[12] found that fluoxetine (mean dose, 40 mg) was helpful in preventing relapse in patients with anorexia. Sixty-three percent of patients on fluoxetine remained on the medication for a year compared with only 16% of those on placebo. Patients on fluoxetine had significant increases in weight and a reduction in obsessional symptoms, anxiety, and depression. As previously asserted by Kaye and colleagues,[13] SSRIs may not be helpful in nutritionally deprived patients. There may not an adequate supply of the essential amino acid tryptophan, the precursor to serotonin, in these individuals. Because neuronal supplies of serotonin would already be low, a drug that inhibits the reuptake of depleted serotonin levels would not be helpful in augmenting this neurotransmitter.

In a preliminary unpublished study by Kaye and colleagues, 6 patients with anorexia were placed on olanzapine. There was some evidence of weight gain with this intervention. Several clinicians at the study institution were already using olanzapine in nonstudy patients, however. In order to assess the patient acceptability of this agent, an uncontrolled retrospective survey of 18 of these patients was conducted. There was some evidence of weight gain, although this was not clearly related to the olanzapine. Patients felt that they were eating meals with greater ease and had less food obsessions and reduced anxiety before and during meals. These results must be considered very preliminary, but a controlled trial appears indicated. A clinical effect, if present, may be related to the 5HT2a receptor-blockading properties of olanzapine. This receptor profile is also seen with other atypical neuroleptics.

Adjunctive Treatment of Anorexia Nervosa by E-mail

The difficulties in the treatment of patients with eating disorders are greatly magnified in a rural setting. Contact with distantly located health care practitioners is difficult and this forms a major barrier to effective intervention. Joel Yager, MD,[14] of the University of New Mexico School of Medicine, Albuquerque, has been confronted with these difficulties and thus uses e-mail to communicate with his patients.

It is estimated that many physicians have communicated with patients via e-mail, however, there are numerous clinical and legal issues that confront physicians who use this tool. All communications should be considered open, and caution must be exercised regarding the nature of the material disclosed by the patient or physician. Furthermore, the computer leaves an "e-mail trail" that may be readily used in malpractice cases.

Nevertheless, there are many advantages of employing e-mail. The faceless, nonjudgmental computer screen acts like a psychoanalytic blank screen and may make information easier to disclose. There is unlimited access to the therapist, and the patient does not have to wait a week or more to communicate relevant information. The clinician may offer a frequent "virtual hug" and thus enhance the therapeutic connection. In patients with eating disorders, daily progress on behavioral programs may be monitored and the "work" of treatment may be done on a more frequent basis.

Dr. Yager indicated that there were many positive reactions by the patients to this intervention. A 17-year-old was able to use an informal tone in the communications that she would not have used in face-to-face encounters. She referred to Dr. Yager as "Dr. Yagermeister" or "Dr. Dude" and this appeared to reflect the ease that she felt in communicating in a more direct way. When more demoralized, she would revert to more formal greetings, such as "Dr. Yager, sir." Responses by the therapist were brief, often consisting of 1-2 lines of encouragement. Nevertheless, she found them comforting. On 1 occasion she wrote that it was "nice to hear back from you, makes me feel like I am more than a once-a-week patient."

Many patients said that e-mail made them think about things daily. It also served as a monitor for clinical deterioration. If e-mail communications lapsed, it was often a sign of difficulties. Patterns, such as sleep, could be assessed, as middle-of-the-night messages were an indication of problems in this area. One patient indicated that this direct communication gave her a "sense of security because there is always an open door between me and someone who is there to help me if I need it. Getting feedback on how I am doing is also very encouraging." Although this "on-demand feeding" may be comforting, the clinician must inform the patient of vacations or other times of unavailability. If there are lapses in clinician communication, this may be taken by the patient as a sign of rejection or disapproval.

The choice of appropriate patient is important to ensure that the experience is a constructive one. Western and colleagues[15] have delineated 3 personality types in patients with eating disorders. There is a high-functioning/perfectionistic type, a constricted/overcontrolled type, and an emotionally dysregulated/undercontrolled type. Dr. Yager suggested that the perfectionistic type is most suited to this intervention, whereas the technique is less effective in the constricted/overcontrolled type. Caution should be exercised with regard to the dysregulated/undercontrolled type because the impulsive element may interfere with the utility of the intervention. Appropriate boundaries may be difficult for this individual to maintain without face-to-face encounters or direct voice contact. Another problem that might particularly arise with such patients is the expression of suicidal tendencies in the middle of the night. Because issues such as these do not get the immediate clinical attention they deserve, planning for such behaviors must done in advance.


Anorexia and bulimia have frequently been some of the most difficult disorders to treat. There is a very high treatment drop out rate and relapse rates, as delineated in an unpublished study presented by Halmi and colleagues.[7] This important symposium on eating disorders summarized some of the new advances and ways of understanding that may help in treatment of these disorders. A better understanding of cognitive, behavioral, and neurobiological underpinnings of these disorders can only serve to enhance the treatment capability and options of the clinician.

References Herzog DB, Sherman BJ, Eddy KT, Blais MA, Connor CE, Deckersback T, et al. Cognitive functioning in anorexia nervosa. Program and abstracts of the 154th Annual Meeting of the American Psychiatric Association; May 5-10, 2001; New Orleans, Louisiana. Symposium 8A.

Bellodi L, Cavallini MC, Bertelli S, Chiapparino D, Riboldi C, Smeraldi E. Morbidity risk for obsessive-compulsive spectrum disorders in first-degree relatives of patients with eating disorders. Am J Psychiatry. 2001;158:563-569.

Lennkh C, Strnad A, Bailer U, Biener D, Fodor G, de Zwaan M. Comorbidity of obsessive compulsive disorder in patients with eating disorders. Eat Weight Disord. 1998;3:37-41.

Okasha A, Rafaat M, Mahallawy N, et al. Cognitive dysfunction in obsessive-compulsive disorder. Acta Psychiatr Scand.2000;101:281-285.

Lauer CJ, Gorzewski B, Gerlinghoff M, Backmund H, Zihl J. Neuropsychological assessments before and after treatment in patients with anorexia nervosa and bulimia nervosa. J Psychiatr Res. 1999;33:129-138.

Mitchell JE, Glass JB. Eating behavior while driving a car. Program and abstracts of the 154th Annual Meeting of the American Psychiatric Association; May 5-10, 2001; New Orleans, Louisiana. Symposium 8B.

Halmi KA, Agras WS, Mitchell JE, Crow SJ, Wilson GT. Cognitive-behavior therapy relapse predictors for bulimia nervosa. Program and abstracts of the 154th Annual Meeting of the American Psychiatric Association; May 5-10, 2001; New Orleans, Louisiana. Symposium 8C.

Mazure CM, Halmi KA, Sunday SR, Romano SJ, Einhorn AM. The Yale-Brown-Cornell Eating Disorder Scale: development, use, reliability and validity. J Psychiatr Res. 1994;28:425-445.

Kaye WH, Lavia M, Frank G, Molina A. New treatment strategies for anorexia nervosa. Program and abstracts of the 154th Annual Meeting of the American Psychiatric Association; May 5-10, 2001; New Orleans, Louisiana. Symposium 8D.

Kaye WH. Anorexia nervosa, obsessional behavior, and serotonin. Psychopharmacol Bull. 1997;33:335-334.

Ferguson CP, La Via MC, Crossan PJ, Kaye WH. Are serotonin selective reuptake inhibitors effective in underweight anorexia nervosa? Int J Eat Disord. 1999;25:11-17.

Kaye WH, Nagata T, Weltzin TE, et al. Double-blind placebo-controlled administration of fluoxetine in restricting- and restricting-purging-type anorexia nervosa. Biol Psychiatry. 2001;49:644-652.

Kaye W, Gendall K, Strober M. Serotonin neuronal function and selective serotonin reuptake inhibitor treatment in anorexia and bulimia nervosa. Biol Psychiatry. 1998;44:825.

Yager J. Adjunctive treatment of anorexia nervosa by e-mail. Program and abstracts of the 154th Annual Meeting of the American Psychiatric Association; May 5-10, 2001; New Orleans, Louisiana. Symposium 8E.

Westen D, Harnden-Fischer J. Personality profiles in eating disorders: rethinking the distinction between axis I and axis II. Am J Psychiatry. 2001;158:547-562.