An Overview of the Soft Bipolar Spectrum
By Arnold L. Lieber, MD

The purpose of this communication is to offer an overview of the soft bipolar spectrum, with its varied clinical presentations and diagnostic challenges. It is intended primarily for mental health professionals. However, I have included parenthetical explanations of medical terms and trade names for drugs so that patients and their families might find the material easy to understand. References are made to other publications offering greater details and clinical examples of the material we will touch upon here.

Bipolar I Disorder (manic depressive illness with or without psychosis), bipolar II disorder (episodes of major depression alternating with episodes of hypomania which are not severe enough to result in impairment of function) and cyclothymic disorder (brief and attenuated episodes of depression and hypomania sometimes known as minor cyclic mood disorder) are defined by explicit criteria sets in the latest edition of the Diagnostic and Statistical Manual of Mental Disorders ( DSM- IV ). Together, these three conditions are thought to have a lifetime prevalence of 3% to 4% of the general population. In recent years, clinical research has begun to validate the observations reported by experienced practitioners of clinical psychopharmacology – that a much larger group of patients demonstrate milder and/or atypical forms of episodic mood disturbances ( sub threshold bipolar disorder ). These patients are frequently resistant to standard antidepressant therapies, and sometimes their conditions are worsened by drug treatment with antidepressants. Efforts at clinical subtyping of the so-called soft bipolar spectrum are ongoing; up to this time such patients have fallen into the DSM-IV diagnostic category of Bipolar Disorder, NOS (not otherwise specified). If we include these patients with the three DSM-IV bipolar subtypes, the lifetime prevalence approaches 5% to 8 % of the general population. This is a far cry from the 1 % prevalence for manic depressive illness postulated by several large scale epidemiological surveys conducted by academic consortiums.



The great German neuropsychiatrist, Emil Kraepelin, described what he termed manic-depressive insanity at the end of the nineteenth century. He conceptualized a continuum that included today’s DSM-IV subtypes, mixed and rapid cycling states, many of the soft bipolar variations and also episodic depressions. This view prevailed until the 1960’s, at which time the creators of the first DSM edition ( DSM-I ) proposed a differentiation between major depression and manic-depressive illness. In later DSM editions, this evolved to the unipolar – bipolar dichotomy. In the 1970’s Fieve and Dunner discriminated bipolar I from bipolar II disorder, a seminal event in the evolution of the soft bipolar spectrum. Gerald Klerman was the first to postulate a further subtyping of bipolar disorders in 1981. Klerman’s Classification of Primary Bipolar Subtypes is summarized as follows:


Klerman’s Primary Bipolar Subtypes

(Psychiatric Annals #17: January 1987)
Bipolar I: Mania and depression
Bipolar II: Hypomania and depression
Bipolar III: Cyclothymic disorder
Bipolar IV: Hypomania or mania precipitated by antidepressant drugs
Bipolar V: Depressed patients with a family history of bipolar illness
Bipolar VI: Mania without depression [unipolar mania]


Today’s leading conceptual thinker in the area of bipolar subtyping is Hagop Akiskal. A fastidious researcher and an astute clinical observer, Akiskal is a devotee of Kraepelin. He believes that the nosologic (classification) pendulum is gradually swinging back towards the master’s original unitary concept of the bipolar spectrum of mood disorders. Akiskal’s description of the bipolar subtypes differs from that of Klerman. He proposes six subtypes, some of which are further subdivided according to their unique clinical features. A summary of his proposed subtype schema is as follows:


Akiskal’s Schema of Bipolar Subtypes

(Psychiatric Clinics of North America 22:3, September 1999; Medscape Family Medicine, 2005;7[1])Bipolar I: full-blown mania
Bipolar I ½: depression with protracted hypomania
Bipolar II: depression with hypomanic episodes
Bipolar II ½: cyclothymic disorder
Bipolar III: hypomania due to antidepressant drugs
Bipolar III ½: hypomania and/or depression associated with substance use
Bipolar IV: depression associated with hyperthymic temperament
Bipolar V: recurrent depressions that are admixed with dysphoric hypomania
Bipolar VI: late onset depression with mixed mood features, progressing to a dementia-like syndrome




Any experienced practitioner of clinical psychopharmacology will attest to the fact that a majority of patients presenting in office or outpatient settings with symptoms of mood disturbance, anxiety and/or depression do not meet strict DSM diagnostic criteria. Their symptoms often do not conform to the time constraints required and they tend to fluctuate over time. Anxiety and depression are likely to have atypical manifestations. Hypomania, when present, tends to be of the dysphoric (irritable) variety rather than the euphoric hypomania described in the DSM. Many of these patients have previously been started on antidepressants or anxiolytic drugs by their primary care physicians or by other psychiatrists. Either they have failed to respond to antidepressant trials (antidepressants are now commonly used for both anxiety and depression), or else they have had only a partial response. Sometimes the antidepressant treatment has worsened the depression or precipitated an episode of euphoric or dysphoric hypomania. Some patients present taking high doses of benzodiazepine tranquilizers, which no longer contain their anxiety. All of the above described patients populate the soft bipolar spectrum. Until a subtyping schema for soft bipolars is widely accepted, I prefer to be known as a “lumper” and to use the term Bipolar Spectrum Disorder to describe these patients. This terminology is also used by Nassir Ghaemi, a leading academician and clinical researcher who writes on this topic. It is finding wider acceptance among today’s practitioners.


Presenting Symptoms

* Episodic mood instability – these patients manifest lifelong episodes of mood swings starting around adolescence. The mood shifts unpredictably among several distinct mood poles: brief depressions lasting hours to one or two days, brief euphorias, brief dysphoric or irritable episodes, brief paranoid episodes, episodes of rage or intense uncontrollable anger, episodic anxiety equivalents (panic attacks, phobias or obsessive ruminations ). This multiplicity of mood options begs the very issue of bipolarity. It appears that multipolar mood disorder might be a more accurate designation for the soft bipolar spectrum.

* Episodic atypical depression – bipolar depressions can manifest the entire gamut of endogenous, nonendogenous and/or atypical depressive symptomatology, and they are always recurrent over time . Soft bipolar depressions usually show atypical depressive features. Patients are mood responsive, which means that they respond to favorable circumstances with a temporary lifting of the mood that can last hours to a day or two before returning to the depressed state. Other symptoms may include eating too much, sleeping too much, feeling worse towards evening and intense tiredness or lethargy. Anxiety and its subtypes (phobias, panic attacks, OCD ) frequently co-exist with atypical depression , as does episodic mood instability. There are a number of atypical depressive subtypes that are distinguished by special features. Since they are often episodic and associated with mood instability, they should be viewed as part of the soft bipolar spectrum. Included are the following: seasonal affective disorder – winter-onset atypical depressions; premenstrual dysphoric disorder – atypical depression associated with irritability, mood swings and dysphoria which occurs a week to ten days on either side of the menstrual period; hysteroid dysphoria – atypical depression mainly in women with histrionic personality features, whose episodes are precipitated by romantic rejection; abulic depression – atypical depression with a deficit syndrome ( apathy, amotivation, lack of will power, lack of energy, lack of pleasure in life, emotional blunting ).

* Hypomania – hypomania is of two types, euphoric and dysphoric or irritable. It is also of two durations, episodic and protracted or characterologic. Bipolar spectrum patients usually show episodic dysphoric hypomania. Euphoric hypomania feels good and is sometimes productive, but dysphoric hypomania produces irritability, emotional discomfiture, impulsiveness, temper dyscontrol and impaired judgment. It tends to interfere with interpersonal relationships and to limit productivity at work. There is a sense of inner speeding combined with restless over activity and racing thoughts, which can lead to a state of desperation. The hypomania frequently alternates with episodes of depression, and mood instability is almost always present. Sometimes brief euphoric episodes are added to the mix. The triad of irritable episodes alternating with rage episodes and paranoid episodes is characteristic of dysphoric hypomania.

* Mixed states – mixed bipolar disorder [ the simultaneous occurrence of both depressive symptoms and mania/ hypomania ] and rapid cycling bipolar disorder [ the patient experiences frequent switches from depression to mania/ hypomania and back ] often produce diagnostic confusion for treaters and treatment resistance for patients. These mixed states are found in bipolar I, bipolar II and bipolar spectrum disorders. They are more common in women and are often associated with thyroid abnormalities, lack of response to lithium (the standard treatment for bipolar I disorder) and antidepressant-induced worsening of symptoms. Outpatient diagnosis of these conditions is difficult at best, even after a detailed history is obtained. Diagnosis of mixed states is most likely to be made by a skilled diagnostician after a patient fails to respond to outpatient treatment or becomes worse on antidepressant medications and is subsequently admitted to the hospital for closer observation. Misdiagnosis of these conditions is all too common, leading to delays in effective treatment and a higher risk of suicide.


Comorbidity means the simultaneous existence of two or more medical conditions in the same patient at the same time. A high percentage of bipolar mood disorders, perhaps more than 50 %, are comorbid with other medical and/or psychiatric conditions. This can complicate both diagnosis and treatment. Significant comorbid conditions and how they may impact bipolar spectrum disorders are presented :

* Thyroid disorders – any thyroid disease that results in either a hyperthyroid or a hypothyroid clinical state can interact adversely with bipolar mood disorders. Hyperthyroidism can resemble hypomania/ mania and it can worsen pre-existing mania/ hypomania. Hypothyroidism can resemble clinical depression and it can cause pre-existing depression to be unresponsive to antidepressant medications. Treatment with lithium can produce hypothyroidism, which then interferes with the effectiveness of psychotropic medications. Subtle or subclinical hypothyroidism is often associated with the development of mixed and rapid cycling bipolar disorders. If present, it must be compensated for in order to successfully treat the mixed state. In general, the co-existence of a thyroid disorder, unless recognized and adequately treated, will interfere with effective treatment of bipolar disorder and also of major and minor depressions.

* Substance abuse – approximately 50% of all patients with bipolar disorder will experience significant alcohol and/or drug abuse at some point during the lifetime course of their bipolar illness. This is a serious problem since the use of any intoxicating substance has psychoactive effects on the brain and can worsen the bipolar condition. Additionally, substances (including excessive caffeine and nicotine use ) interfere with effective treatment of the mood disorder. Alcohol and drugs can mimic both depression and hypomanic states, but they do not cause bipolar illness. Substance use may, however, unmask a pre- existing depression or bipolar disorder. Bipolar mood disorders can and frequently do give rise to secondary alcohol and substance abuse. It is quite common that adolescents and adults who have not yet been diagnosed, attempt to self-medicate their unpredictable and uncontrollable mood swings with whatever is readily available – alcohol, marijuana, amphetamines, cocaine, opiates. By the time they reach evaluation, these patients are manifesting two separate and related conditions that feed upon one another, vastly complicating the treatment of both. I will not undertake the treatment of co-morbid bipolar/ substance abuse patients until they have undergone detoxification and have been able to remain substance-free for thirty days. The bipolar illness will not yield to treatment until the body is thoroughly cleansed of all intoxicating substances. The saving grace here is that once the underlying bipolar disorder is adequately controlled, the desire and craving to use substances is often dramatically reduced and sometime eliminated entirely.

* Attention deficit hyperactivity disorder ( ADHD ) – this condition is most frequently associated with hyperactive children, who have great difficulty sitting still and paying attention in school. It is now known to often persist into adulthood and its symptoms may overlap with those of adults diagnosed with bipolar spectrum disorder. Overlapping symptoms may include restlessness, motor hyperactivity, easy distractibility, impulsiveness, inability to concentrate or focus attention and temper dyscontrol. ADHD tends to be present continuously, whereas bipolar conditions are almost always episodic. The two conditions can co-exist in the same patient, although the incidence of their co morbidity is unknown. Stimulants such as methphenindate (Ritalin), which are the treatment of choice for ADHD, tend to worsen the symptoms of bipolar spectrum disorder. Antidepressants, especially the older tricyclic drugs s uch as desipramine and nortriptyline, may be effective for both disorders, frequently in combination with a mood stabilizing medication.

* Borderline personality disorder – these patients have elements of any or all of the DSM- IV personality disorder categories combined, leading to a stormy and unstable lifestyle. They tend to be overly dramatic, to have intense but unstable relationships, to be acutely sensitive to abandonment, to place unrealistic demands upon their families and their treaters and to exhibit self-defeating and often self-destructive behaviors. Substance abuse, self- mutilation and suicidal behaviors are frequently present. At one time they were considered to be the most difficult and demanding of all patients, and many books have been written detailing special technics of psychotherapeutic treatment. In recent years it has become apparent that a high percentage of these patients have co morbid bipolar spectrum disorders. This has proved to be a very important development for these patients. Those who are able to comply with psychopharmacologic management, and achieve mood stabilization, have been found to become amenable to standard psychodynamic psychotherapy. I have found that about 75 % of these patients will respond to combination pharmacotherapy using a serotonin reuptake inhibiting ( SSRI ) antidepressant, a mood stabilizer and an atypical antipsychotic medication such as olanzapine or risperidone. Once the mood has become stable, they can then benefit from competent psychotherapy to deal with their emotional backwash. The prognosis (future outlook ) for patients who are accepting of this treatment approach is much better than it was a few years ago. Today they are often able to get better and to lead normal lives.

* Personality disorders – many patients with bipolar illness will demonstrate some features of the various personality disorders described in DSM-IV during episodes of mood disturbance. I have encountered bipolar patients showing excessive dependency, passive aggressiveness, histrionic traits, paranoid features, narcissism or extreme self-centeredness, hypochondriasis and manipulative antisocial traits. Usually, when they are effectively treated for their bipolar conditions, these characterological features become insignificant or may disappear entirely. However, when major character pathology co-exists with bipolar disorder, treatment is often problematic. Patients with fixed personality disorders tend to be demanding, defiant, manipulative and self-defeating, often undermining the efforts of their treaters. They often abuse alcohol and drugs and they tend to be noncompliant with efforts to treat them, both pharmacologically and psychotherapeutically. They understand why they might need and benefit from treatment, but most crystallized personality disorders present an insurmountable obstacle to effective treatment. Bipolar conditions have a high prevalence in prison populations.


Bipolar I and II disorders are not difficult to diagnose, given an acquaintance with DSM-IV diagnostic criteria and the clinical presence of clear cut mania or hypomania. When bipolar disorder presents initially with depression, the diagnosis is often not made unless and until a later episode of mania or hypomania appears. The clinician should have a high index of suspicion for bipolarity when the depression is episodic and there is a family history of bipolar illness or its concomitants. These include alcoholism, drug abuse, uncontrolled episodes of rage and/or violence, suicide attempts, postpartum depression, psychiatric hospitalizations for depression or psychotic states. The diagnosis of bipolar spectrum disorder is more difficult. This is not surprising when one considers the above noted comorbidities, the nuances of the various clinical subtypes and their differing symptom sets, the rapidly fluctuating mood poles and the presence of mixed states. The basic mood disorders evaluation consists of the following elements: detailed history of the presenting symptoms, history of previous episodes and their response to treatment, family psychiatric history, medical history including results of a recent physical examination, a listing of all medications taken presently and during the past month, a listing of any psychotropic medications used now or previously including the patient’s response or non-response to these drugs and a current mental status examination. It is advisable to administer a structured clinical interview designed to detect and diagnose mood disorders. Several are now available and some are structured to detect subtypes based on certain symptom clusters. Some laboratory tests may also be helpful. The TRH Stimulation Test will assess for the presence of clinical or subclinical hypothyroidism; it has also proved helpful in the discrimination of unipolar from bipolar depressions. All the brain imaging technologies have been used to examine groups of depressives. They have shown promise in the delineation of brain changes specific to certain depressive states, including bipolar depressions. Included here are CAT scans, MRI scans, PET scans, SPECT scans and quantitative EEG with auditory average evoked potentials (brain mapping). These tests are expensive, they are not always available to psychiatrists and their findings are considered by many to belong in the realm of research. They hold promise for the emerging objectification of psychiatric diagnoses.


For a comprehensive review of the treatment of bipolar disorders, the reader may wish to access an excellent website maintained by Peter M. Brigham titled The Psychopharmacology of Bipolar Disorders.

I will present a personalized approach to the assessment and treatment of bipolar spectrum disorders. It may differ from that of other practitioners, but we psychopharmacologists tend to become intuitive and sometimes creative alchemists in our search for effective treatments. As a specialist in treatment refractory mood and anxiety disorders, most of the patients I see have failed to adequately respond to one or more trials of antidepressant and/or anxiolytic medications. They undertake an assessment that includes a specially designed clinical interview for mood and anxiety disorders, a structured symptom inventory administered and scored by computer, depression and anxiety symptom severity rating scales, quantitative EEG with averaged auditory evoked potential ( P-300 ) testing and , when indicated, a TRH Stimulation Test. Around 75% of my patients turn out to have bipolar spectrum disorder. If they have been taking benzodiazepines I inform them that they will be tapered off these drugs concomitant with their treatment. If they have recently abused drugs or alcohol , I defer treatment until they have been substance-free for thirty days. If they have complicating medical illnesses or mixed states, or if they are on very high doses of benzodiazepines (Ativan, Xanax, Valium, Klonopin, Tranxene, etc.), I usually insist on inpatient stabilization in a hospital setting that is designed for rapid detoxification and psychopharmacological stabilization. All patients are counseled regarding the importance of strict compliance with medication instructions and the need for long term follow-up monitoring. Some will require individual or group psychotherapy, but that determination is generally made after pharmacological stabilization has been achieved. The treating physician should be easy to reach by telephone. Patients may need reassurance during the early stages of treatment when they are more likely to experience transient medication-related side effects.

Patients who have been diagnosed with bipolar spectrum disorder and present with predominantly anxiety or depression symptoms are started initially on an SSRI antidepressant (Prozac, Zoloft, Paxil, Luvox, Celexa, lexapro). I have found these drugs to have excellent anti-anxiety and antidepressant efficacy and also to be reasonably good mood stabilizers. I monitor the patients every one or two weeks until they are stabilized. If symptoms of hypomania occur during the course of treatment, I add a mood stabilizer such as Depakote, Tegretol, Lamictal or Topamax, all of which are also anti-epileptic medications. If the patient fails to respond to the SSRI within four weeks or is unable to tolerate it due to side effects, I will switch to a dual neurotransmitter antidepressant (Effexor, Welbutrin, Remeron, Serzone, Cymbalta). Once the patient is mood stable and without symptoms, I monitor at one to three-month intervals. I advise every patient with this diagnosis to remain permanently on the medication at the same doses that relieved the symptoms, at the risk of symptom recurrence if they choose to discontinue their treatment.

Patients who present with clear cut symptoms of either euphoric or dysphoric hypomania, or who have a history of uncontrollable rage or violent outbursts are started on a mood stabilizer. If necessary, an antidepressant drug can be added later after the mood has been stabilized.

As previously mentioned, patients with mixed or rapid cycling states, as well as those requiring detoxification from addictive anti-anxiety drugs are treated in a hospital setting. They usually respond to combination therapy with mood stabilizers and antidepressants.

Patients with co morbid borderline personality disorder require combination therapy with a mood stabilizer, an antidepressant and an atypical anti-psychotic drug (Risperdal, Zyprexa, Seroquel, Geodon). Once they have achieved mood stability, psychotherapy, which is usually necessary, has a good chance of being effective. The outcome is good with long-term psychotherapy and permanent maintenance medication monitoring.

Patients with co morbid ADHD are treated with a mood stabilizer and an older tricyclic antidepressant such as desipramine or nortriptyline. If this combination proves unsuccessful, they may be tried on a mood stabilizer combined with a stimulant such as Ritalin, Dexidrene, Adderall, Provigil, or Concerta.

Patients with co morbid alcohol or substance abuse, who are able to remain substance-free long enough to enable adequate mood stabilization, often find their craving for substances disappears or diminishes considerably. More than 50% of these patients are able to remain off drugs or alcohol. This compares to the 10% of true recoveries reported by Alcoholics Anonymous for primary alcoholism.

Patients with co morbid personality disorders do well initially so long as they are motivated to please the treater (authority figure). When they become defiant and act rebellious, they usually undermine their treatment. Long term prognosis for this group of patients is generally poor.

Patients with co morbid medical conditions tend to remain well stabilized as long as their medical conditions are under good control. Any metabolic disruption such as fever, infection, uncontrolled diabetes, uncontrolled thyroid states, pain syndromes requiring narcotic pain killers, etc. can interfere with adequate mood stabilizing medication, resulting in recurrence of mood disorder symptoms.

Finally, a word about an old standby class of medications, that are sometimes effective for patients who have failed on or cannot tolerate any of the newer antidepressant drug classes. The monoamine oxidase inhibitor antidepressants (MAOI’s) were the first drugs discovered to be effective as antidepressants in the late 1950’s. They went into eclipse because of dangerous hypertensive reactions that occurred in patients taking these drugs who ate certain common foods (cheese reaction ) or used certain medications. It was later found that patients with atypical depression were uniquely responsive to this class of drugs if one avoided the foods and medications known to cause adverse reactions. These drugs remain a useful emergency backup for patients unable to benefit from the newer drug classes. They can be used safely if patients are educated about what foods and medications they need to avoid. MAOI drugs like Nardil, Marplan and Parnate have been very effective in controlling anxiety and depression in patients with bipolar spectrum disorder.

Psychosocial Dilemma: Suicidal Behavior Among Youngsters Treated with Antidepressants

A great deal of media attention during the past decade has centered on the occurrence of suicidal behavior among children and adolescents who have been treated with antidepressant drugs, especially the selective serotonin reuptake inhibitors. Black Box warnings now appear on the packaging inserts of most of these drugs and many physicians have become wary of treating depressed youngsters with antidepressants. In addition, several of the perpetrators in the rash of school shootings that have rocked our nation had been or were being treated with one or another of these drugs at the time of their violent crimes. These unfortunate events focused the glare of publicity on treatment issues in child and adolescent psychiatry. Where the problem really lies, I suspect, is with issues of diagnosis. However, this aspect has received scant if any public attention.

Ever since the advent of the SSRI’s , treaters have encountered an occasional patient who developed intense suicidal pre-occupation shortly after starting an SSRI. This usually resolved quickly after stopping or changing the medication. Patients who experienced this phenomenon rarely went on to actual suicidal behavior. What appeared to be happening here was that the drug induced sudden and intense obsessional ideation – a clue that we might have unmasked a bipolar spectrum disorder. Child and adolescent psychiatrists are starting to recognize that a significant number of their depressed patients, especially those demonstrating symptoms of anger / agitation / irritability, may actually fit into the soft bipolar spectrum. Multi-site research studies carried out by collaborating members of The Spectrum Project, an international consortium of academic researchers, suggest that more than half of all patients diagnosed with unipolar depression will eventually develop bipolar symtoms. Treatment with antidepressants may fail with these patients, or it may precipitate the rapid onset of bipolar symptoms. The occurrence of the latter could account for the suicidal behavior and/or the murderous rage evidenced by the troubled youngsters mentioned earlier.

Clearly, we must continuously strive to refine our diagnostic skills, to search for objective diagnostic parameters and to educate our trainees to recognize the nuances of clinical diagnoses that lie outside the confines of the official psychiatric nomenclature. At the time of my retirement from active practice in early 2005, only a few academic psychiatry training programs in the United States were teaching their residents about the soft bipolar spectrum. Since effective pharmacotherapy is contingent upon accurate diagnosis, prior educational deficiencies must be targeted and corrected by informed continuing education programs for physicians.

About the author

Dr. Arnold L. Lieber is currently retired, after practicing psychiatry for 32 years in Miami and Miami Beach. During his career as a clinician and researcher, he held the following positions: Chairman, Department of Psychiatry, St. Francis Hospital, Chairman, Department of Psychiatry, Miami Heart Institute, both in Miami Beach; Medical Director of the Clinical Neuroscience Unit of Miami Heart Institute; Clinical Associate Professor of Psychiatry, University of Miami School of Medicine; President, American Academy of Clinical Psychiatrists. He has authored or co-authored thirty scientific publications, including two books. His email address is