The Bipolar Spectrum

Diagnosis or Fad?

Joel Paris

• Publisher: Routledge
• Year: 2012
• ISBN:
  9781136325328
• Edition: 1
• Topic: Psychology
• Subtopic: Abnormal Psychology
• Index: Psychology

In the 1970s, author Joel Paris was one of the first doctors in his hospital to prescribe lithium to a psychiatric patient. In the wake of the drug's success, both in that case and countless others, why this book? As Dr. Paris' historical examination of bipolar diagnosis and critique of the spectrum demonstrates, medicine has often been prone to fads that are assumed correct until proven wrong by science. This book opens discussion about the overdiagnosis of bipolar disorder and the negative impact of this development on clinical care. Dr. Paris explores why patients are being classified as bipolar on dubious grounds and are being prescribed drugs they do not need. He explains the differences between bipolar disorder and depression without mania, personality disorders characterized by unstable mood, and impulsive disorders. A separate chapter discusses the unique issues present in the field of child psychiatry. Fads remain popular as long as they answer elusive and complex questions. Unfortunately, the bipolar spectrum being used to explain a wide variety of psychopathological phenomena has caused classic bipolar disorder to become almost lost in the shuffle. Combining research findings and personal experiences, Dr. Paris documents the damage of overdiagnosis and explores alternative treatments patients could benefit from.

PART I

Concepts

1
THE BIPOLAR DIAGNOSIS

Emil Kraepelin: The Pioneer

It has been recognized since antiquity that some mental illnesses are characterized by dramatic shifts in mood. This clinical picture was known to Hippocrates (Angst & Sellaro, 2000), but descriptions of a specific illness marked by alternating mania and depression only appeared in the nineteenth century (Healy, 2009).

For modern psychiatrists, the pioneer and crucial figure in defining this illness was Emil Kraepelin (1856–1926). Serving as a professor at several leading German universities, Kraepelin was the most seminal figure in twentieth-century psychiatry (Shorter, 1997), and his influence continues to be felt today. Before Kraepelin’s time, the only meaningful distinction in diagnosis was between psychosis and non-psychosis. Within psychoses, classification was a confusing hodge-podge of overlapping syndromes. Kraepelin, although not a researcher in a contemporary sense, was a hero of systematic observation, and the first person to make sense out of mania.

Kraepelin was the first psychiatrist to diagnose mental illness by delineating a specific course rather than diagnosing by symptoms alone. He reorganized the psychoses into two overarching categories. Schizophrenia (then called dementia praecox) had a continuous course that became steadily worse with time. In contrast, manic-depression had a cyclical course in which patients could be normal, or at least near-normal, between episodes (Kraepelin, 1921). While these distinctions are not absolute, they remain crucial for classification.

Kraepelin died in 1926, but his ideas never lost influence in Europe. In North America, during the period after World War II when psychoanalysis was dominant, Kraepelinian psychiatry went into a temporary eclipse (Shorter, 1997). In the United States, the classification of mental disorders was often seen as a dry and unrewarding subject. This was reflected in a general lack of interest in earlier editions of the Diagnostic and Statistical Manual of Mental Disorders (DSM-I and DSM-II). Psychoanalysts, as well as many other psychiatrists, viewed psychopathology as a reaction to circumstance, not as a biologically determined pattern.

Recent decades have been marked by a sea change in the orientation of American psychiatry. DSM-III was associated with the triumph of what has been called a “neo-Kraepelinian” school of thought (Klerman, 1986). Psychiatry took a U-turn away from psychoanalysis and rejoined the mainstream of medicine, becoming oriented towards neuroscience and psychopharmacology. As drug treatment advanced, it became possible to offer specific treatments for specific diagnoses. That is why classification gained importance. That is also why Kraepelin came back in a big way in the 1970s. Shorter (1997) has described him as the most important psychiatrist of the twentieth century. Under the influence of neo-Kraepelinian ideas, diagnosis became tied to family history, outcome, and the possibility of discovering biological markers of disease.

Kraepelin was the first person to suggest the existence of a bipolar spectrum, which he envisaged as a dimension of mood disturbances ranging from psychotic illness to near-normal variants. As we will see later in this book, contemporary psychiatrists have resurrected and greatly expanded this idea. But since Kraepelin’s experience was confined to severely ill patients who were hospitalized, he might have been surprised by claims that 10% or more of the general population suffer from some form of bipolar disorder.

Changing the Name

For decades after Kraepelin, the concept of manic-depressive illness underwent little change. But under the influence of another German psychiatrist, Karl Leonhard (1902–1988), the name was changed. The term “bipolar disorder,” introduced in 1957 (Goodwin & Jamison, 2007), was one of many suggestions made by Leonhard (1999) for reclassifying mental disorders, but the only one to be generally adopted. Leonhard separated mood disorders into a unipolar type (depression or mania but not both) and a bipolar type. While unipolar depression is common, unipolar mania is very rare; however, almost all bipolar patients eventually suffer from both depressive and manic episodes.

With time, the term “bipolar” came to replace manic-depression entirely. While the older terminology points to the necessity of manic episodes and suggests a severe illness, the newer label is more neutral. The term “bipolar” has made it easier to expand the boundaries of the disorder.

Before and After Lithium

Fifty years ago, bipolar disorder could often go unrecognized, particularly in North America. In the 1960s, a “New York–London Study” documented wide divergences in diagnostic practices between American and British psychiatrists (Cooper, Kendell, & Gurland, 1972). When presented with identical filmed interviews of psychotic patients, the Americans diagnosed most as schizophrenic, while the British diagnosed the same patients as having manic-depression. American psychiatrists at the time had a very broad concept of schizophrenia, and almost anybody with psychotic symptoms, especially paranoid delusions, tended to get that diagnosis. Yet a few years later, the Americans came around to see things the British way. The reason was that bipolar disorder now had a specific treatment. Psychiatry had arrived in the age of lithium.

Treatment with lithium carbonate was the most important event in the history of bipolar disorder. An Australian psychiatrist, John Cade (1949), was the first to report that this simple salt offered an effective treatment for mania. However his discovery was not followed up, mainly because lithium had been tried for cardiac patients and discarded due to side effects. The Danish psychiatrist Mogens Schou (2001) reintroduced lithium in the late 1960s. The results of this treatment could sometimes be miraculous (see the Introduction to this book).

Before lithium, it did not make much difference whether a psychotic patient was diagnosed with mania, schizophrenia, or some other disorder. Either way, treatment, at least since the 1950s, had depended mainly on antipsychotic drugs. But these agents did not prevent recurrences of mania. If bipolar disorder responded specifically to lithium, and if many patients who were lithium-responsive could be maintained without antipsychotic drugs, the clinical advantage was enormous.

With a lower threshold for a bipolar diagnosis, some patients who had not been recognized as having manic episodes were rediagnosed and given more appropriate treatment. But patients unlikely to respond to lithium (or other mood stabilizers) were also prescribed these drugs, sometimes for long periods. Most suffered from other psychotic conditions, particularly schizophrenia. If the original diagnosis was correct, these patients would not benefit. The problem was that it was impossible to determine if adding lithium to an antipsychotic regime was making any difference. This was the first indication that enthusiasm for a bipolar spectrum could create trouble.

Variants of Bipolar Disorder

Illness is not the same in every patient. That is most certainly true for bipolar disorder. For example, cases that begin early in life tend to have a poorer prognosis (Goodwin & Jamison, 2007). Other cases start late and are less severe. Some patients have a strong family history of disorder, while others do not. Sometimes response to treatment can define clinical subgroups. Patients who respond to lithium may have a fundamentally different form of illness than those who do not (Alda, 1999).

The most important variant of classic manic-depression is bipolar-II disorder. This category, described in the literature for decades (Dunner, Fleiss, & Fieve, 1976), was only officially accepted as a diagnosis in the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV; American Psychiatric Association–APA, 1994). Essentially, bipolar-II is a milder form of the illness, with hypomanic rather than full manic episodes. It has been suggested that this disorder could have a separate etiology and pathogenesis (Goodwin & Jamison, 2007). As we will see, bipolar-II has boundary problems, mainly relating to the definition of a hypomanic episode.

Manic episodes, the defining feature of bipolar-I disorder, are fairly unmistakable. They are characterized by what has been called a “classical triad” of elevated affect, psychomotor excitement, and racing thoughts, all of which need to last for at least a week. Other characteristic symptoms include grandiose ideas, a decreased need for sleep, distractibility, and a variety of impulsive behaviors (such as overspending). However, since manic episodes are usually associated with psychosis, and with either grandiose or paranoid delusions, they have to be differentiated from schizophrenia. The main twist is that some manic patients present more with irritability than with euphoria (Winokur & Tsuang, 1975), a possibility that DSM-IV later included in its definition. That does not, however, mean that everyone who is highly irritable must be manic. Irritability is also a common symptom in depression, in which it is more related to severity than to latent bipolarity (Perlis et al., 2009).

Bipolar-II requires episodes of both major depression and hypomania. As currently defined, it is a heterogeneous construct (Vieta & Suppes, 2008) that is often not diagnosed with great precision. If you have never had hypomania, you cannot have bipolar-II. That makes the definition of a hypomanic episode a crucial issue.

In DSM-IV-TR (APA, 2000), the requirement is for “a distinct period of persistently elevated, expansive, or irritable mood, lasting throughout at least four days, that is clearly different from the usual non-depressed mood.” To be classified as hypomanic, patients must then have at least three of the following (four if the mood is irritable and not euphoric): inflated self-esteem or grandiosity, decreased need for sleep, talking more than usual or pressure to keep talking, flight of ideas or the subjective experience that thoughts are racing, distractibility, increase in goal-directed activity (either socially, at work or school, or sexually), psychomotor agitation, and excessive involvement in pleasurable activities that have a potential for painful consequences.

A hypomanic episode also must be associated with an unequivocal change in functioning that is uncharacteristic of the person when not symptomatic, and (crucially) is observable by other people. In contrast to full mania, hypomania need not be severe enough to cause marked impairment in social or occupational functioning, or to necessitate hospitalization; and it does not show psychotic features.

If all these criteria are met, bipolar-II can be diagnosed. This diagnosis forms a distinct category of illness; every clinician will have seen such cases. But as Dunner and Tay (1993) have noted, patients do not always report mental states accurately. For example, patients may remember a period in which mood was unstable as continuous, rather than variable. This makes it crucial to interview relatives or friends to determine whether hypomania has been present. It is also worth noting that patients with both bipolar-I and bipolar-II may spend more time feeling low than high. Since bipolar depression is difficult to treat, it has been the subject of a great deal of recent research (Nivoli et al., 2011).

Another bipolar variant is a “mixed state,” defined by DSM-IV-TR as at least a week in which a patient meets criteria for both major depression and mania. This category is problematic. Research is thin, and it is difficult to distinguish a mixed state from agitated depression (Goodwin & Jamison, 2007), or from agitation associated with other mental disorders. The result is that mixed states are diagnosed too readily, placing patients with other forms of psychopathology in the bipolar spectrum.

Finally, DSM-IV-TR allows for a diagnosis of bipolar disorder not otherwise specified (NOS). The DSM system has an NOS category in each major grouping to account for patients who have some but not all features of specific diagnostic entities. In this case, the diagnosis describes any disorder with bipolar features that does not meet criteria for any of the above types.

One of the examples given in the DSM manual for a bipolar-NOS diagnosis is “rapid cycling of mood over brief durations.” But while the subcategory of rapid-cycling bipolar disorder (defined as more than four episodes in a year) has been much studied (Bauer, Beaulieu, Dunner, Lafer, & Kupka, 2008), there is no reference to shifts over brief durations, and it is doubtful whether the term “rapid cycling” should be used to describe such phenomena. What is actually being described is affective instability, which is not necessarily bipolarity (see Chapter 3).

Bipolar-NOS itself has not been well researched. One chart review study (Bader & Dunner, 2007) found that some patients have first-degree relatives with classic bipolar disorder, but many do not. There is a case to be made that subclinical features of bipolarity can be a risk factor for developing the full disorder. However, that does not mean that one can jump from these clinical features to a firm diagnosis. The strength of the relationship depends on how broadly the diagnosis is used. The problem is that the NOS diagnosis as currently defined leaves the door open for clinicians to diagnose bipolarity in almost any patient they wish.

These definitions will not change dramatically in DSM-5, except for one likely revision: the four-day rule for hypomania. This time scale has been challenged on the grounds that it is arbitrary. And this criticism is absolutely correct—except that any other length would be equally arbitrary. A two-day instead of a four-day rule may well prevail in the new manual. Yet one would still need to determine whether mood has been consistently abnormal within that time frame. As Chapter 3 shows, patients with symptoms of affective instability rarely remain in the same mood for as long as two full days.

Redefining hypomania to include even briefer periods of mood change (as short as a day or less), as well as reducing the number of required symptoms, would lead to a radical expansion of bipolar diagnosis. That is already happening—in spite of the definitions provided by DSM-IV. Clinicians do not necessarily pay attention to all these details. Instead of following algorithms, diagnoses tend to be based on features considered to be characteristic of bipolar disorder (especially mood swings of any kind).

Diagnostic Validation

How do psychiatrists go about validating the diagnosis of any mental disorder? In modern medicine, diagnosis is initially based on signs and symptoms, but is usually confirmed by laboratory tests, imaging procedures, and pathological findings. Since the nineteenth century, under the influence of the German pathologist Rudolf Virchow and the Canadian-American internist William Osler, it has been accepted that medical diagnoses should be grounded in objective data.

Psychiatrists have not been able to meet that standard. They still depend almost entirely on clinical observation to make diagnoses. In this respect, they are in much the same position as other specialists were 100 years ago. (There are also important syndromes in internal medicine, such as migraine, that are no better understood than mental disorders.) One can quantify the measurement of signs and symptoms with structured interviews or by using clinical rating scales. But such procedures can hardly be considered “gold standards” since they can be no more accurate than the clinical phenomena by which they are validated. While physicians will always continue to assess signs and symptoms, the convergence of a characteristic clinical picture with biological findings provides the data that make diagnoses valid.

Over 40 years ago, Eli Robins and Samuel Guze, two psychiatrists at Washington University in St. Louis, who founded the neo-Kraepelinian school, proposed ways to validate diagnoses. Robins and Guze (1970) recommended a combination of disciplined observation, supported by a characteristic course of illness and buttressed by the eventual discovery of laboratory methods.

Robins and Guze listed five pathways to validity. The first was precise clinical description. That principle became the basis of the DSM-III system, published 10 years later. The second pathway was laboratory studies to identify biological markers. That idea turned out to be rather futuristic; biological markers did not exist for any categories of mental illness in 1970, and psychiatrists still have none. The third criterion was clear delineation from other disorders. Yet in practice, patients often meet criteria for multiple diagnoses that can lead to a massive degree of “comorbidity.” This term, which implies the presence of more than one disorder in the same patient, is misleading because separate disorders can have overlapping symptoms. The problem is that DSM manuals have no way to determine what is primary and what is secondary, so that multiple diagnoses are both allowed and encouraged. The fourth pathway is a characteristic outcome in follow-up studies—Kraepelin’s approach. That criterion has been frustrated by heterogeneity in outcome, raising questions about the validity of many categories. The fifth Robins–Guze criterion was a genetic pattern in family history studies, a concept of relevance to bipolar disorder. Identifying bipolar relatives has often been used to delineate the boundaries of the spectrum. However, problems with defining bipolar disorder itself makes its identification even more difficult in a relative than in a proband.

Psychiatry is still struggling to achieve validity for most diagnoses. The Robins–Guze criteria represented a noble attempt to offer useful general guidelines. But until specific and sensitive biological markers are identified, we cannot be sure whether patients do or do not merit any diagnosis. For this reason, the DSM manual should ...

Table of contents

• Cover Page
• Half Title Page
• Title Page
• Copyright Page
• Dedication
• Contents
• Introduction
• Part I Concepts
• Part II Disorders
• Part III Implications
• References
• Index