SEARCH

SEARCH BY CITATION

Keywords:

  • Bipolar disorders;
  • Manic–depressive disorders;
  • Epidemiology;
  • Mood stabilizer

Abstract

  1. Top of page
  2. Abstract
  3. DISEASE CLASSIFICATION AND LIFETIME PREVALENCE RATES
  4. THE CHALLENGE OF RECOGNIZING BIPOLAR DISORDERS
  5. HEALTH-RELATED QUALITY OF LIFE AND SOCIOECONOMIC BURDEN
  6. NEEDS AND PERSPECTIVES
  7. REFERENCES

Summary:  Bipolar, or manic–depressive, disorders are frequent and severe mental illnesses associated with considerable morbidity and mortality. Epilepsy and bipolar disorder could probably share some aspects of pathophysiology because manic as well as depressive symptoms are seen in patients with seizures, and a number of antiepileptic drugs are effectively used in the acute and prophylactic treatment of bipolar disorder. Epidemiologic research suggests a dimensional composition of bipolar illness at the population level. Apart from the DSM-IV diagnostic features of bipolar I (mania and depression) and bipolar II (hypomania and depression), the concept of bipolar spectrum disorders comprises a range of bipolar conditions with less obvious manifestations with estimated lifetime prevalence rates ranging from 2.8 to 6.5%. Expanding the definition of bipolar II disorders shows that half of the patients currently diagnosed with a unipolar depressive episode could suffer from unrecognized bipolar II disorder, and about the same number of mild depressive patients could be minor bipolars. Research efforts to refine the diagnostic criteria of bipolar disorder aim at an earlier and complete recognition of the disease to provide appropriate pharmacological and nonpharmacological treatment early in the course of the illness to anticipate individual suffering, suicidal behavior, and increased socioeconomic costs for society. This article also discusses risk factors, comorbid conditions, course of illness, as well as the individual and socioeconomic impact of bipolar disorders.

Conclusions: The findings suggest reconceptualizing bipolar illnesses as highly recurrent, malignant disorders that occur far more frequently than previously thought. Interdisciplinary knowledge transfer could help to increase our understanding of the pathophysiology of these disorders as well as provide grounds for better recognition and treatment of patients with manic and/or depressive symptoms.

Bipolar, or manic–depressive, disorders are frequent, severe, and often chronic mental illnesses. They are associated with considerable morbidity and mortality, and for many patients, an initial episode of mania or depression evolves into a life-long illness. To prevent recurrence, suicidal behavior, and chronicity, long-term pharmacologic treatment is indicated early in the course of the disease. Population data suggest an increased prevalence of depression and bipolar disorders in patients with seizures, and a number of antiepileptic drugs (AEDs) are used, apart from lithium, in the short-term therapy and as long-term mood stabilizers. Important epidemiologic aspects of bipolar disorders are discussed in this article.

DISEASE CLASSIFICATION AND LIFETIME PREVALENCE RATES

  1. Top of page
  2. Abstract
  3. DISEASE CLASSIFICATION AND LIFETIME PREVALENCE RATES
  4. THE CHALLENGE OF RECOGNIZING BIPOLAR DISORDERS
  5. HEALTH-RELATED QUALITY OF LIFE AND SOCIOECONOMIC BURDEN
  6. NEEDS AND PERSPECTIVES
  7. REFERENCES

Bipolar disorders are classified within the framework of the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV, published by the American Psychiatric Association), which differentiates between Bipolar I (manic or mixed episodes in the course of disease apart from depressive episodes), Bipolar II disorder (hypomania for ≥4 days but no manic states in the course of disease, apart from depressive episodes), and bipolar disorder not otherwise specified (Bipolar NOS) (1). The last mentioned includes patients experiencing bipolar features that do not meet the criteria for bipolar disorder (i.e., hypomania without depressive episodes or bipolar disorder due to medical conditions). Cyclothymia forms a distinct diagnosis requiring ≥2 years of fluctuating mood disturbances with hypomanic and depressive states insufficient in severity to meet the criteria for Bipolar I or II.

According to the International Statistical Classification of Diseases and Related Health Problems, 10th Revision (ICD-10; 2), a diagnosis of bipolar affective disorder requires manic, hypomanic, or mixed episodes. The term “other bipolar affective disorders” includes Bipolar II and recurrent manic episodes, and Bipolar NOS again comprises patients experiencing bipolar features that do not meet the criteria for bipolar disorder. A separate classification exists for manic episodes independent of bipolar disorder available in case only one manic episode has been observed. Cyclothymia is defined as persistent instability of mood involving mild depression and mood elations.

Rapid-cycling bipolar disorder is a malignant form of bipolar illness characterized by four or more disease episodes during the previous 12-month period (3). Rapid cycling affects ∼15% of bipolar patients and is established either from the beginning or develops in the course of the disease. This subtype is associated with increased depressive symptoms and risk for suicidal behavior (4).

Lifetime prevalence rates of bipolar disorder are estimated to range from 1 to 5% in the general population (5–8). The median lifetime prevalence found in 11 studies reviewed by Wittchen (9) is 1.3% (range, 0.6–3.3). Lifetime prevalence of Bipolar II disorder is suggested to range from 0.5 to 3.0% (10–14, see Table 1 for prevalence rates estimated in different studies).

Table 1. Lifetime prevalence rates of bipolar disorders
Diagnosis (DSM-IV)Study/authorYearCountryLifetime prevalence
  1. NCS, National Comorbidity Study; ECA, Epidemiologic Catchment Area study; NHANES III, 3rd National Health and Nutrition Examination Survey.

Bipolar INCS/Kessler et al.1994U.S.A.0.4
Bipolar I and IIECA/Regier et al.1988U.S.A.1.2
NCS/Kessler et al.1994U.S.A1.6
Weissman et al.1996Cross-national0.3–1.5
NHANES III/Jonas et al.2003U.S.A.1.6
Bipolar IIOliver and Simmons1985U.S.A.3.0
Weissman et al.1991U.S.A.0.6
Stefansson et al.1991Iceland0.5
Bipolar spectrumWeissman and Myers1978U.S.A.3.0
Oliver and Simmons1985U.S.A.3.3
Heun and Maier1993Germany6.5
Angst1998Switzerland2.8

Epidemiologic research suggests a dimensional composition of bipolar illness at the population level. The concept of bipolar spectrum disorders (proposed by 15; refined by 16 and 17) comprises a range of bipolar conditions with less-obvious manifestations. Estimates of lifetime prevalence rates for bipolar spectrum disorders range from 2.8 to 6.5% (10,11,18–20). Table 2 shows the classification approach used by Angst in 2003 (21).

Table 2. Bipolar spectrum conceptualized by Angst et al. 2003 (ref. 21)
ClassificationDescription
  1. DSM-III-R, Diagnostic and Statistical Manual of Mental Disorders, 3rd revised edition.

Bipolar IHospitalized mania plus major depression
Bipolar IIDSM-III-R major depressive episodes associated with
 (a) a hypomanic syndrome as defined above, or
 (b) hypomanic symptoms only
Minor bipolar disorderDysthymia, minor depression, or recurrent brief depression associated with
 (a) the hypomanic syndrome or
 (b) hypomanic symptoms only
Pure hypomania (a) a hypomanic syndrome without any diagnosis of depression, and
 (b) hypomanic symptoms only

Strong support has been given for expanding the definition of bipolar II disorders (22–24). In an attempt to validate the Zurich study data, Angst et al. challenged the DSM-IV diagnostic requirements for hypomania, suggesting that the demanded length of the hypomanic state of ≥4 days may be too restrictive. With the established DSM-IV criteria, 0.5% of patients had Bipolar I disorder, 1.7% had Bipolar II disorder, and 21.3% had Major depressive disorder. With newly defined “hard” and “soft” Zurich criteria, the prevalence of patients with Bipolar II was 5.3 and 11.0%, respectively, and the percentage of patients with Major depressive disorder was reduced to 17.1 and 11.4%. The findings imply that up to half of the patients currently diagnosed with a unipolar depressive episode could suffer from unrecognized Bipolar II disorder and about the same proportion of mild depressive patients from minor bipolar disorder (21).

THE CHALLENGE OF RECOGNIZING BIPOLAR DISORDERS

  1. Top of page
  2. Abstract
  3. DISEASE CLASSIFICATION AND LIFETIME PREVALENCE RATES
  4. THE CHALLENGE OF RECOGNIZING BIPOLAR DISORDERS
  5. HEALTH-RELATED QUALITY OF LIFE AND SOCIOECONOMIC BURDEN
  6. NEEDS AND PERSPECTIVES
  7. REFERENCES

Bipolar disorders are not easy to recognize because many symptoms overlap with other psychiatric disorders, comorbidity is common (both psychiatric and somatic), patients often lack insight into their illness (especially for hypomanic states), and stigmatizing societal environments are only slowly changing (see also 25,26). In a recent survey using a validated screening instrument (the Mood Disorder Questionnaire) in 127,000 adult U.S. citizens, the overall rate for Bipolar I and II disorders, weighted to match the 2000 U.S. Census demographics, was 3.4% (3.7% after adjustment for nonresponse bias). Supporting previous findings, only 19.8% of individuals with positive screening reported having previously been diagnosed with bipolar disorders by a physician, whereas 31.2% reported having received a diagnosis of unipolar depression (27; see also 28). Identifying hypomanic episodes is not an easy task because patients are usually not functionally impaired; instead, they often show increased achievement potential. However, hypomania is by no means benign, as it is accompanied by a high risk for suicidal behavior and social impairment and is followed by manic episodes in 10–15% of cases in the course of illness (21,29).

The scientific and clinical endeavors to refine the diagnostic criteria of bipolar disorders aim at building the groundwork for an earlier and complete recognition of the disease. This provides the chance for choosing the appropriate pharmacological and nonpharmacological treatment early in the course of the disease to anticipate individual suffering, suicidal behavior, and increased socioeconomic costs for society.

Pathophysiology

Despite major research achievements, the underlying pathophysiology of bipolar disorders, including molecular causes and mechanisms, is still obscure. Imaging studies (reviewed by Bruno in this volume) have been variable, with suggested abnormalities in prefrontal cortex, limbic regions, striatum, ventricular volume, and in some instances, cerebellum (30–34). With proton resonance spectroscopy, reduced levels of a potential marker for neuronal density and viability were found in the dorsolateral prefrontal cortex of euthymic bipolar patients (35,36). Neuroanatomic studies show decreased neuronal and glial density in prefrontal cortex that appears to be more extensive than in unipolar depression (37–40). Neurotransmitter systems probably involved in the pathophysiology of bipolar illnesses include serotonergic pathways, as well as the hypothalamic–pituitary–thyroid and –adrenal systems (41–44). Examining the mechanisms of action of mood stabilizers has provided clues to potential underlying neurobiologic abnormalities in bipolar disorders (see also 45,46).

Family studies have consistently demonstrated the genetic liability of bipolar disorders. First-degree relatives of affected individuals have a 10-fold increased risk for developing the disease compared to relatives of unaffected controls (47). Twin and adoption studies have provided evidence that genes strongly contribute to familial transmission of the illness. Early-onset bipolar disorder appears to be associated with increased familial risk, and studies have identified other putative familial subtypes, including lithium-responsive bipolar disorder and bipolar disorder with psychosis or comorbid panic disorder (48,49). McGuffin et al. (50) applied statistical models to the data of 67 twin pairs, estimating a heritability of 85% for bipolar disorder, leaving only 15% of variance to individual specific and/or nonfamilial environmental effects. They estimated the genetic correlation of bipolar disorder and depression to be 0.65 and of nonfamilial environment to be 0.59. Interestingly, 71% of the genetic variance in liability to mania was not shared by depression (50).

Implications of gender

Gender influences both presentation and course of bipolar disorders. Bipolar I disorder affects men and women with equal frequencies. However, Bipolar II appears to be more common in women. About 80–90% of patients with the malignant type of rapid-cycling bipolar disorder are women (51). Additionally, women are more likely to experience mixed mania and manic switches during antidepressant treatment (51). Comorbidity of somatic and psychiatric disorders is more common in women than in men (52). Finding the right treatment regimen can be challenging in women who wish to conceive as well as in pregnancy and breast-feeding periods. Individualized risk–benefit assessments are required to promote the health of the woman and avoid or limit exposure of the fetus or infant to potential adverse effects of medication (52).

Psychiatric and somatic comorbidities

Lifetime prevalence rates of comorbid psychiatric and medical conditions are increased in patients with bipolar disorders, complicating the treatment and overall management of the disorder (53). About 22% of bipolar patients have an additional somatic condition, and 39% a second psychiatric one (of that, two thirds are related to substance abuse; 54). The Epidemiologic Catchment Area (ECA) study demonstrated that a diagnosis of substance abuse or dependence is found more often in bipolar patients than in those with any other Axis I disorder (12). Substance-abusing bipolar patients show higher rates of suicide attempts, panic attacks, and very fast cycling (55). Goodwin and Hoven (56) found that the co-occurrence of bipolar disorders and panic attacks was associated with an earlier onset and more severe expression of the panic disorder.

Prevalence rates of both depressive and bipolar spectrum disorders in persons with epilepsy appear to be higher than those in the general population. Recent data from community samples show rates significantly above those in patients with and without other chronic diseases (57). Mania is reported less frequently than depression (58).

Age at onset and course of disease

Unlike unipolar depression, bipolar disorders are manifested to a great extent in the adolescent and young-adult years (see 59). The peak age at onset of the first symptoms of bipolar disorder is between age 15 and 19. The median age at onset of 2,839 patients with bipolar disorders recorded by the Stanley Center Bipolar Disorder Registry was 17.5 years (mean, 19.8 years; 60). Bipolar disorders can appear before puberty; however, this is rare and difficult to distinguish from other severe psychiatric disorders such as schizophrenia, unipolar depression, and attention-deficit/hyperactivity disorder (ADHD) (61,62).

Elderly patients also can have symptoms of bipolar disorders for the first time in old age. However, manic and depressive symptoms in this age group have usually been present for many years, and in mild cases, have often gone untreated. Onset of mania after age 60 is less likely to be associated with a family history of bipolar disorder and more likely secondary to medical causes other than bipolar disorders [e.g., neurologic (trauma, neoplasm, multiple sclerosis, stroke, epilepsy), endocrine (hyperthyroidism, Cushing disease), infectious (AIDS), and inflammatory (systemic lupus erythematosus) disorders (63)].

Usually a delay of 5–10 years is found between the first episode of a bipolar disorder and the age at first treatment or hospitalization (64). In a study of 450 bipolar patients, the observed latency from the first episode of bipolar illness to the start of maintenance treatment averaged 7.8 years (median, 5 years), during which time patients experienced a mean of 9.4 episodes (median, five episodes). In the years before maintenance treatment, patients were ill ∼46.5% of the time, and 28.7% of subjects had at least two episodes per year (65). Leverich et al. (66) found a mean age at first treatment of 29.2 years in their sample of 631 bipolar patients.

Manic–depressive disorders have a high rate of recurrence (>90% of individuals who have a single manic episode will have future episodes). The natural course of the illness and its phenotypic expression are highly variable. It has been estimated that without adequate treatment, the average patient experiencing the onset of bipolar disorders at age 25 years will have a few episodes; however, 10–15% of patients will undergo >10 episodes (64,67). The prevalence of residual depressive symptoms was estimated to be 40–85% in bipolar patients, showing positive associations with recurrences (68).

An early onset of bipolar disorders increases the risk for a severe course and poor outcome of the disease. The risk for having the rapid-cycling type (69), suicidal ideation, and comorbidity with substance abuse–related disorders also is increased (69–71). Other predictors of poor outcome include childhood psychopathology (72), being female (see gender differences and 73), and poor adherence to treatment (74).

Increased suicide risk

Suicidal behavior is the most serious consequence of bipolar disorders, reaching a lifetime prevalence of suicide attempts of up to 30% (5,75,76). The odds ratio for suicide attempts was 6.2 in the subjects of the ECA database versus controls (compared with 3.1 in unipolar depression (77,78). Up to 20% of bipolar patients die of suicide. Data from the Stanley Foundation Bipolar Network show that several factors are associated with an increased risk for suicidal behavior in bipolar patients, including genetic and familial variables, childhood and recent environmental adversities, loss of social and medical support, psychiatric comorbidity, and a more severe course of bipolar illness (79). An early and appropriate treatment provides the background for effective prevention of suicidal mortality (5). Lithium proved to be the most efficient drug in long-term treatment of both unipolar and bipolar patients, reducing the risk for suicidal acts in Bipolar I and II patients by 82 and 67% (80).

HEALTH-RELATED QUALITY OF LIFE AND SOCIOECONOMIC BURDEN

  1. Top of page
  2. Abstract
  3. DISEASE CLASSIFICATION AND LIFETIME PREVALENCE RATES
  4. THE CHALLENGE OF RECOGNIZING BIPOLAR DISORDERS
  5. HEALTH-RELATED QUALITY OF LIFE AND SOCIOECONOMIC BURDEN
  6. NEEDS AND PERSPECTIVES
  7. REFERENCES

Studies assessing the impact of bipolar disorders on the health-related quality of life show outcomes resembling or even worse than that seen in patients with chronic somatic and some other psychiatric diseases (81–84).

Adding to the suffering of bipolar patients and their families, the condition significantly increases costs for society (85). At present, estimated annual total disease costs in Germany amount to ∼Euro 5.9 billion, mainly arising from indirect costs (86); in the United States, the costs for society in 1991 were calculated to reach ∼US$ 45 billion (87). Work impairment in bipolar patients is even greater than that seen in unipolar depressed patients (88–90).

NEEDS AND PERSPECTIVES

  1. Top of page
  2. Abstract
  3. DISEASE CLASSIFICATION AND LIFETIME PREVALENCE RATES
  4. THE CHALLENGE OF RECOGNIZING BIPOLAR DISORDERS
  5. HEALTH-RELATED QUALITY OF LIFE AND SOCIOECONOMIC BURDEN
  6. NEEDS AND PERSPECTIVES
  7. REFERENCES

The findings discussed suggest a reconceptualization of bipolar illnesses as highly recurrent and malignant disorders that occur more frequently than previously thought. Because early recognition and intervention may be able to ameliorate adverse outcomes, efforts should be taken to implement structured screening and diagnosis tools and treatment guidelines, including both pharmacological and nonpharmacological strategies into research and clinical practice. Interdisciplinary transfer of knowledge is required, bearing in mind high comorbidity rates of bipolar disorders with other medical, neurologic, and psychiatric illnesses and the likely shared aspects of pathophysiology.

REFERENCES

  1. Top of page
  2. Abstract
  3. DISEASE CLASSIFICATION AND LIFETIME PREVALENCE RATES
  4. THE CHALLENGE OF RECOGNIZING BIPOLAR DISORDERS
  5. HEALTH-RELATED QUALITY OF LIFE AND SOCIOECONOMIC BURDEN
  6. NEEDS AND PERSPECTIVES
  7. REFERENCES