| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Department of Psychiatry and Behavioral Sciences, Duke University Medical Center, Durham, North Carolina.
Address correspondence and reprint requests to Ranga Krishnan, MB, ChB, Department of Psychiatry and Behavioral Sciences, Duke University Medical Center (3050A), 4584 Hospital South, Box 3950, Durham, NC 27710. E-mail: krish001{at}mc.duke.edu
| ABSTRACT |
|---|
|
|
|---|
Methods: Articles were obtained by searching MEDLINE from 1970 to present with the following search words: bipolar disorder AND, comorbidity, anxiety disorders, eating disorder, alcohol abuse, substance abuse, ADHD, personality disorders, borderline personality disorder, medical disorders, hypothyroidism, obesity, diabetes mellitus, multiple sclerosis, lithium, valproate, lamotrigine, carbamazepine, atypical antipsychotics. Articles were prioritized for inclusion based on the following considerations: sample size, use of standardized diagnostic criteria and validated methods of assessment, sequencing of disorders, quality of presentation.
Results: Although the literature establishes a strong association between bipolar disorder and substance abuse, the direction of causality is uncertain. An association is also seen with anxiety disorders, attention-deficit/hyperactivity disorder, and eating disorders, as well as cyclothymia and other axis II personality disorders. Medical disorders accompany bipolar disorder at rates greater than predicted by chance. However, it is often unclear whether a medical disorder is truly comorbid, a consequence of treatment, or a combination of both.
Conclusion: To ensure prompt, appropriate intervention while avoiding iatrogenic complications, the clinician must evaluate and monitor patients with bipolar disorder for the presence and the development of comorbid psychiatric and medical conditions. Conversely, physicians should have a high index of suspicion for underlying bipolar disorder when evaluating individuals with other psychiatric diagnoses (not just unipolar depression) that often coexist with bipolar disorder, such as alcohol and substance abuse or anxiety disorders. Anticonvulsants and other mood stabilizers may be especially helpful in treating bipolar disorder with significant comorbidity.
Key Words: bipolar disorder psychiatric comorbidity medical comorbidity medical complications anxiety disorders alcohol and substance abuse
Abbreviations: ADHD = attention-deficit/hyperactivity disorder; BMI = Body Mass Index; CADASIL = Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy; CNS = central nervous system; DSM-IV = Diagnostic and Statistical Manual of Mental Disorders, fourth edition; ECA = epidemiologic catchment area; MDQ = Mood Disorder Questionnaire; NHANES III = National Health and Nutrition Examination Survey, 19881994; MS = multiple sclerosis; OCD = obsessive-compulsive disorder; PCOS = polycystic ovarian syndrome; PTSD = posttraumatic stress disorder; TRH = thyrotropin-releasing hormone; TSH = thyroid-stimulating hormone; VCFS = velocardiofacial syndrome.
| INTRODUCTION |
|---|
|
|
|---|
Relevant comorbidity can be illustrated in clinical, familial, or epidemiological settings. Clinical comorbidity is defined as one disorder influencing the course of, outcome of, or treatment response to a second coexisting disorder (1). In familial comorbidity, which may be caused by genetic linkage or shared environmental exposures, relatives of a patient with one disorder are more likely to have the comorbid disorder than are individuals in a family without either disorder (1). This review synthesizes the available knowledge on both the psychiatric and medical comorbidities that exist with bipolar disorder.
| METHODS |
|---|
|
|
|---|
Psychiatric Comorbidity
Bipolar disorder often coexists with other Axis I and Axis II disorders (Table 1), which hinders diagnosis and complicates treatment. Reported rates of lifetime psychiatric comorbidity in bipolar I samples range from 50% to 70%, perhaps reflecting study populations of severe and refractory patients, many of whom are hospitalized and not representative of most community-based bipolar patients (2).
|
In a Stanley Foundation Bipolar Treatment Outcome Network study of 288 patients with bipolar disorder, 187 patients (65%) also met DSM-IV criteria for at least 1 comorbid lifetime Axis I disorder, whereas 42% had 2 or more Axis I comorbidities, and 24% had 3 or more (1). Axis I comorbidity was correlated with earlier age at onset of affective symptoms, rapid cycling, worsening severity of episodes over time, and higher incidence of drug misuse in first-degree relatives (1).
Other associations with comorbidity include poorer overall outcome, high rates of suicidality (3), depression onset (4), and less favorable response to lithium (5). In a study from the University of Barcelona in Spain, bipolar patients with psychiatric comorbidity had more mixed features, depressive episodes, and suicide attempts; poorer outcome and treatment compliance; and greater frequency of depressive onset than their counterparts without psychiatric comorbidity (2).
Comorbid psychiatric disorders usually precede the onset of bipolar disorder. In a 10-year, community-based, longitudinal study of 717 youths and their mothers, adolescent psychiatric disorders were associated with increased risk for early adulthood bipolar disorder, even after accounting for adolescent bipolar disorder (6). However, in another study of 77 patients hospitalized for bipolar disorder for the first time, obsessive-compulsive disorder (OCD) occurred more often after the onset of bipolar disorder (7). In another large cohort, substance use disorders also tended to follow the onset of bipolar disorder (8).
Ongoing research should shed additional light on the prognostic and treatment response implications of Axis I comorbidity in bipolar disorder. Although the pathophysiology resulting in comorbidity is still unknown, bipolar disorder may be a risk factor for some comorbid Axis I diagnoses or vice versa; coexisting conditions may share similar end states but have different etiologies, or all comorbid diagnoses may be related and may share a common underlying pathophysiology (1). Because bipolar symptoms of mood disturbance, anxiety, and psychosis overlap with those of other psychiatric conditions, evaluating patients during remission may help differentiate among these possible mechanisms.
Alternatively, poorly defined or "fuzzy" categories of DSM-IV resulting in frequent overlap with similar disorders may give rise to artifactual comorbidity. For example, the symptoms of generalized anxiety disorder overlap with those of depression, which may explain some of the comorbidity between bipolar disorder and anxiety disorder. Both symptom overlap with other disorders and true comorbidity complicate the diagnosis of early-onset bipolar spectrum disorder (9).
Age
Several studies suggest a lower rate of psychiatric comorbidity among geriatric populations, but a cross-sectional study of 182 depressed subjects aged 60 and older seen in primary care and psychiatric settings showed relatively high rates of current (23%) and lifetime (35%) anxiety disorders (10).
Bipolar I/II
A large Stanley Foundation study showed no differences in psychiatric comorbidity between patients with bipolar I and II disorder (1). However, patients with bipolar II disorder may be up to five times more likely to have comorbid migraine (11). In studies with relatively few bipolar II subjects, rates of lifetime substance abuse or dependence were higher in bipolar I than in bipolar II subjects (60.7% vs. 48.1% for any drug, 46.2% vs. 39% for alcohol, 11% vs. 5.6% for cocaine, and 20% vs. 5.6% for marijuana) (12).
Gender
Except for substance use disorders, medical and psychiatric comorbidity is more common in women than in men, and adversely affects recovery from bipolar disorder more often in women. Thyroid disease, obesity, and anxiety disorders are more prevalent in bipolar women than in bipolar men (13). Bipolar men are much more likely to have a comorbid alcohol or substance use disorder than bipolar women, although bipolar women have much higher prevalence of alcohol and substance use disorders than do women in community samples (14). Men and women with bipolar disorder have similar rates of migraine headache (15), whereas migraine is more frequent among women than men in community samples. Bipolar men treated with lithium are less likely than women to develop hypothyroidism (16).
Alcohol and Substance Use Disorders
Despite evidence suggesting that substance use follows bipolar disorder (8), the directionality remains uncertain (2,12). In the ECA study of a household sample of 20,291 adults aged 18 and older from 5 US cities, and 56.1% of bipolar patients had alcohol or substance abuse or dependence (17). In the National Comorbidity Survey of a community-based US household sample of 8,098 people aged 15 to 54 years, 6.5% of alcoholic men and 10.6% of alcoholic women had a lifetime history of mania (18). In 392 patients hospitalized at John Umstead Hospital (Butner, NC) for manic or mixed episodes, rates of lifetime substance abuse were 48.5% for alcohol, 43.9% for other drugs, and nearly 60% for any substance (19).
Other estimates of comorbid lifetime substance abuse in bipolar disorder range from 6% to 69% for alcohol abuse, with rates of at least 30% in most studies (19), and from 14% to 60% for drug abuse (19). Males with bipolar disorder have higher rates of substance abuse than females (59.7% vs. 37.8% for alcohol; 54.5% vs. 33.8% for other drugs) (19). Lifetime marijuana abuse rates were higher in males than in females, but rates of cocaine and opiate use were similar (19).
A chart review of 131 patients meeting DSM-IV criteria for bipolar disorder revealed that bipolar men were twice as likely as bipolar women to have a comorbid substance use disorder. However, women with bipolar disorder had four times the rate of alcohol use disorders and seven times the rate of other substance use disorders than women from community-derived samples (14). Among bipolar patients committing suicide, 10 of 18 male victims, but none of 13 female victims, had alcohol dependence (20). Older patients had lower rates of active substance abuse.
Complications of substance abuse in bipolar disorder include higher rates of mixed and rapid cycling mania (21), prolonged recovery time (21), higher prevalence of medical disorders including liver disease (18), more suicide attempts (22), and suicide (20,23,24). Poor premorbid adjustment in childhood may predispose bipolar patients to develop substance abuse comorbidity and an increased risk for rapid cycling and suicide attempts (25). In a genetic linkage study of 337 subjects with major affective disorder in 71 families, the lifetime rate of attempted suicide was 38.4% in subjects with both bipolar disorder and alcoholism and 21.7% in those without alcoholism (p < .005) (24).
The high frequency of substance abuse in bipolar disorder mandates urine testing and third-party corroboration of history in all bipolar patients. Prescription of benzodiazepines to patients with bipolar disorder and comorbid substance use disorders is controversial, because it may double rates of benzodiazepine abuse compared with those who were not prescribed benzodiazepines (15% vs. 6%) (26). Drug abuse may lead to misdiagnosis of bipolar disorder, because patients intoxicated with stimulants such as cocaine or methamphetamine can appear manic.
Anxiety Disorders
Given the prevalence of anxiety symptoms and the fuzzy overlap in terms of common symptoms in both manic and depressive episodes, the frequent coexistence of anxiety disorders with bipolar disorder is not surprising. In the National Comorbidity Survey, 92% of those who met criteria for lifetime bipolar I disorder also met criteria for a lifetime anxiety disorder, including comorbid social phobia (47.2%) and posttraumatic stress disorder (PTSD) (38.8%) (27). This was a retrospective study, which may be limited by recall bias. These comorbidity rates were higher than in other studies, possibly because of the exclusion of subjects 55 years and older, which is the group with the lowest rates of comorbidity (28).
In a Hungarian epidemiological study of a random sample of 149 adults, aged 18 to 64 years, interviewed using the Diagnostic Interview Schedule for DSM-III-R, the rate of lifetime anxiety disorder in patients with bipolar disorder was 48.9%, with generalized anxiety disorder the most common and panic disorder the second most common (29).
Comorbid anxiety disorders may be unrecognized and hence neglected in children and adolescents with bipolar disorder. A community study showed that anxious high school students were 7 times more likely to have comorbid bipolar disorder than were students without anxiety disorder. In a longitudinal study of depressed adolescents, half of those with comorbid anxiety disorders, but none without anxiety, developed bipolar disorder (30). This suggests that anxiety disorder may be a risk factor for bipolar disorders. Of 43 outpatient youths with bipolar disorder, only 11.6% had no psychiatric comorbidity, and only 23.5% did not have a comorbid anxiety disorder. Patients with comorbid anxiety disorders more often reported pharmacologically induced hypomanic episodes (31).
Persistent subsyndromal symptoms in patients with bipolar disorder are associated with high rates of comorbid anxiety disorders and eating disorders (32).
Anxiety disorders and substance use disorders that cluster together in bipolar patients can pose a therapeutic challenge as well as a diagnostic dilemma (26). Preferable therapeutic options may include a cognitivebehavioral approach, or use of mood stabilizers before carefully attempting the addition of antianxiety compounds with a relatively lower risk of mania induction, such as selective serotonin reuptake inhibitors rather than tricyclic antidepressants (5).
Panic Disorder
In the ECA survey (33), 20.8% of pooled bipolar I and II patients had lifetime panic disorder, which was significantly greater than in patients with major depression (10.0%) and in the total population (0.8%). Panic disorder, therefore, appears to be epidemiologically comorbid with bipolar disorder.
In a study of 606 subjects with bipolar disorder from the NIMH Bipolar Disorder Genetics Initiative, familial panic and the diagnosis of panic disorder in an individual subject increased the odds for rapid mood switching (34). In a smaller study, 12 of 13 bipolar patients with depressive mania, but none with pure mania, met the criteria for intraepisode social phobia and panic disorder concurrently (p < .0001) (35).
ObsessiveCompulsive Disorder
In the ECA survey, 21.0% of patients with bipolar disorder had a lifetime diagnosis of OCD, compared with 2.5% of the general population and 12.2% of those with major depression (36). OCD typically occurs after the onset of bipolar disorder (6), suggesting epidemiological comorbidity. Bipolar subjects with OCD were more likely than those without OCD to have higher lifetime rates of thoughts of death and suicide, suicide attempts, and panic disorder (27).
In contrast to these findings, the Suffolk County (NY) Mental Health Project found obsessivecompulsive and panic symptoms at baseline in 10% to 20% of psychotic inpatients in each of 3 groups studied. In the bipolar group, only 2% of patients met criteria for a lifetime diagnosis of OCD, although 13% had exhibited symptoms of the disorder. Only 3% met criteria for a lifetime diagnosis of panic disorder, whereas 19% experienced symptoms of panic disorder (37). The clinical significance of the comorbidity between OCD and bipolar disorder is therefore unclear (7).
Compared with patients with pure bipolar disorder, those with comorbid OCD had higher rates of panic disorderagoraphobia but lower rates of attention-deficit/hyperactivity disorder (ADHD)conduct disorder (38).
Social Phobia
In the National Comorbidity Survey (36), having bipolar disorder quadrupled the risk of having social phobia, with an odds ratio of 4.6. However, the baseline rate of social phobia in this survey was also high (13.3%), falling just behind major depressive episode and alcohol use disorder as the third most common disorder.
Bipolar II disorder may be the common link in comorbidity between social phobia and alcohol abuse, as demonstrated in epidemiological and clinical studies (39). Of 153 outpatients with social phobia studied retrospectively at the University of Pisa in Italy, 34 (22.2%) had a lifetime history of alcohol abuse of at least 1 year. Lifetime comorbidity of bipolar II disorder was 3.4% in patients with social phobia alone and 29.4% in patients with social phobia and comorbid alcohol abuse, suggesting that the link between social phobia and alcohol abuse may be their overlap with bipolar disorder.
| OTHER COMORBID AXIS I DISORDERS |
|---|
|
|
|---|
Attention-Deficit/Hyperactivity Disorder
Childhood bipolar disorder is a chronic, rapid-cycling, mixed-manic state often comorbid with ADHD and conduct disorder (42), but epidemiological data are limited. Features of bipolar disorder that overlap with those of ADHD include distractibility, inattention, impulsivity, and hyperactivity, and those overlapping with conduct disorder include impulsivity, substance abuse, aggressiveness, and problems with the law.
These overlapping symptoms may cause overestimates of comorbidity (42). Misdiagnosing bipolar disorder as ADHD can result in treatment with psychostimulants, which may induce mania or rapid cycling in a truly bipolar patient (43).
In 104 consecutive children referred to a community mental health clinic for the treatment of ADHD, 62 children (59.6%) had a mood disorder (44). The frequent onset of ADHD before mania and of oppositional defiant disorder/conduct disorder after mania further reinforces the need to examine for manic symptoms in children diagnosed with either of these comorbid conditions (45).
However, not all children meeting criteria for mania and ADHD have long-term persistence of manic symptoms, which casts some doubt on a link between manic symptoms associated with ADHD in childhood and later bipolar disorder (46). Conversely, children meeting criteria for ADHD but not mania may be diagnosed with bipolar disorder at long-term follow-up (46).
Axis II Personality Disorders
Axis II personality disorders may complicate the diagnosis and course of bipolar disorder. In a study of 52 remitted DSM-III-R bipolar patients, comorbid personality disorder occurred in 28.8%. Dramatic/emotionally erratic and fearful/avoidant personality disorders were more common than odd/eccentric disorders. Bipolar patients with comorbid personality disorders had greater severity of residual mood symptoms than those without comorbid personality disorders, even during bipolar remission (47).
In a sample of 117 patients with unipolar and 60 with bipolar affective disorders, 38% of the bipolar group met criteria for a comorbid personality disorder, and narcissistic personality disorder was significantly more common in bipolar than in unipolar patients (48).
A retrospective chart review of 52 euthymic male bipolar I outpatients revealed that 38% met criteria for an axis II diagnosis. The presence of a personality disorder predicted a more difficult course, with lower rate of current employment, use of more psychiatric medications, and more frequent history of alcohol and substance use disorders (49).
In a long-term study using a "life-charting" approach in 87 patients with bipolar disorder, patients with fewer personality disorder symptoms in 7 of 10 disorder categories had better outcomes, and Cluster A personality disorder symptoms best distinguished euthymic and symptomatic patients (50).
The mood cycling in bipolar II disorder may be difficult to distinguish from cyclothymic temperamental and borderline personality disorders (51). Furthermore, mood lability and interpersonal sensitivity traits appear to be related by a cyclothymic temperamental diathesis underlying the complex pattern of anxiety, mood, and impulsive disorders characterizing atypical depression, bipolar II disorder, and borderline personality disorder (52). Of 194 patients with bipolar II enrolled in a French multicenter EPIDEP study, 74 had cyclothymic temperament. Compared with those without cyclothymic temperament, they had younger age at onset and at seeking medical attention, higher scores on the HAM-D and Rosenthal atypical depressive scale, longer delay between onset and bipolar diagnosis, higher rate of psychiatric comorbidity, higher scores on irritable-risk-taking, and different profiles on Axis II, with more histrionic and passiveaggressive and fewer obsessivecompulsive personality disorders (53). These findings support the recent international consensus favoring the diagnosis of cyclothymic and bipolar II disorders over erratic and borderline personality disorders in patients meeting criteria for both sets of disorders.
The overlap between bipolar disorder and borderline personality may have therapeutic implications. In a retrospective study of 35 bipolar patients, 40% met criteria for borderline personality disorder. This subgroup had a more frequent history of substance abuse and childhood symptoms of ADHD. Borderline symptoms in both groups improved significantly with lamotrigine treatment, corresponding with improvement in bipolar symptoms (54).
Similarly, results of a double-blind placebo-controlled pilot study suggest that divalproate may be safe and effective in women with borderline personality disorder and comorbid bipolar II disorder, decreasing irritability, anger, volatility in relationships, and impulsive aggression (55). Given the overlap in symptoms, it is best not to diagnose comorbid personality disorders when patients are not in remission.
| MEDICAL COMORBIDITIES |
|---|
|
|
|---|
Many neurological disorders can cause "secondary mania," including strokes, tumors, head trauma, CNS infection, and degenerative disorders (56,57). DSM-IV defines these as "mood disorder due to a general medical condition, with manic features" (58). In these cases, the psychiatric disorder is not, strictly speaking, comorbid with the medical disorder, but is an expression of it. Nonetheless, making a proper diagnosis of the underlying medical disorder is crucial to effective treatment. Often it is unclear whether a medical disorder is truly comorbid, a consequence of treatment, or a combination of both, as is the case for obesity, diabetes, and hypothyroidism (59,60). Optimal management of bipolar disorder demands awareness of comorbidity and its complications.
Medical Disorders Comorbid With Bipolar Disorder
Migraine
Migraine headache is a common, debilitating disease, affecting approximately 11% of the US population (61). However, migraine tends to be underdiagnosed and undertreated in patients with bipolar disorder (62). In a representative US sample, the adjusted lifetime prevalence of migraine headache was 15.2% among the 3.7% of the sample that screened positive for bipolar disorder, compared with 7.0% for those who screened negative (63).
Migraine prevalence is higher in bipolar II disorder. In a Norwegian study of 27 patients with bipolar disorder (11), 77% of bipolar II subjects (n = 13) but only 14% of bipolar I subjects (n = 2) suffered from migraine (p = .001). In 108 outpatients with bipolar disorder, the overall lifetime prevalence of migraine was 39.8% (43.8% for women and 31.4% for men), but it was 64.7% in the bipolar II subgroup. Compared with patients without migraine, those with migraine were younger, more educated, more likely to be employed, had fewer psychiatric hospitalizations, and initially presented for symptoms of depression rather than hypomania. These differences in course suggest that bipolar disorder with migraine may represent a subtype of bipolar disorder (62).
Velocardiofacial Syndrome
VCFS is a genetic disorder, caused by a microdeletion on chromosome 22q11, with an incidence of approximately 1 in 3,000 people (64). It is characterized by multiple somatic anomalies (64). Psychiatric diagnoses that have been found in excess of population norms in VCFS patients include schizophrenia, bipolar disorder, and ADHD (65,66). There may be an age-related progression, with ADHD predominating in school-age children, bipolar spectrum disorders in adolescents, and psychotic illness (bipolar I or schizophrenia) in adults (64).
Multiple Sclerosis
MS is the neurological disorder most consistently identified as comorbid with bipolar disorder. Several epidemiological studies from the United States and Canada have documented this comorbidity at twice the expected rate based on the prevalence of each disorder in the population (67,68), while clinical studies have identified bipolar disorder in more than 10% of MS patients (69,70).
Bipolar symptoms may precede other neurological signs of MS (71) or may occur in tandem with relapse or exacerbation (7274). Although the interaction between the psychology and biology of MS is complex, most researchers hypothesize that the comorbidity of bipolar disorder is related to the location of the MS lesions (73,74).
Cushings Syndrome
Cushings syndrome is caused by the prolonged hypersecretion of cortisol. Patients may present with a clinically significant mood disturbance (75): typically depression (7578), but sometimes mania or hypomania (75,78,79).
Vascular Bipolar Disorder
Vascular mania, or vascular bipolar disorder, refers to subcortical infarcts presenting with manic symptoms (56). Patients are typically older, with late-onset bipolar disorder, cognitive impairment, and extensive subcortical white matter hyperintense lesions on brain magnetic resonance images (80,81). Younger bipolar patients may have similar hyperintensities, but their pathophysiologic significance is still unknown (80).
Cerebral Autosomal Dominant Arteriopathy With Subcortical Infarcts and Leukoencephalopathy
CADASIL, caused by mutations of the Notch 3 gene, typically presents with lacunar infarcts in the absence of classic cerebrovascular risk factors (82). Subcortical dementia and focal neurologic deficits may ensue, as well as migraine-like headache and psychiatric disturbances including manic or mixed affective symptoms (82).
Brain Tumor
A few patients with brain tumors have had documented bipolar symptoms, and some have initially been diagnosed with a bipolar disorder (57,83). Tumor location is presumed to be most important in causing the psychiatric symptoms, although tumor location and type have been variable (57).
Head Trauma
Manic episodes may follow nonpenetrating traumatic head injury (84,85), but depressive disorders are more common (85). In a 1-year, prospective study of 66 consecutive patients suffering from a closed-head injury, 6 patients developed mania. This incidence (9%) is higher than reported for either stroke or CNS tumors (85).
Dariers Disease
Dariers disease also involves a single genetic locus, in this case the ATP2A2 gene coding for the SERCA 2 protein, which plays a key role in intracellular calcium signaling. Vesicular lesions of the skin are the hallmark of this rare, autosomal dominant disorder. CNS lesions are also common, leading to mental retardation, epilepsy, and psychiatric symptoms (86), especially bipolar (87) and other mood disorders.
Asthma
In a representative sample of the general population in Dresden, Germany, aged 18 to 65 years, lifetime severe asthma was associated with a fivefold increase in the probability of bipolar disorder (odds ratio, 5.64; 95% confidence interval, 1.9516.35) (88).
| MEDICAL DISORDERS RELATED TO PHARMACOLOGIC TREATMENT OF BIPOLAR DISORDER |
|---|
|
|
|---|
Of 644 outpatients with DSM-IV bipolar disorder in the Stanley Foundation Bipolar Treatment Outcomes Network (59), 58% were overweight, 21% were obese (36.2% of those who were overweight), and 5% were extremely obese (8.6% of those who were overweight). However, data from NHANES III (19881994) showed that 55% of Americans were overweight and 27% were obese (59). Thus, these data do not constitute evidence for epidemiologic comorbidity between obesity and bipolar disorder.
A critical issue is the role of treatment in promoting weight gain. During acute treatment of bipolar disorder, 14 subjects (28%) gained at least 5% of their baseline BMI, and during maintenance treatment, 13 subjects (26%) gained more than 5% of their BMI (89). Higher scores on the Hamilton Rating Scale for Depression and negative scores on the Bech-Rafaelsen Mania Scale predicted an increase of BMI during acute treatment. Weight at treatment initiation was inversely related to weight gain during treatment, and the obese group had no significant weight gain during the acute phase.
Lithium has been associated with weight gain (91), which is a major factor leading to poor compliance with treatment (92). Anticonvulsants, especially valproate, may also increase risk of weight gain. Over 12 months, 71% of patients taking valproate and 43% of those taking carbamazepine gained weight (93). Weight gain is an adverse effect of most novel antipsychotics, most prominently olanzapine and clozapine (94), and more moderately for risperidone and quetiapine (95). Risk of weight gain is increased for antipsychotic-naïve patients and may not be dose dependent (96).
Diabetes Mellitus (Type 2)
The prevalence of diabetes mellitus was significantly higher in 345 hospitalized bipolar patients with manic or mixed subtype (9.9%) than in the overall population (3.4%) (60). Although the possible mechanisms of this observed comorbidity are unclear, possibilities include a genetic relationship, a causal relationship (in which hypercortisolemia induces diabetes, or diabetic vascular lesions contribute to mania), or the effect of psychotropic medications and associated weight gain. Complications of treatment with some atypical antipsychotics may include sudden, severe, and occasionally fatal diabetes ketoacidosis, as well as a marked increase of serum lipids, especially triglycerides (94).
Hypothyroidism
There is little evidence linking bipolar disorder and hypothyroidism. In the only study evaluating baseline thyroid indices and TSH response to TRH infusion in bipolar patients who had never been exposed to lithium or carbamazepine, only 5 of 54 patients (9.2%) had any evidence of hypothyroidism, and T4 was low in only 1 of these patients (97). However, hypothyroidism could be an important consequence of treatment with lithium, or it could be related to autoimmune thyroiditis (98). Prospective studies may help determine whether pharmacological thyroid supplementation facilitates recovery from bipolar depression (99).
Polycystic Ovarian Syndrome
Abnormalities of the hypothalamicpituitarygonadal axis can occur in bipolar disorder, especially high rates of menstrual disturbances (100), which in one study occurred in conjunction with obesity, independent of therapeutic agent used (101). However, menstrual abnormalities, hyperandrogenism, and polycystic ovarian syndrome (PCOS) in bipolar disorder have been linked to valproate use (102). Of 140 women with bipolar disorder, nearly half of those taking valproate had menstrual abnormalities and abnormally high androgen levels, and 41% had PCOS. Only 13% of those not receiving valproate reported menstrual problems (102).
Renal Failure
Lithium-induced nephrogenic diabetes insipidus is common, and there are some reports of renal failure with lithium. In one study of 24 bipolar patients who had been taking lithium for a mean of 13.6 years (range 225) and who presented with elevated serum creatinine (mean 2.8 mg/dl; range, 1.38.0), all had lithium-related renal damage confirmed by biopsy, 87% had nephrogenic diabetes insipidus, and 33.3% had hypertension (98). Despite discontinuation of lithium, 7 of 9 patients with serum creatinine greater than 2.5 mg/dl had subsequent progression to end-stage renal disease (98).
Given the association with cardiovascular risk factors, it is not surprising that cardiovascular disease is a leading cause of early death in bipolar disease, second only to suicide.
Skin Rash
Skin rash, and rarely Stevens-Johnson syndrome, is associated with some of the medications used in the management of bipolar disorder, including carbamazepine and lamotrigine.
| TREATING COMORBID CONDITIONS |
|---|
|
|
|---|
Evidence of the efficacy of atypical antipsychotics in mania has led to their widespread use (104,105). However, when weight gain is an issue, it is important to consider gradually decreasing the dose and switching to another atypical antipsychotic, such as ziprasidone or aripiprazole, which are not associated with weight gain (106,107), or discontinuing the atypical antipsychotic. Regulatory agencies in some countries have suggested that avoiding the use of certain atypicals as first-line treatment in patients with a personal or family history of obesity, diabetes, and hyperlipidemias can reduce the risk of metabolic disorders (94).
Topiramate may reduce psychotropic drug-induced weight gain and binge eating in euthymic patients without affecting mood (108). By inducing weight loss, it may offset the weight gain associated with some psychotropic drugs and with bipolar disorder. However, its efficacy in the management of bipolar disorder is unproven, and it may produce adverse effects of cognitive dulling, depression, and paresthesia that limit its utility.
Used alone or as adjunctive therapy, carbamazepine or divalproex may be more effective than lithium for mixed, dysphoric, and rapid-cycling mania, which often accompanies substance abuse (109). Divalproex may also reduce drug cravings and promote abstinence (18). Valproate decreased panic symptoms in 1 of 13 mood-unstable patients with comorbid panic attacks (55).
In patients with rapid cycling accelerated by substance abuse or anxiety disorder, there is a clear need for pharmacotherapy that is effective as an acute antidepressant without inducing switching or cycle acceleration. Data suggests that valproate and lamotrigine may be of benefit in rapid-cycling bipolar disorder (110). A case report suggests that lamotrigine is safe and effective in rapid-cycling refractory bipolar disorder with comorbid alcohol and polysubstance abuse (111). Lamotrigine does not promote weight gain (112), but caution should be exercised in combining lamotrigine with valproate in view of increased serum levels of lamotrigine and the potential for Stevens-Johnson syndrome. Gabapentin appears to be ineffective in the treatment of bipolar disorder, although it may be useful in bipolar patients with social phobia or other anxiety disorders (113).
| CONCLUSION |
|---|
|
|
|---|
The author acknowledges Karen Oberheim for help in the preparation of this article.
| NOTES |
|---|
|
|
|---|
Supported by grants from National Institutes of Mental Health (MH60451) and GlaxoSmithKline; an additional grant have been received from Novartis. Dr. Krishnan is also a consultant to Abbott, Amgen, GlaxoSmithKline, Johnson & Johnson, Merck, NPS, Organon, Otsuka, Pfizer, Somerset, Synaptic, Vela, and Wyeth.
DOI:10.1097/01.psy.0000151489.36347.18
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. Harrison Doctors' health and fitness to practise: the need for a bespoke model of assessment Occup. Med., August 1, 2008; 58(5): 323 - 327. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. J. Kupfer The Increasing Medical Burden in Bipolar Disorder JAMA, May 25, 2005; 293(20): 2528 - 2530. [Full Text] [PDF] |
||||
![]() |
Reviews of Note Journal Watch Psychiatry, April 6, 2005; 2005(406): 6 - 6. [Full Text] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |