The World Journal of Biological Psychiatry, 2010; 11: 656–666
ORIGINAL INVESTIGATION
Bipolar disorder and comorbid attention deficit hyperactivity disorder. A distinct clinical phenotype? Clinical characteristics and temperamental traits1
SILVIA BERNARDI1,2, SAMUELE CORTESE3, MARY SOLANTO1, ERIC HOLLANDER4 & STEFANO PALLANTI1,2 1Department
of Psychiatry, Mount Sinai School of Medicine, New York, USA, 2Department of Psychiatry, University of Florence, Italy, 3Child and Adolescent Psychopathology Unit, Robert Debré Hospital, Paris, France, and 4Department of Psychiatry, Montefiore Medical Center University Hospital for Albert Einstein College of Medicine, New York, USA
Abstract Objectives. It has been suggested that bipolar disorder (BD) with comorbid ADHD represents a distinct clinical phenotype of BD. There are no data regarding potential heterogeneity between BD subjects with a diagnosis of ADHD in childhood whose ADHD remitted in adulthood (cADHD-BD) vs. BD patients with persistent ADHD diagnosis in adulthood (aADHD-BD). This heterogeneity may constitute a confounder in investigations of the nature of the co-occurrence between BD and ADHD. The aim of this paper is to compare BD patients without ADHD, to those with aADHD-BD, and those with cADHD-BD on clinical and temperamental characteristics, hypothesizing that maladaptive temperament will be increased in BD subjects with a stable diagnosis of ADHD in adulthood compared to those whose ADHD remitted. We further hypothesize that maladaptive temperament will be associated with the severity of both illnesses. Methods. A total of 100 outpatients (aged 18–30 years) with BD in remission were included. The assessment of ADHD was made according to a procedure aimed to reduce potential recall biases. Subjects had to have a parent available and had never been treated with stimulants. Temperamental traits were assessed with the California Child Q-sort (CCQ) and the Early Adolescent Temperament Questionnaire (EATQ). Results. Rate of cooccurrence of ADHD-BD was 18% lifetime and 10% current diagnosis. Patients with ADHD-BD (aADHD-BDcADHD-BD) reported a significantly earlier onset of mood disorder, higher number of previous mood episodes, and significantly higher impulsivity than BD patients without ADHD. aADHD-BD showed a significantly earlier BD onset, higher number of previous mood episodes, higher impulsivity, decreased Reactive Control and higher Negative Emotionality temperamental scores than cADHD patients. Conclusion. Findings suggest that patients with aADHD-BD present a clinical phenotype distinct from that of patients with BD without ADHD or with a childhood ADHD diagnosis that remitted with the age. This appealing hypothesis of a BD-distinct phenotype that can be detected early due to its associated maladaptive temperamental traits requires further investigation in larger samples, supported by neuropsychological, genetic and imaging data.
Key words: ADHD, bipolar disorder, impulsivity, temperament, adult Abbreviations: ADHD-BD, bipolar subjects with co-diagnosis ADHD lifetime; aADHD-BD, bipolar subjects with persistent ADHD diagnosis during adulthood; cADHD-BD, bipolar subjects with full ADHD diagnosis during childhood and symptom remission during adulthood.
Introduction Attention-deficit/hyperactivity disorder (ADHD) is one of the most common childhood-onset psychiatric disorders. It is characterized by pervasive,
impairing, and developmentally inappropriate symptoms of inattention and/or impulsivity-hyperactivity. ADHD symptoms may persist into adulthood in up to 60% of cases (Wood et al. 1976;
1A
portion of the data has been previously presented by the authors in a lecture entitled “ADHD in Bipolar Adult Sample” at the XII meeting of the Italian Society of Psychopathology (SOPSI), Rome, Italy, February 19–23, 2008 Correspondence: Silvia Bernardi, MD, Mount Sinai School of Medicine, Department of Psychiatry, One Gustave L. Levy Place, Box 1230, New York, NY 10029, USA. Tel: 1 212 241 5288. E-mail:
[email protected] (Received 19 July 2009; accepted 26 January 2010) ISSN 1562-2975 print/ISSN 1814-1412 online © 2010 Informa UK Ltd. (Informa Healthcare, Taylor & Francis AS) DOI: 10.3109/15622971003653238
Temperament in adult bipolar disorder and comorbid ADHD 657 Kessler et al. 2006). ADHD cases that persist into adulthood have generally more severe symptoms (Young and Gudjonsson 2008). Bipolar disorder (BD) with comorbid ADHD has been described as a serious condition: patients with both these disorders are reported to exhibit an approximately 5 years earlier onset of BD, more depressive and mixed episodes and a generally worse course of illness and treatment response (Sachs et al. 2000; State et al. 2004; Nierenberg et al. 2005). Based on the increment of the clinical severity, some authors have suggested that ADHD-BD comorbidity may represent a distinct clinical phenotype of BD (Faraone et al. 2001; Masi et al. 2006). This hypothesis has been supported by recent familial studies (Doyle and Faraone 2002). Furthermore, there is increasing evidence of neuroanatomical differences between ADHD, BD, and ADHD BD (Biederman et al. 2008; Monuteaux et al. 2008). However, several inconsistencies about the comorbidity between BD and ADHD still need to be addressed. For instance, the prevalence of ADHDBD comorbidity in adults has been reported to vary between 27% (Tamam et al. 2008) and 9.5% (Nierenberg et al. 2005). This discrepancy may be due to methodological artefacts, related to the retrospective nature of the ADHD diagnosis (Miller et al. 2009), or to the similarity between ADHD and BD symptoms that has been shown to often lead to mistakenly assumed ADHD symptoms as part of BD (Klassen et al. 2009). Controversially, the rate of ADHD comorbidity has been reported to decrease as the population ages (Biederman et al. 2000). It is unclear whether the age-dependent symptom decline constitutes remission or a methodological artefact, given that ADHD symptoms are reported to change with age. Specifically, when compared to children with ADHD, adults with ADHD present with less externalizing symptoms and with a higher rate of psychiatric comorbidities, including major depressive disorder, anxiety symptoms and substance abuse (Karam et al. 2009). Mostly on the base of the decrease of the ADHD-BD co-occurrence with age, some authors have doubted the nature of ADHD-BD comorbidity and hypothesized that, due to symptom overlap, it may represent a “phenocopy“, instead of a phenotype (Geller et al. 1998). Children with ADHD-BD whose symptoms do not meet the full ADHD criteria in adulthood (cADHD-BD) may constitute a different population from children with ADHD-BD whose ADHD symptoms persist in adulthood (aADHD-BD). There are currently no data regarding potential differences. The lack of attention to the possible heterogeneity of patients with ADHD-BD co-occurrence may constitute a confounder of investigations of the ADHD-
BD phenotype, potentially leading to inconsistencies and consequent skepticism regarding the nature of this comorbidity. The present article evaluates the rate of cooccurrence and clinical characteristics of comorbid ADHD and BD compared to patients with BD without ADHD, with cADHD, and with aADHD. The impact of ADHD comorbidity on BD severity is also investigated. The analysis of temperamental traits may provide further understanding regarding the etiology of aADHD-BD and cADHD-BD.Temperamental traits are presumed to have a genetic and familial basis and manifest themselves from childhood (Rothbart and Bates 2006). They have been reported to differ between ADHD children and normal controls (Martel and Nigg 2006). Interestingly, Martel et al. (2009a,b) reported that personality/temperamental traits in children are associated with different ADHD symptom dimensions. Cognitive/Effortful Control, defined as the relatively deliberate modulation of emotional states, behaviours and focusing and shifting attention (Eisenberg et al. 1996), is associated with inattentive symptoms. Neuroticism/Negative Emotionality, defined as the tendency to respond intensely to negative emotions (Eisenberg et al. 1996), is instead associated with hyperactivity/impulsivity symptoms. Reactive Control, which is defined as the relatively automatic modulation of emotion and behaviour in response to an immediate incentive (Eisenberg et al. 2000), is associated with impulsive/hyperactive symptoms. Given the relationship between temperamental traits and ADHD, the current study attempts to investigate how patients with aADHD-BD differ from patients with cADHD-BD. In relation to this aim, we hypothesized that the impulsivity and mood instability typical of ADHD and BD symptoms have an additive effect on the severity of BD and map within the Reactive Control temperamental domain. Thus, we expected that maladaptive temperament (high negative emotionality, low resiliency, diminished effortful control and, in particular, low reactive control) would be increased in BD subjects with a stable diagnosis of ADHD in adulthood, and would also be associated with the severity of the illnesses.
Method Subjects The sample included 100 consecutive adult selfreferred out-patients at the Institute of Neuroscience, Florence, Italy during a 4-year period (January 2004–January 2008). The final sample was constituted by 59 males and 41 females and a diagnosis of BD Inclusion criteria were: (1) an age range of at least
658 S. Bernardi et al. 18 and no greater than 30 years to avoid a retrospective ADHD diagnosis time that was potentially too long; (2) remission of BD for at least 3 months, as assessed by Young Mania Rating Scale (12) (Young et al. 1978) and Hamilton Depression Scale (7) scores (HAMILTON 1960), in order to reduce possible diagnostic confounders of symptom overlap between ADHD and acute mania; and (3) the presence of at least one parent able to describe the patient’s lifetime course of disease symptoms, in order to confirm th e age of onset of ADHD. The presence of at least one parent was considered necessary also to report about the onset and the course of BD symptoms and to distinguish them from symptoms of oppositional defiant disorder or conduct disorder. Exclusion criteria were: (1) any clinically significant medical conditions, organic brain disorders, (2) current substance/alcohol abuse or dependence (in the last 6 months), since it may precipitate BD episodes and impair ADHD symptoms at present increasing their intensity; and (3) any lifetime comorbid mental disorders (except previous secondary symptoms of anxiety or substance abuse as based on temporal onset and symptoms severity in remission from at least 6 months), unwillingness or inability to comply with study assessments, or inability to provide informed consent. A total of 159 patients with a DSM-IV-TR BD diagnosis, with not known history of psychotic disorders and major medical conditions, in the age-range of 18–30 years that consecutively entered the Institute for treatment and/or assessment were screened. Of the subjects initially screened 47 have been excluded because of the absence of a relative able to provide information about their childhood, 50 were excluded because of current substance/alcohol abuse and 34 were excluded because of the presence of psychiatric comorbidities other than ADHD. After a complete description of the study, all patients provided written informed consent before participating in the study. Local institutional review boards approved the study. Procedure Diagnosis of BD was based on a clinical psychiatric interview and was further confirmed using the affective disorder module of the Structured Clinical Interview for DSM-IV (SCID-I) (First et al. 1997). In order to assess ADHD comorbidity and to assess other psychiatric comorbidities, certified psychiatrists conducted detailed diagnostic interviews with the subjects and interviews with their relatives that lasted approximately 2.5 h. ADHD diagnosis was based on DSMIV-TR criteria, and supported by the Wender Utah Rating Scale (WURS) (Fossati et al. 2001). Patient age
at the time of onset was defined as the first emergence of clinically impairing symptoms as assessed by the WURS. The first assessment included a careful review of the medical charts and direct communication with the previous treating psychiatrist, if needed. The WURS assessed ADHD severity and classified the participants in the following groups: (1) patients without any lifetime ADHD diagnosis (BD, 6 previous hyperactive symptoms, 6 previous inattentive symptoms, 6 current hyperactive symptoms, and 6 current inattentive symptoms); (2) patients who received a diagnosis of childhood ADHD but did not meet the criteria for a diagnosis of adult ADHD (cADHD-BD; at least six previous hyperactive symptoms and/or at least six previous inattentive symptoms but 6 current hyperactive symptoms and 6 current inattentive symptoms); and (3) patients with a persistent diagnosis of comorbid ADHD (aADHD-BD: at least six previous hyperactive symptoms and/or at least six previous inattentive symptoms, along with at least six current hyperactive symptoms and/or at least six current inattentive symptoms). The absence of ADHD symptoms has been evaluated retrospectively with the same questionnaire within the 6 months previous the interview. The WURS assessment of childhood ADHD symptoms were conducted with the presence of at least a parent. Only medical chart bearing previous diagnoses of hyperactivity/impulsivity and associated dysfunctional impact were considered. Previous treating psychiatrists were consulted for four cases, all of them with a previous diagnosis of ADHD confirmed at the enrollment time by the WURS score. In the case of disagreement regarding the ADHD diagnosis, an additional board certified psychiatrist blinded to the case condition was consulted. All participants were followed for at least 6 months at intervals of at least 15–30 days at the Institute of Neuroscience clinic in order to control for the onset of a bipolar episode that may constitute a confounding factor with ADHD symptoms. Temperament trait measures were assessed for cADHD-BD and aADHD-BD.The California Q-sort (CCQ) Adult Revised Edition was administered (by staff persons) to subjects and parents (Block and Block 1980). In case of lack of accordance between the raters, a senior psychiatrist during the clinical interview elaborated the item. The CCQ consists of 100 descriptive statements of behavioural, affective, and cognitive characteristics of the subject. The temperamental dimensions of Resiliency, Reactive Control and Negative Emotionality were computed by previously published CCQ items (Eisenberg et al. 1996). To measure Effortful Control, parents and subjects completed a short form of the Early Adolescent Temperament Questionnaire (EATQ), whose
Temperament in adult bipolar disorder and comorbid ADHD 659 items had been adapted for adulthood (Capaldi and Rothbart 1992). Subjects and relatives rated how well each statement described the subject currently on a scale from 1 (almost always untrue of the subject) to 5 (almost always true of the subject), thereby yielding 10 scales. As a secondary clinical measure, the Barratt Impulsiveness Scale (BIS-11) (Fossati et al. 2001) was included. The BIS-11 is a 30-item self rating scale that assesses the severity of impulsivity according to three factors: behavioural impulsivity, cognitive impulsivity, and non-planning impulsivity.
Statistical analyses Statistical analyses were performed using SPSS 17.0 software (SPSS Inc., Chicago, IL). First, we compared demographic and clinical characteristics between patients with ADHD-BD (cADHD-BD and aADHD-BD) and those with BD without any history of ADHD (BD). Chi-squared cross-tabulation analyses and one-way analyses of variance were employed. Given the differences in the size of the samples, the tests were corrected with the Brown-Forsythe and Welch robust F-test to adjust for differences in the variances. Correlations were assessed using Pearson correlation coefficients. Predictors of the age of onset of mood symptoms were assessed using a multiple regression with age at the BD onset as dependent factor, WURS as independent and age, gender, education, BIS score, the number of previous episodes, and previous history of substance abuse as predictors entered into the model using a backward stepwise elimination. Similarly, predictors of WURS score were tested using multiple regression with WURS score as dependent factor, the BIS total score as independent, age, gender, education, number of previous mood episodes, the BIS subscales’ score as predictors entered with backward stepwise method. All tests were twotailed and assumed a 5% significance level. Given the non-normality of the cADHD-BD subject data, even after logarithmic transformation, further comparisons between the means of the patients with a persistent ADHD diagnosis (aADHDBD) and patients with a previous ADHD diagnosis (cADHD-BD) were carried out using non-parametric rank sum tests (Mann–Whitney U-test, as alternative to the two independent sample t-test) in which a significance level of P0.05 was applied. Relationships among temperamental and demographic/ clinical variables were tested by non-parametric Spearman’s rank correlation tests. We controlled for the effects of gender, age, age at BD onset, age at ADHD onset, and the number of previous mood episodes on clinical and temperamental measures
by stratifying the results via the Mantel–Haenszel χ2-test. This test permitted a comparison within each stratum (by gender, age, age at BD onset, age at ADHD onset, and the number of previous mood episodes) of the number of individuals who scored differently at the BIS, the WURS and temperamental measure converted in categorical variables.
Results Forty-nine women and 51 men with a diagnosis of BD in remission comprised the sample. The mean age of the 100 patients was 28.53; SD2.3 years (range: 18–30 years). Sixty-five subjects had a family history of affective disorder. The mean age of onset for bipolar disorder was 17.25, SD3.76 years. The first affective episode in the majority of cases (76%) was a depressive episode. The predominant affective episode during the course of the illness was depressive in 75% (n75) of the subjects. Previous depressive episodes were more frequent than (hypo)-manic episodes (mean2.79; SD1.16 and mean1.58; SD0.57, respectively; t–8.997, df81, P0.000). Subjects with type one bipolar disorder constituted the majority of the sample (67.4% type I vs. 32.6% type II, valid percent). Differences in clinical and demographic data among the sample of patients with BD (n82) and patients with ADHD-BD (n18) are illustrated in Table I. At the time of assessment, 86 subjects (86%) were under pharmacological treatment. Differences in treatment conditions between BD subjects with and without ADHD comorbidity were not statistically significant (Table II). No patients were treated with stimulants. Eighty-four percent of the BD group and 94.4% of the BD-ADHD group were treated with mood stabilizers (P0.306) (Table II).
Characteristics of BD subjects compared to ADHD-BD subjects (cADHD-BD and aADHD-BD) The prevalence of ADHD comorbidity in our sample of patients with BD was 18% for the lifetime of the patient and 10% at the time of the current diagnosis. Patients exhibiting a diagnosis of ADHD included thirteen men and five women. The mean age at ADHD symptom onset was 6.50; SD1.04 years. Eight subjects (44.4%) reported late ADHD onset after the seventh year, and none reported an onset after the tenth year. Subjects with type I constituted 66.7% of the sample with co-morbid ADHD and 67.5% of the sample without ADHD co-morbidity. Difference in the ratio of type I and type II were not statistically
660 S. Bernardi et al. Table I. Clinical and demographic characteristics of patients with Bipolar Disorder in remission with and without a lifetime ADHD diagnosis (aADHD-BD and cADHD-BD). ADHD-BD Subject (aADHD-BD and cADHD-BD)
BD Subjects
Analysis
Variable
N
Mean
SD
95% Confidence Interval for Mean
Age (years) Education (years) BD onset (years) Previous mood episodes (N) Lifetime depressive episodes (N) Lifetime (Hypo) manic episodes (N) WURS CARS-M total score CARS-M Mania subscale CARS-M Psychosis subscale BIS-11 total score BIS attentional subscale BIS motor subscale BIS non planning
80 81 82 82
28.65 10.76 17.76 4.30
2.29 3.21 3.78 1.30
28.11–29.13 10.05–11.47 16.93–18.60 4.01–4.59
18 18 18 18
28.11 9.66 14.88 5.33
2.42 3.06 2.65 1.49
26.90–29.31 8.14–11.19 13.56–16.20 4.58–6.07
0.67 1.85 14.64 7.30
1 1 1 1
24.31 26.00 34.13 23.00
0.42 0.18 0.001 0.01
82
2.79
1.16
2.53–3.04
18
4.00
1.53
3.23–4.76
9.90
1
21.48
0.005
82
1.51
0.57
1.38–1.63
18
1.33
0.48
1.09–1.57
1.87
1
28.40
0.18
82 80 80
24.74 15.12 11.71
4.87 3.15 2.43
23.67–25.81 14.42–15.82 11.17–12.25
18 18 18
53.11 15.72 12.33
24.89 3.92 4.07
40.73–65.48 13.77–17.67 10.30–14.35
23.18 0.36 0.38
1 1 1
17.28 0.001 22.21 0.55 19.81 0.54
80
3.71
1.56
3.36–4.05
18
3.38
1.91
2.43–4.34
0.44
1
22.35
82 82
54.53 17.24
9.35 4.82
52.48–56.59 16.18–18.30
18 18
63.00 19.50
7.59 5.33
59.22–66.77 16.84–22.15
16.77 2.72
1
29.53 0.001 23.48 0.11
82 82
13.12 24.14
4.52 6.00
12.12–14.11 22.82–25.46
18 18
16.33 27.05
4.74 4.45
13.97–18.69 24.84–29.26
6.88 5.49
1 1
24.26 32.20
N
Mean
SD
95% Confidence BrownInterval for Forsythe Mean F df1
df2
P
0.51
0.01 0.02
BD, subjects with a bipolar disorder diagnosis; ADHD, attention deficit hyperactivity disorder; ADHD-BD, subjects with a current or previous lifetime ADHD diagnosis; aADHD-BD, bipolar subjects with a current diagnosis of ADHD; cADHD-BD, bipolar subjects with a diagnosis of ADHD during childhood CARS-M, Clinician-Administered Rating Scale for Mania; WURS, Wender Utah Rating Scale.
correlated with the WURS score (n98, Pearson coefficient–0.36, P0.001). The final linear regression model with BD onset as a dependent factor, WURS as an independent factor, and previous manic episodes and education as predictors reported a coefficient B of –0.087 (SE0.02, t– 3.89, P0.001, 95% CI–0.13 to –0.04). Similarly, the linear regression model with the WURS score as a dependent variable, previous depressive episodes as an independent variable, and
significant between groups with and without ADHD (χ20.005; P0.575). Patients with ADHD-BD reported a significantly earlier onset of mood disorder, a significantly higher number of previous mood episodes, and a particularly notable increase in the number of depressive episodes). Patients with ADHD-BD scored higher than BD patients on the BIS 11, in particular on the motor and non-planning impulsivity subscales. The age at the time of BD onset was significantly
Table II. Demographic characteristics and pharmacological treatment of patients with bipolar disorder in remission with and without a lifetime ADHD diagnosis (aADHD-BD and cADHD-BD): cross tabulation analyses.
BD Subjects Variable Gender (male/female) History of substance/alcohol abuse Psychotropic medication Mood stabilizer (lithium, valproate) Selective Serotonin Reuptake Inhibitor Benzodiazepines Atypical antipsychotics aContinuity bFisher's
correction. exact test.
ADHD-BD Subject (aADHD-BD and cADHD-BD)
Analysis
N
%
N
%
χ2
Correctiona
df
Exact Sig.b
38/44 17
46.3/53.7 20.7
13/5 7
72.2/27.8 38.9
3.956 2.668
2.988 1.765
1 1
0.680 0.129
69 4
84.1 4.9
17 2
94.4 11.1
1.300 0.986
0.585 0.200
1 1
0.306 0.588
17 9
21.8 11.3
3 2
17.6 11.1
1.44 0.000
0.003 0.000
1 1
0.761 1.000
Temperament in adult bipolar disorder and comorbid ADHD 661 Table III. Clinical and demographic characteristics of patients with bipolar disorder in remission with a previous ADHD diagnosis during childhood (cADHD-BD) and current ADHD diagnosis (aADHD-BD): Mann–Whitney U-test.
cADHD-BD
Analysis (Mann–Whitney U-test)
AADHD-BD
N
Sum of rank
Mean
SD
N
Sum of Rank
Mean
SD
U
Z
P∗
8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8
67.00 82.50 109.00 86.00 36.50 40.50 79.00 36.00 36.00 36.00 68.00 43.00 38.50 86.50 38.00 99.00 75.00 114.00 36.00
27.37 10.12 16.87 6.75 4.00 2.62 1.37 26.50 6.33 6.44 15.12 56.12 15.00 17.75 23.37 5.00 3.30 3.60 2.87
3.46 3.31 0.99 0.88 1.06 1.18 0.51 6.50 0.67 0.69 4.64 5.76 2.92 5.67 2.38 1.06 0.67 0.67 0.64
10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10
104.00 88.50 62.00 85.00 131.50 130.50 92.00 135.00 135.00 135.00 103.00 128.00 132.50 84.50 133.00 72.00 96.00 57.00 135.00
28.70 9.30 13.30 6.30 6.40 5.10 1.30 74.40 0.87 1.37 16.20 68.50 23.10 15.20 30.00 3.80 3.37 5.50 5.80
0.94 2.98 2.49 1.15 0.69 0.56 0.48 2.11 1.35 1.76 3.42 2.71 3.84 3.76 3.36 1.03 0.74 0.53 0.63
31.50 33.50 7.00 30.00 3.50 4.50 37.00 0.00 0.00 0.00 32.00 7.00 2.50 29.50 2.00 17.00 39.00 2.00 0.00
–0.842 –0.734 –2.974 –0.928 –3.465 –3.317 –0.326 –3.593 –3.749 –3.686 –0.714 –2.952 –3.353 –0.941 –3.405 –2.116 –0.102 –3.468 –3.670
0.426 0.578 0.002 0.372 0.001 0.001 1.000 0.001 0.001 0.001 0.498 0.002 0.001 0.371 0.001 0.041 1.000 0.001 0.001
Age (years) Education (years) BD onset (years) ADHD onset (years) Previous mood episodes (N) Lifetime depressive episodes (N) Lifetime (Hypo) manic episodes (N) WURS Hyperactivity/Impulsivity DSM items (N) Inattention DSM items (N) CARS-M total score BIS-11 total score BIS attentional subscale BIS motor subscale BIS non planning Resiliency Effortful control Reactive control Negative emotionality ∗Exact.
Sig two-tailed. BD, subjects with a bipolar disorder diagnosis; ADHD, attention deficit hyperactivity disorder; ADHD-BD, subjects with a current or previous lifetime ADHD diagnosis; aADHD-BD, bipolar subjects with a current diagnosis of ADHD; cADHD-BD, bipolar subjects with a diagnosis of ADHD during childhood CARS-M, Clinician-Administered Rating Scale for Mania; WURS, Wender Utah Rating Scale.
the BIS attentional subscale, BIS non-planning subscale, and previous manic episodes as predictors reported a coefficient B of 4.01 (SE1.01, t3.95, P0.001, 95%CI1.99–6.02).
subjects with cADHD-BD. Subjects with aADHDBD also showed a higher number of previous mood episodes compared to subjects with cADHD-BD, in particular for depressive episodes. Subjects with a type I bipolar diagnosis constituted 50% of the aADHDBD sample and 62.5% of the cADHD-BD sample. This difference was not significant (χ20.596; P0.664). Patients with aADHD-BD scored higher on the BIS 11, especially with respect to the attentional and non-planning impulsivity. Average resiliency scores were higher in the cADHD-BD group,
Characteristics of subjects with a current adult ADHD diagnosis (aADHD-BD) and subjects with a remitted childhood ADHD diagnosis (cADHD-BD) Table III shows that subjects with aADHD-BD demonstrated significantly earlier BD onset compared to
Table IV. Demographic characteristics and pharmacological treatment of patients with bipolar disorder in remission with a previous ADHD diagnosis during childhood (cADHD-BD) and current ADHD diagnosis (aADHD-BD). cADHD-BD
AADHD-BD
Analysis
Variable
N
%
N
%
χ2
Correctiona
df
Exact Sig.b
Gender (male/female) History of substance/alcohol abuse Psychotropic medication Mood stabilizer (lithium, valproate) Selective Serotonin Reuptake Inhibitor Benzodiazepines Atypical antipsychotics
6/2 3
75.0/25.0 37.5
7/3 4
70.0/30.0 40.0
0.055 0.012
0.000 1.000
1 1
1.000 0.914
7
87.5
10
100.0
1.324
0.013
1
0.906
1
12.7
1
10.0
0.028
0.000
1
1.000
2 0
25.0 0.0
1 1
10.0 20.0
0.720 1.800
0.047 0.345
1 1
0.832 0.557
aContinuity bPearson
correction. Chi square test.
662 S. Bernardi et al.
Figure 1. Differences in temperamental traits (resiliency, negative emotionality, reactive control) in a sample of subject with bipolar disorder in remission and persistent ADHD comorbidity (aADHD) versus a sample of subject with bipolar disorder in remission and remitted ADHD symptoms (cADHD).
although this difference did not reach statistical significance. Reactive Control scores were significantly higher in the cADHD-BD group, whereas Negative Emotionality scores were higher in the aADHD-BD group (Tables III and IV) (Figure 1). Mantel–Haenszel χ2 stratification allowed excluding the possibility of a gender effect. Age at the time of bipolar disorder onset had an effect on the severity
of the impulsivity (BIS score) (Mantel–Haenszel χ29.96, df1, P0.002), Reactive Control (Mantel–Haenszel χ27.94, df1, P0.005), Negative Affectivity (Mantel–Haenszel χ26.09, df1, P0.014), and ADHD severity (WURS score) (Mantel–Haenszel χ26.04, df1, P0.014) scores, as well as the number of previous episodes (Mantel– Haenszel χ24.58, df1, P0.032).
Temperament in adult bipolar disorder and comorbid ADHD 663 BD onset, a greater number of episodes, and higher impulsivity compared to subjects whose ADHD did not persist into adulthood. Patients with aADHDBD also demonstrated lower resiliency and reactive control scores, but higher negative affectivity. Again, ADHD severity was positively associated with impulsivity and a measure of negative affectivity, but was inversely associated with reactive control scores. As described, the prevalence rate of ADHD comorbidity in BD adults has been reported as 27% (Tamam et al. 2008) and 9.5% (Nierenberg et al. 2005). Our rate is closer to the rate reported by Nierenberg (2005). This discrepancy with Tamam (2008) may be explained by the lower age inclusion criterion (30 years). Our clinical description of ADHD-BD comorbidity confirms previous data (State et al. 2004; Nierenberg et al. 2005) and provides direct evidence for an association between ADHD severity and BD onset and severity (number of episodes). These findings indicate the existence of an association between ADHD symptom persistence and higher BD course severity, as well as earlier onset. Interestingly, eight subjects reported late ADHD onset after the seventh year, none reported an onset after the tenth year. This finding supports Faraone et al.’s results that suggested that late-onset adult ADHD is valid and that the DSM-IV age-at-onset criterion is too stringent (Faraone et al. 2006). This study avoided some of the common potential confounders; for example, stimulants have been reported to induce mania in 40% of BD cases
Impact of temperamental traits on affective and ADHD symptoms Spearman’s test confirmed a negative correlation between Negative Emotionality and BD onset (P 0.001), the number of previous episodes (P0.001), and depressive episodes (0.001). Moreover, Negative Emotionality was positively correlated with all ADHD severity measures (DSM items [P0.001], the WURS score [P0.009]), and the impulsivity measure BIS-11 (P0.001). In contrast, Reactive Control was negatively correlated with all ADHD severity measures, the WURS score [P0.001]), and the number of previous mood (especially depressive) episodes (Table V). As expected, Resiliency was negatively correlated with the DSM ADHD inattention items (P0.030). No correlations were found for Effortful Control.
Discussion In the present BD sample, the co-occurrence rate of ADHD comorbidity was 18% over the lifetime of the patient and 10% at the time of the current diagnosis. Patients with BD and comorbid ADHD (aADHDBDcADHD-BD) showed earlier BD onset, a greater number of episodes, and higher impulsivity. ADHD severity was positively correlated with the number of mood episodes and BD onset. Among subjects with BD and comorbid ADHD, patients with a persistent ADHD diagnosis exhibited earlier
Table V. Association of temperamental traits (resiliency, effort control, reactive control, and negative emotionality) with clinical ADHD and BD measures: Spearman's rho non-parametric correlation in ADHD bipolar patients (aADHD-BD and cADHD-BD). Resiliency
Age (years) Education (years) BD onset (years) ADHD onset (years) Previous mood episodes (N) Lifetime depressive episodes (N) Lifetime (Hypo) manic episodes (N) WURS Hyperactivity/Impulsivity DSM items (N) Inattention DSM items (N) CARS-M total score BIS-11 total score BIS attentional subscale BIS motor subscale BIS non planning ∗Correlation
Effort control
Reactive control
Negative emotionality
N
r
P
N
r
P
N
r
P
N
r
P
18 18 18 18 18 18 18 18 18
–0.427∗ –0.130 0.270 0.219 –0.393 –0.378 –0.065 –0.337 –0.517
0.037 0.546 0.279 0.283 0.057 0.061 0.771 0.079 0.011
18 18 18 18 18 18 18 18 18
0.310 0.101 0.046 –0.010 –0.092 –0.108 –0.050 –0.009 –0.089
0.150 0.654 0.823 0.963 0.671 0.610 0.829 0.964 0.675
18 18 18 18 18 18 18 18 18
–0.077 –0.043 0.486∗ 0.168 –0.566∗∗ –0.586∗∗ 0.129 –0.664∗∗ –0.447∗
0.708 0.841 0.012 0.410 0.006 0.004 0.562 0.001 0.027
18 18 18 18 18 18 18 18 18
–0.042 –0.224 –0.644∗∗ –0.025 0.723∗∗ 0.694∗∗ –0.042 0.497∗∗ 0.667∗∗
0.836 0.292 0.001 0.902 0.001 0.001 0.847 0.009 0.001
18 18 18 18 18 18
–0.434∗ 0.053 –0.236 –0.260 –0.039 –0.284
0.030 0.781 0.218 0.177 0.842 0.141
18 18 18 18 18 18
0.019 0.020 –0.284 –0.009 –0.224 –0.081
0.927 0.894 0.156 0.965 0.266 0.689
18 18 18 18 18 18
–0.596∗ –0.008 –0.348 –0.471∗ 0.122 –0.459∗
0.003 0.968 0.068 0.014 0.525 0.017
18 18 18 18 18 18
0.782∗∗ 0.254 0.531∗∗ 0.623∗∗ –0.144 0.673∗∗
0.001 0.179 0.005 0.001 0.451 0.001
is significant at the 0.05 level (two-tailed). is significant at the 0.01 level (two-tailed). BD, subjects with a bipolar disorder diagnosis; ADHD, attention deficit hyperactivity disorder; ADHD-BD, subjects with a current or previous lifetime ADHD diagnosis; aADHD-BD, bipolar subjects with a current diagnosis of ADHD; cADHD-BD, bipolar subjects with a diagnosis of ADHD during childhood CARS-M, Clinician-Administered Rating Scale for Mania; WURS, Wender Utah Rating Scale. ∗∗Correlation
664 S. Bernardi et al. (Wingo and Ghaemi 2008) and to exacerbate mania symptoms (Faedda et al. 2004). Stimulants have also been reported to be related to earlier onset of BD (DelBello et al. 2001). Our study comprised an Italian population for whom stimulants were not fully available in the Italian market at the time of recruitment and when they became to be on the market indications included only childhood population (18). Subjects with a persistent ADHD diagnosis in adulthood were characterized by maladaptive temperamental traits, such as high Negative Affectivity, low Resiliency, and diminished Reactive Control. Overall, the findings seem to suggest several differences in clinical severity and temperamental dimensions between aADHD-BD and cADHD-BD. Diminished Resiliency (defined as flexible modulation of control in response to contextual demands) and high Negative Emotionality scores have been previously shown to correlate with ADHD development (Martel et al. 2009a, 2009b). Low levels of Reactive Control have been hypothesized to be characteristic of several externalizing disorders (Eisenberg et al. 2001), and such levels have recently been shown to be the only temperamental traits responsible for mediating the risk for ADHD and subsequent development of substance abuse (Martel et al. 2009a, 2009b). It is noteworthy that, in our sample, both Reactive Control and Negative Emotionality were also associated with early onset of mood symptoms and strongly associated with ADHD severity. Reactive Control is thought to be associated with automatic control of emotion (Rothbart 1989), and has been hypothesized to reflect the “motivational aspect of impulsivity” (Kindlon et al. 1995). In support of this hypothesis, impulsivity as measured by the BIS was higher in the aADHD-BD group, which was characterized by the lowest reactive control scores. This could be explained as an additive effect stemming from the presence of both diseases. In fact, impulsivity has been reported as a core feature of BD that is especially relevant during inter-episode periods (Swann et al. 2001). Effortful Control has been defined as the ability to inhibit a dominant response in order to perform a subdominant response (Rothbart and Bates 2006). Its neural basis has been argued to be formed by the executive attention network, which is located within the frontal cortex, as well as the anterior cingulate cortex (ACC) (Rothbart et al. 2003). In our sample, as expected and consistent with previous reports, subjects with ADHD scored lower in Effortful Control than subjects with BD and without ADHD comorbidity (Martel and Nigg 2006). However, scores obtained for Effortful Control were not able
to differentiate between aADHD-BD and c-ADHDBD patients. This failure may be explained by a lack of statistical significance due to the small sample size or by the therapeutic action of mood stabilizers, which are reported to have therapeutic effects on the prefrontal cortex (Brambilla et al. 2005). Temperamental traits are thought to be based on and influenced by genetics (Rothbart and Bates 2006). However, the temperamental trait analysis presented in this paper does not definitively prove the existence of a familial nature for the ADHD-BD model. In fact, the relationship between temperament and psychopathology is not clear; temperament may predispose patients to psychopathology, interact with psychopathology, or reflect the same underlying biological process (Watson et al. 2006). Furthermore, correlation does not imply causation. The present study was a cross-sectional investigation. Longitudinal studies are needed to exclude the possibility of an influence of psychopathology on the temperament measurements and confirm the validity of temperament as a risk or protective factor.The relatively limited sample size for comparison among aADHD-BD and cADHD-BD constitutes a limitation to this study as well as the self-refereed nature of the sample. The diagnosis of ADHD, obtained through clinical interview, was based on the criteria of the DSM-IV that, however, are not validated for adult populations. However, all of the patients received a very rigorous and methodologically sound evaluation according to the state-of-the-art assessment measures of BD and ADHD for adults. Similarly, the strict selection of a population aged less than 30 years, with at least one parent available and without other psychiatric comorbities, included current substance abuse, limits the general validity of these results when applied to the population of subjects with BD. Conclusion In addition to presenting higher BD severity, earlier onset, and higher impulsivity, subjects with BD and a persistent ADHD diagnosis in adulthood are characterized by maladaptive temperamental traits like higher negative affectivity, low resiliency, and diminished reactive control. Taken together, these findings suggest that patients with aADHD-BD present a clinical phenotype distinct from that of patients with BD without ADHD symptoms or with a childhood ADHD diagnosis that remitted with the age. This appealing hypothesis of a BD-distinct phenotype that can be detected early due to its associated maladaptive temperamental traits requires further investigation of larger samples that are supported by neuropsychological, familial, and genetic and imaging
Temperament in adult bipolar disorder and comorbid ADHD 665 data. If the model of aADHD-BD as a distinct phenotype with particular temperamental features is confirmed by additional studies, it may provide useful insight into the clinical management of this particular subgroup of BD patients.
Acknowledgements None.
Statement of interest No financial support has been provided for this study. Drs Bernardi Hollander and Pallanti report no competing interests. Dr Solanto is a member of the Medical Advisory Board of Shire Pharmaceuticals. Dr Cortese is a consultant for Shire Pharmaceuticals, has received financial support to attend medical meetings from Eli Lilly and Co. and Shire Pharmaceuticals, and has been a co-investigator in studies sponsored by Glaxo SmithKline, Eli Lilly and Co., and Genopharm.
References Biederman J, Mick E, Faraone SV, Spencer T, Wilens TE, Wozniak J. 2000. Pediatric mania: a developmental subtype of bipolar disorder? Biol Psychiatry 48:458–466. Biederman J, Makris N, Valera EM, Monuteaux MC, Goldstein JM, Buka S, et al. 2008. Towards further understanding of the co-morbidity between attention deficit hyperactivity disorder and bipolar disorder: a MRI study of brain volumes. Psychol Med 38:1045–1056. Block J, Block J. 1980. The role of ego-control and ego-resiliency in the organization of behavior. In: Collins W, editor. Development of cognition, affect, and social relations: The Minnesota symposia on child psychology. Hillsdale, NJ: Lawrence Erlbaum. p. 39–101. Brambilla P, Stanley JA, Nicoletti MA, Sassi RB, Mallinger AG, et al. 2005. 1H magnetic resonance spectroscopy investigation of the dorsolateral prefrontal cortex in bipolar disorder patients. J Affect Disord 86:61–67. Capaldi D, Rothbart M. 1992. Development and validation of an early adolescent temperament measure. J Early Adolesc 12:153–173. DelBello MP, Soutullo CA, Hendricks W, Niemeier RT, McElroy SL, Strakowski SM. 2001. Prior stimulant treatment in adolescents with bipolar disorder: association with age at onset. Bipolar Disord 3:53–57. Doyle AE, Faraone SV. 2002. Familial links between attention deficit hyperactivity disorder, conduct disorder, and bipolar disorder. Curr Psychiatry Rep 4:146–152. Eisenberg N, Fabes R, Guthrie I, Murphy B, Maszk P, Holmgren R. 1996. The relations of regulation and emotionality to problem behavior in elementary school children. Dev Psychopathol 8:141–162. Eisenberg N, Fabes RA, Guthrie IK, Reiser M. 2000. Dispositional emotionality and regulation: their role in predicting quality of social functioning. J Pers Soc Psychol 78:136–157. Eisenberg N, Cumberland A, Spinrad TL, Fabes RA, Shepard SA, Reiser M, et al. 2001. The relations of regulation and emotion-
ality to children’s externalizing and internalizing problem behavior. Child Dev 72:1112–1134. Faedda GL, Baldessarini RJ, Glovinsky IP, Austin NB. 2004. Treatment-emergent mania in pediatric bipolar disorder: a retrospective case review. J Affect Disord 82:149–158. Faraone SV, Biederman J, Monuteaux MC. 2001. Attention deficit hyperactivity disorder with bipolar disorder in girls: further evidence for a familial subtype? J Affect Disord 64:19–26. Faraone SV, Biederman J, Spencer T, Mick E, Murray K, Petty C, et al. 2006. Diagnosing adult attention deficit hyperactivity disorder: are late onset and subthreshold diagnoses valid? Am J Psychiatry 163:1720–1729. First M, Spitzer R, Gibbon M, Williams JBW. 1997. Structured Clinical Interview for DSM-IV Axis I (SCID I), Clinical Version. Washington, DC: American Psychiatric Press. Fossati A, Di CA, Acquarini E, Barratt ES. 2001. Psychometric properties of an Italian version of the Barratt Impulsiveness Scale-11 (BIS-11) in nonclinical subjects. J Clin Psychol 57:815–828. Fossati A, Di CA, Acquarini E, Donati D, Donini M, Novella L, Maffei C. 2001. The retrospective assessment of childhood attention deficit hyperactivity disorder in adults: reliability and validity of the Italian version of the Wender Utah Rating Scale. Compr Psychiatry 42:326–336. Geller B, Williams M, Zimerman B, Frazier J, Beringer L, Warner KL. 1998. Prepubertal and early adolescent bipolarity differentiate from ADHD by manic symptoms, grandiose delusions, ultra-rapid or ultradian cycling. J Affect Disord 51:81–91. Hamilton M. 1960. A rating scale for depression. J Neurol Neurosurg Psychiatry 23:56–62. Karam RG, Bau CH, Salgado CA, Kalil KL, Victor MM, Sousa NO, et al. 2009. Late-onset ADHD in adults: milder, but still dysfunctional. J Psychiatr Res 43:697–701. Kessler RC, Adler L, Barkley R, Biederman J, Conners CK, Demler O, et al. 2006. The prevalence and correlates of adult ADHD in the United States: results from the National Comorbidity Survey Replication. Am J Psychiatry 163:716–723. Kindlon D, Mezzacappa E, Earls F. 1995. Psychometric properties of impulsivity measures: temporal stability, validity and factor structure. J Child Psychol Psychiatry 36:645–661. Klassen LJ, Katzman MA, Chokka P. 2009. Adult ADHD and its comorbidities, with a focus on bipolar disorder. J Affect Disord. Martel MM, Nigg JT. 2006. Child ADHD and personality/ temperament traits of reactive and effortful control, resiliency, and emotionality. J Child Psychol Psychiatry 47:1175– 1183. Martel MM, Nigg JT, von EA. 2009a. How do trait dimensions map onto ADHD symptom domains? J Abnorm Child Psychol 37:337–348. Martel MM, Pierce L, Nigg JT, Jester JM, Adams K, Puttler LI, et al. 2009b. Temperament pathways to childhood disruptive behavior and adolescent substance abuse: testing a cascade model. J Abnorm Child Psychol 37:363–373. Masi G, Perugi G, Toni C, Millepiedi S, Mucci M, Bertini N, Pfanner C. 2006. Attention-deficit hyperactivity disorder – bipolar comorbidity in children and adolescents. Bipolar Disord 8:373–381. Miller CJ, Newcorn JH, Halperin JM. 2009. Fading memories: retrospective recall inaccuracies in ADHD. J Atten Disord. Monuteaux MC, Seidman LJ, Faraone SV, Makris N, Spencer T, Valera E, et al. 2008. A preliminary study of dopamine D4 receptor genotype and structural brain alterations in adults with ADHD. Am J Med Genet B Neuropsychiatr Genet 147B:1436–1441.
666 S. Bernardi et al. Nierenberg AA, Miyahara S, Spencer T, Wisniewski SR, Otto MW, Simon N, et al. 2005. Clinical and diagnostic implications of lifetime attention-deficit/hyperactivity disorder comorbidity in adults with bipolar disorder: data from the first 1000 STEP-BD participants. Biol Psychiatry 57:1467–1473. Rothbart M. 1989. Temperament and development. In: Kohnstamm G, Bates J, Rothbart M, editors. Temperament in childhood . New York: Wiley. p. 187–247. Rothbart M, Bates J. 2006. Temperament. In: Damon W, Lerner R, editors. Handbook of child psychology. New York: Wiley. p 99–166. Rothbart MK, Ellis LK, Rueda MR, Posner MI. 2003. Developing mechanisms of temperamental effortful control. J Pers 71: 1113–1143. Sachs GS, Baldassano CF, Truman CJ, Guille C. 2000. Comorbidity of attention deficit hyperactivity disorder with early- and late-onset bipolar disorder. Am J Psychiatry 157: 466–468. State RC, Frye MA, Altshuler LL, Strober M, DeAntonio M, Hwang S, Mintz J. 2004. Chart review of the impact of attention-deficit/hyperactivity disorder comorbidity on response to lithium or divalproex sodium in adolescent mania. J Clin Psychiatry 65:1057–1063.
Swann AC, Anderson JC, Dougherty DM, Moeller FG. 2001. Measurement of inter-episode impulsivity in bipolar disorder. Psychiatry Res 101:195–197. Tamam L, Karakus G, Ozpoyraz N. 2008. Comorbidity of adult attention-deficit hyperactivity disorder and bipolar disorder: prevalence and clinical correlates. Eur Arch Psychiatry Clin Neurosci 258:385–393. Watson D, Kotov R, Gamez W. 2006. Basic dimension of temperament in relation to personality and psychopatology. In: Om Krueger R, Tackett J, editors. Personality and psychopatology. New York: Guildford. p. 7–38. Wingo AP, Ghaemi SN. 2008. Frequency of stimulant treatment and of stimulant-associated mania/hypomania in bipolar disorder patients. Psychopharmacol Bull 41:37–47. Wood DR, Reimherr FW, Wender PH, Johnson GE. 1976. Diagnosis and treatment of minimal brain dysfunction in adults: a preliminary report. Arch Gen Psychiatry 33:1453–1460. Young RC, Biggs JT, Ziegler VE, Meyer DA. 1978. A rating scale for mania: reliability, validity and sensitivity. Br J Psychiatry 133:429–435. Young S, Gudjonsson GH. 2008. Growing out of ADHD: the relationship between functioning and symptoms. J Atten Disord 12:162–169.
