Evidence for cognitive subgroups in bipolar disorder and the influence of subclinical depression and sleep disturbances.

European Neuropsychopharmacology (]]]]) ], ]]]–]]]

www.elsevier.com/locate/euroneuro

Evidence for cognitive subgroups in bipolar disorder and the influence of subclinical depression and sleep disturbances J. Volkertn, J. Kopf, J. Kazmaier, F. Glaser, K.C. Zierhut, M.A. Schiele, S. Kittel-Schneider, A. Reif Department of Psychiatry, Psychosomatics and Psychotherapy, University of Wuerzburg, Fuechsleinstrasse 15, D-97080 Wuerzburg, Germany Received 7 April 2014; received in revised form 18 July 2014; accepted 26 July 2014

KEYWORDS

Abstract

Bipolar disorder; Cognition; Neuropsychological functioning; Depression; Sleep disorder

Recent research in bipolar disorder (BD) points to the relevance and persistence of cognitive deficits even in euthymia. Up to now, the mechanisms behind why some bipolar patients (BP) do not reach their former level of cognitive performance and psychosocial functioning while others remit completely, are not understood. In this study we aimed to identify a “cognitive deficit” vs. “non-deficit” subgroup within BD by using an extensive neuropsychological test battery. The test performance of 70 euthymic outpatients (BD-I and II, recruited as a sample of convenience from our bipolar disorder programme) was compared to 70 matched, healthy controls (HC). Furthermore, we investigated the association between demographic/clinical variables and the cognitive performance of BP. As expected, our sample of euthymic BP performed significantly worse than HC in psychomotor speed, divided attention, working memory, verbal memory, word fluency and problem solving. However, 41.4% of the patients did not have any neurocognitive deficits at all, and whether or not a patient belonged to the non-deficit group was not influenced by disease severity. Instead, our results demonstrate that patients suffering from persistent sleep disturbances and sub-threshold depressive symptomatology show more severe cognitive dysfunctions. In addition, antipsychotic treatment and comorbid anxiety disorder were associated with cognitive deficits. In sum, these results suggest that a major part of cognitive impairment is due to current symptomatology, especially sleep disorder and subsyndromal depression. Rigorous treatment of these symptoms thus might well improve cognitive deficits and, as a consequence, overall functioning in BD. & 2014 Elsevier B.V. and ECNP. All rights reserved.

n

Corresponding author. Tel.: +49 931 20177540. E-mail address: [email protected] (J. Volkert).

http://dx.doi.org/10.1016/j.euroneuro.2014.07.017 0924-977X/& 2014 Elsevier B.V. and ECNP. All rights reserved.

Please cite this article as: Volkert, J., et al., Evidence for cognitive subgroups in bipolar disorder and the influence of subclinical depression and sleep disturbances. European Neuropsychopharmacology (2014), http://dx.doi.org/10.1016/j.euroneuro.2014.07.017

2

J. Volkert et al.

1.

Introduction

Cognitive deficits are common during acute episodes of depression and mania in bipolar disorder (BD). However, a great number of bipolar patients (BP) report persistent cognitive impairments even after remission of an acute episode. Recent studies verified cognitive impairments during euthymia with deficits up to one standard deviation below average in psychomotor speed, attention, working memory, long term memory and executive functioning (Ancin et al., 2013; Burdick et al., 2010; Hellvin et al., 2012; Martinez-Aran et al., 2004; Ryan et al., 2012; Thompson et al., 2005). With the exception of crystallised abilities (e.g. vocabulary, word reading) and premorbid IQ, these deficits seem to be global and comparable to schizophrenia, albeit less severe (Vohringer et al., 2013). However, current studies suggest that cognitive deficits in BD are not as severe as previously assumed. For instance, Bourne et al. (2013) conducted an “individual patient data metaanalysis” by including unpublished studies and considering more confounding factors. The authors confirmed cognitive impairments in euthymic BP, but found considerably lower effect sizes (d =.26–.63) than previous meta-analyses (d= .5–1). In line with this result, other studies reported that only a part of their bipolar sample had significant cognitive impairments (Altshuler et al., 2004; Aminoff et al., 2013; Iverson et al., 2011; Martino et al., 2008). Thus, there seem to be subgroups within the cluster of BD; however, the underlying reasons are unclear: Some patients do not reach their former level of cognitive performance and become more disabled with progressing illness, while others remit completely after each episode and maintain a high occupational and social functioning (Gilbert and Marwaha, 2013). The disparity of cognitive impairments depend on the applied criteria of the normal range (Reichenberg et al., 2009). However, due to the heterogeneity in the cognitive profile of BP, group statistics are not sufficient. Mean values convey the risk to suggest a global deficit for all patients and can obscure inter-individual differences. Nevertheless, the accurate description of neurocognitive impairment is essential for the understanding of aetiopathogenesis. An accurate description of neuropsychology in BD allows conclusions about disease specific cortical or subcortical pathologies. More homogeneous subgroups are needed in the search for genetic underpinnings and mechanisms underlying treatment response. In the clinical setting, personalised interventions are needed to improve not only affective symptoms but also cognitive dysfunctioning. This is crucial because cognitive impairments are significant predictor of psychosocial and occupational outcomes in BD (Gilbert and Marwaha, 2013). Given that low psychosocial functioning in turn leads to a high risk of relapse, reduced life quality and rising economic costs (Morselli et al., 2004), the identification of moderator variables associated with cognition in BD has relevant implications for further research and the improvement of treatment strategies. To address this issue, the relationship between cognition and demographic, clinical and treatment variables has been investigated in recent years. Numerous studies found an association between symptom severity and neurocognitive impairments. For instance, significant associations between

cognitive deficits and greater number of episodes have been found (Martinez-Aran et al., 2004; Thompson et al., 2005). Furthermore, higher incidences of manic episodes (Aminoff et al., 2013; Bourne et al., 2013; Lopez-Jaramillo et al., 2010), longer duration of disease (Ryan et al., 2012; Torrent et al., 2012), and more hospitalisations were described to be associated with cognitive impairment (Ancin et al., 2013; Ryan et al., 2012). Therefore, some authors suggested that severe mood symptoms could act like stress-induced neurotoxins (Lopez-Jaramillo et al., 2010). In contrast, other studies found that first-episode BP patients have the same neurocognitive performance as patients with multiple episodes in the past (Bombin et al., 2013; Hellvin et al., 2012). Torrent et al. (2012) described in a recently published longitudinal study that cognitive impairments (except for a worsening of executive functions) remained stable over nine years, irrespective of severe relapses in the meantime. Hence, several authors concluded that cognitive impairments are present in the early beginning of BD and that there is an evidence of a neurocognitive decline with illness progression. These results led to considerations of cognitive deficits as trait effect and the idea of them being a bipolar endophenotype (Bora et al., 2009). Another often reported finding is that patients diagnosed with BD Type-I have more cognitive deficits than patients with BD Type-II (Aminoff et al., 2013; Ancin et al., 2013; Palsson et al., 2013; Torrent et al., 2006). However, Dittmann et al. (2008) and Chaves et al. (2011) found no neuropsychological differences between these two diagnosis types. The fact that patients with BD Type-I have full-blown manic episodes and more psychotic symptoms could contribute to these inconsistent findings (Palsson et al., 2013), as some authors found indeed a significant association between psychotic symptoms and cognition (Aminoff et al., 2013; Bora et al., 2011), while others did not (Brissos et al., 2011). Furthermore, individuals diagnosed with BD Type-I are more often treated with antipsychotics compared to BD-II, which could contribute to cognitive impairments in this patient group (Arts et al., 2011; Jamrozinski et al., 2009; Palsson et al., 2013; Torrent et al., 2011). If other drugs or polypharmacy have negative effects on cognition has not been fully understood yet. Mood stabilisers, especially lithium, interestingly only exerts a marginal (Dias et al., 2012) or no influence on cognition (Altshuler et al., 2004; Arts et al., 2011; LopezJaramillo et al., 2010). Moreover, a longitudinal study pointed out neuroprotective effects of lithium (Diniz et al., 2013). Antidepressants seem to have no remarkable adverse cognitive effects, aside from the anticholinergic effects of tricyclics (Amado-Boccara et al., 1995). In summary, only antipsychotics seem to have a negative effect on cognitive performance in BD. However, cognitive dysfunctions cannot fully be explained by drug side effects because even in medication-free euthymic BP, neurocognition is impaired (Bourne et al., 2013; Goswami et al., 2009). In summary, cognitive deficits in remitted BP have been repeatedly demonstrated, but findings are inconsistent regarding the severity of impairments and associated disease characteristics. In the present study, we aim to scrutinise neuropsychological functioning in BD. Our hypotheses were that BP as a group show cognitive deficits compared to healthy controls (HC). Furthermore, we


Evidence for cognitive subgroups in bipolar disorder and the influence of subclinical depression and sleep disturbances assume that part of the patients on the individual level however do not have any cognitive impairments. In order to identify such a “cognitive deficit” vs. “non-deficit” subgroup and to delineate clinical correlates of cognitive impairments in BD, we conducted an extended neuropsychological test battery with euthymic BP and HC. We were interested in the severity of cognitive impairments in BP compared to HC and aimed to detect possible clinical differences, or moderators, between BP with and without cognitive deficits, by recording numerous sociodemographic and disease specific variables of patients.

2.

Experimental procedures

The study was performed in the outpatient clinic of the Bipolar Programme of the Psychiatric University Hospital Wuerzburg. All procedures followed the Declaration of Helsinki in its latest version and were approved by the Ethical Committee of the University Hospital Wuerzburg. Written informed consent was obtained from all participants.

2.1.

Participants

The patient group consisted of 70 outpatients with BD (diagnosed by DSM-IV criteria based on clinical interviews). 33 (47.1%) patients suffered from BD Type-I and 37 (52.9%) from BD Type-II. Patients were recruited irrespective of subjective complaints of cognitive impairments. At the time of the neuropsychological testing, affective symptoms were measured with clinical interviews and questionnaires. We applied the Montgomery–Asperg Depression Ratings Scale (MADRS; Montgomery and Asberg, 1979), the Beck Depression Inventory, Second Edition (BDI-II; Beck et al., 1996) and the Young Mania Rating Scale (YMRS; Young et al., 1978). BPs were included in the sample only if they were fully remitted for at least three months (according to patients' subjective appraisal and the above ratings). Criteria for euthymia were rating scores of MADRSo12, BDI-IIo13 and YMRSo5 points. Furthermore patients had to be on a stable medication for at least three months. Participants were 18– 55 years old and native German speakers. Intelligence Quotient (IQ) 485, estimated via the German multiple-choice word test (MWT-B; Lehrl et al., 1995), was another inclusion criterion. Exclusion criteria were previous head trauma, neurological illnesses, schizoaffective disorder or present substance abuse. Patients who received electroconvulsive therapy (ECT) in the preceding six months were excluded as well. The sample of HC consisted of 70 volunteers which were recruited by web-based announcements. Controls were screened by the Mini International Neuropsychiatric Interview, German Version 5 (Sheehan et al., 1998) to exclude persons with history of Axis I disorder. Furthermore, we applied the following exclusion criteria for HC: substance abuse (life time), neurological diseases and taking of psychotropic drugs (life time). HC were selected to have a stringent matching (gender, age, and years of education) to the BP sample.

2.2.

Clinical assessment

The clinical parameters of the bipolar sample were recorded via structured interviews with the patients, their relatives, and clinical records. The following clinical variables were collected (lifetime presence): number of episodes (depression, mania/hypomania, mixed episodes), bipolar subtype, age at illness onset, number of hospitalisations, polarity of the first episode, predominant polarity, duration of euthymia, attempted suicides, substance abuse, psychotic symptoms, medication, ECT, family history of mental disorders, and comorbid somatic (e.g. thyroid diseases) or mental

3

illnesses, especially anxiety disorders and attention deficit disorder (ADHD). Patients were asked if they had persistent sleep disturbances (initial or middle insomnia) in the four weeks preceding the interview. Furthermore, current mood of all participants was recorded by the Positive and Negative Affect Scale (PANAS; Watson et al., 1988) prior to neuropsychological testing. In order to assess psychosocial functioning, all patients were evaluated with the Global Assessment of Functioning Scale (GAF; Hall, 1995).

2.3.

Neurocognitive assessment

The neurocognitive testing was administered in a fixed order by the same trained, experienced clinical psychologist. It lasted 90 min, including a 5-min break. The standardised test battery consisted of eight tests, each measuring one part of the following cognitive domains. Psychomotor speed was assessed by the Subtest Alertness of the Test battery of Attentional Performance (TAP; Zimmermann, 2011) and the compatible trials of the Stroop Interference Test (Words/ Colours; Puhr and Wagner, 2011). Attention was measured using the Subtest Divided Attention of the Test battery of Attentional Performance (TAP), where patients were asked to pay attention on two stimuli at the same time. The California Verbal Learning Test (CVLT; Niemann et al., 2011) was used to measure Short- and Long Term Memory. Scores of total verbal learning, immediate and delayed recalls were considered. Executive functioning was assessed by the subtest working memory (TAP test battery) which is an n-back task, and incompatible trials of the Stroop Interference Test (Reading/ Naming) which is considered to measure the ability to control automatic processes. Planning and problem solving was assessed by the Tower of London Test (TOL; Kaller et al., 2011). Furthermore, the number of generated words in a phonemic word fluency test (RWT; Aschenbrenner et al., 2000) and the Subtest Cognitive Flexibility of the TAP test battery (Shifting task) were used as measures for executive functioning. For all tests, individual reactions times (RTs), number of correct reactions, errors or omissions were conducted. Except CVLT and word fluency, all tests were computerised.

2.4.

Statistical analysis

Statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS), version 19.0. In order to detect group differences in demographic variables between BP and HC independent samples t-tests or chi-square analyses were used. Group differences in neuropsychological test measures were tested by independent sample t-tests. Before these analyses, the neurocognitive variables were checked for outliers. For all tests the significance level was set at po.05. Furthermore, effect sizes (Cohen's d value) were calculated for each comparison. Patients were divided into cognitive deficit and non-deficit subgroups on the basis of their cognitive test performance. Since in the subtest Alertness, 10 HC (compared to 12 BP) had scores below average, this test score does not appear to be a valid feature to indicate cognitive deficits. Because of this, this test score was not considered in regard to the classification of cognitive subgroups. Consequently, the cut-off score relies on 12 test scores (without the subtest Alertness). If a patient scored below the average of the normative data group (at least 1.5 SD) in one of the 12 tests, he was assigned to the deficit-subgroup (Hellvin et al., 2012). Again, group comparisons (deficit- vs. non-deficit BD) in demographic and clinical variables were conducted using independent sample t-tests or chi-square tests. Due to higher validity, normative data of the test manuals were used as reference value. Thereby we were able to check potential cognitive impairments in our healthy control sample. In addition, multiple linear regression analyses (stepwise approach) were calculated in order to test the impact of clinical and demographic


4

J. Volkert et al.

variables on cognition in the deficit and non-deficit subgroups. Due to the fact that we wanted to include categorical variables (e.g. antipsychotics), and to avoid multiple testing, we decided to use a multiple regression instead of correlation analyses. To restrict the number of predictors, we examined only the variables that differed significantly between deficit and non-deficit subgroups. Furthermore, we included antipsychotic medication because the literature suggests a high association to cognition. In sum, age, antipsychotics, depressive symptoms (BDI-II), comorbid anxiety and sleep disorder (MADRS sleep item) were entered as predictors. Furthermore, correlation analyses (Pearson coefficients) were calculated between MADRS score and number of previous episodes; and between cognitive measures and total number of medications. In order to control the influence of age on the cognitive measures we performed a mediator analyses and calculated partial correlations with age as covariate (Supplemental material S2).

3.

Results

3.1.

Sample characteristics

Demographic and clinical characteristics of BP and HC are presented in Table 1. There were no group differences regarding age, premorbid IQ and gender between groups. Despite our effort of a stringent matching between groups, BP shows significantly more years of education than HC. Participants' actual mood (measured by PANAS score) showed a significant reduced positive and increased negative affect in patients with BD compared to HC (Table 1) despite all patients met strict criteria for euthymia. All patients except five with BD-II were medicated with mood stabilisers: lithium (57.1%), valproate (20%), and lamotrigine (4.3%). Furthermore 51.4% were treated with secondgeneration antipsychotics (SGA) and one patient with a firstgeneration antipsychotic. 60% were on monotherapy (particularly lithium or SGA) and 40% on polypharmacy with the combination of lithium and antipsychotics or lithium and other mood stabilisers. In addition to mood stabilisers, 54.3% were treated with antidepressants. None of the patients were medicated with benzodiazepines or barbiturates. HC were completely medication free. Subclinical depression, defined by a MADRS score 47 (Vieta et al., 2010) was reported in 35.7% of the patients. 37.1% of the patients reported initial or middle insomnia during the last four weeks prior the test session. No patients showed subclinical

Table 1

manic symptoms (assessed by the YMRS). Furthermore, patients reported being in a euthymic state over a period of 38.5 (SD=15.8) weeks. On an average, mood ratings showed a score of 6.9 points (SD=2.9) in the MADRS, 5.6 points (SD=4.2) in the BDI-II and 1.4 points (SD=1.5) in the YMRS. Interestingly, we found a significant positive correlation between MADRS score and the number of previous episodes (r=.282; p=.019). It seems that more episodes a patient underwent, the more subthreshold depressive symptoms are present after remission.

3.2. Comparison of cognitive performance between bipolar patients and healthy controls BP showed a significant reduced performance compared to HC in eight of 13 cognitive tasks: In detail, patients had cognitive deficits in word fluency, the compatible trials of the Stroop Test, the delayed recall in the CVLT, the Tower of London Test, the working memory, and the immediate recall of the CVLT (for more detail see Table 2). There were no significant differences in the Alertness subtest, the total score of Verbal Learning (CVLT), the Cognitive Flexibility and the incompatible trials of the Stroop Test between BP and HC.

3.3. Identification of a deficit vs. non-deficit subgroup Based on the severity of cognitive deficits (cut-off score: 1.5 SD below the average of normative data group as defined before data collection), the patients sample was subsequently divided into two subgroups. 29 (41.4%) patients had no clinically significant cognitive impairments, while 41 (58.6%) showed deficits in at least one of 12 cognitive domains. Among this cognitive deficit-subgroup, 21 patients (30%) were impaired on one neuropsychological test, 12 patients (17.1%) on two, and 9 (12.8%) on 3 or more tests. Interestingly, some HC (N=16; 22.9%) also showed clinically significant cognitive impairments after controlling for outliers (see Supplemental material, Table S1). Specifically, these HC were also impaired on the tests Alertness and Divided attention (see Supplemental material, Table S1).

Demographic characteristics of bipolar patients (BP) and healthy controls (HC).

Age Gender f/m Years of education Premorbid IQa PANASb PAc PANAS NAd

BP (N= 70) N/Mean (SD)

HC (N =70) N/Mean (SD)

39.3 (11) 42/28 11.7 (1.5) 30.5 (5.8) 30.3 (5.6) 11.9 (2.4)

39 (9.2) 40/30 11.0 (1.5) 31.4 (2.7) 32.7 (5.5) 11.2 (2)

t/χ2 .15 .12 2.74 1.91 2.49 1.89

p-Value .881 .731 .007nn .058 .014n .060

a

Verbal IQ via multiple-choice word test (MWT-B), Positive and Negative Affect Scale (momentary mood prior the testing) c Score Positive Affect d Score Negative Affect. b


Evidence for cognitive subgroups in bipolar disorder and the influence of subclinical depression and sleep disturbances

Table 2

5

Group analyses of cognitive performance in bipolar patients (BP) and healthy controls (HC).

Cognitive domain

BP (N =70) Mean (SD) HC (N =70) Mean (SD) t test

Psychomotor speed TAP alertness (RT) Stroop compatible words (RT) Stroop compatible colours (RT)

273.6 (54.6) 762.8 (109) 712.1 (110.5)

277.9 (68.9) 707.2 (83) 664.8 (85.4)

1.7 (1.7)

1.5 (1.5)

Memory CVLT (total verbal learning) CVLT (immediate recall) CVLT (delayed recall)

58.7 (9.2) 12.1 (2.3) 12.2 (2.3)

59.4 (9.5) 13.1 (2.7) 13.4 (2.4)

.421 .677 2.36 .019n 2.94 .004nn

Executive functions TAP working memory (omissions) TAP cognitive flexibility (errors) Stroop incompatible (reading) Stroop incompatible (naming) Word fluency (number of words) Tower of London (problem solving)

2.4 1.8 .137 .12 14.9 6.2

1.5 1.9 .136 .104 21.9 7.4

2.59 .552 .110 .853 7.09 3.29

Attention TAP divided attention (omissions)

(2.2) (1.8) (.1) (.1) (5.2) (2.4)

(1.9) (2.7) (.1) (.66) (6.3) (1.9)

p-Value Cohen’s effect size (d)

.403 .688 3.38 .001nn 2.83 .005nn 2.15

.033n

.07 .58 .48 .38 .08 .42 .51

.44 .010n .582 .04 .913 .01 .395 .03 .000nnn 1.22 .56 .001nn

TAP= Test Battery of Attentional Performance, CVLT=California Verbal Learning Test, RT= Reaction Time

3.4. Clinical differences between cognitive subgroups of bipolar patients The aim of dividing the patient sample into a deficit vs. nondeficit subgroup was to detect variables which are either associated with moderate or cognitive impairment. In doing so, BP with cognitive deficits reported significantly more sub-threshold depressive symptoms (BDI-II Score), higher scores on the item “reduced sleep” in the MADRS, and more often persistent initial and middle insomnia. Patients with cognitive deficits had more often comorbid anxiety disorder. Furthermore, there was a numerical difference between the subgroups in comorbid ADHD (see Table 3). Interestingly, our results showed that subgroups did not significantly differ from each other in disease characteristics like duration of illness, number of previous episodes (i.e., variables indicating disease severity) or bipolar subtype (Table 3). It should be noted that the deficit subgroup was significantly older compared to the non-deficit subgroup, which could have an influence on cognitive functioning. However, partial correlations showed that the associations between test scores and sleep disorder were lower after controlling for age but still remained significant (see Supplemental material, Table S2).

3.5. Cognitive performance and associated disease variables In addition to the subgroup analysis we conducted a multiple linear regression analyses to test associations of clinical variables with cognitive functioning in the bipolar sample. The results again showed significant influences of persisting sleep disturbances, sub-threshold depressive symptoms and antipsychotic medication on cognitive deficits in BD (see Table 4). In detail, the variance of Alertness [R2 =.282; F(2,67)=14.3,

p=.000], the compatible trials of Stroop Test Words [R2 =.441; F(2,69)=28.2, p=.000] and Colours [R2 =.466; F (2,68)=31.1, p=.000] were best predicted by sleep disorder and age. Long term memory, measured by CVLT, seemed to be influenced by sleep disorder and antipsychotic medication [R2 =.301; F(1,68)=26.1, p=.000]. The immediate recall (CVLT) was also predicted by sleep disorder, followed by age and antipsychotics [R2 =.326; F(3,68)=12.1, p=.000]. Delayed recall (CVLT) was best predicted by sleep disorder and antipsychotics [R2 =.426; F(2,68)=26.6, p=.000]. The Interference score (Reading) was best explained by sub-threshold depressive symptoms (BDI-II) and sleep disorder [R2 =.147; F(2,67)=6.85, p=.002]. The Interference Score (Naming) was predicted by sub-threshold depressive symptoms, followed by age [R2 =.264; F(3,67)=13.2, p=.000]. Executive functions like Word fluency and Problem solving could not be explained well by the entered variables. Cognitive flexibility and Working Memory could not be explained by any of the chosen variables. In summary, the multiple regression analysis of our data showed that psychomotor speed and attention could be excellently explained by sleep disturbances. Verbal memory was influenced by sleep disturbances and a prescription of antipsychotics. Apart from interference (incompatible trials of Stroop) which was predicted well by sub-threshold depression, the other executive functions like Working Memory, Cognitive flexibility, Word fluency and Problem Solving could only marginally be explained by our model and hence might be genuine disease specifics of the deficit subgroup. Comorbid anxiety did not explain any variance of our test measures. Given that many of the patients were medicated with more than one drug, correlation analyses were conducted in order to test whether there is a relationship between the total number of drugs and cognitive deficits. At this, we found significant positive correlations between the total number of drugs and Cognitive flexibility (r=.293, p=.014). Furthermore


6

J. Volkert et al.

Table 3

Clinical characteristics in bipolar subgroups deficit- vs. nondeficit.

Demographics Gender f/m Age Years of education Verbal IQ Disability pension Current mood and sleepa MADRS score BDI II score YMRS score Item “reduced sleep” (MADRS) Initial/middle insomnia (y/n) Clinical variablesb (Life time) Diagnosis type: BD-I Family history Age of onset Duration of illness (years) Duration of euthymia (weeks) No previous depressive episodes No previous (hypo-)/manic episodes No previous hospitalisations Attempted suicide Predominate polarity History of psychosis Alcohol abuse Comorbid ADHD Comorbid anxiety Chronic thyroid disease ECT GAF score Medication Lithium Other mood stabilizerc Antipsychotics Antidepressants

BP (N= 41) deficit-subgroup N (%) /Mean (SD)

BP (N = 29) nondeficit-subgroup N (%) /Mean (SD)

24/17 41.6 (11.2) 11.6 (1.6) 30.3 (2.2) 10 (24.4%)

18/11 36.2 (10.2) 11.9 (1.43) 30.8 (2.5) 2 (6.9%)

t or χ2

p-Value

.088 2.04 1.01 .775 6.45

.766 .045n .315 .441 .168

7.4 7.1 1.5 1.5 19

(3.3) (4.5) (1.7) (1.4) (46.3%)

6.5 4.1 1.3 .7 7

(2.1) (3.1) (1.2) (.9) (24.1%)

1.22 3.03 .701 2.44 3.59

.227 .003nn .486 .017n .049n

22 35 26.4 15.1 35.9 7.9 5.6 3.7 22 21 15 6 6 9 7 3 65.6

(53.7%) (85.4%) (10.6) (10.7) (14.3) (7.3) (6.4) (3.1) (53.7%) (51.2%) (36.6%) (14.6%) (14.6%) (22%) (17.1%) (7.3%) (14.9)

11 23 23.6 12.6 42.1 7.5 5.1 3 9 17 5 6 1 1 3 3 74.3

(37.9%) (79.3%) (8.1) (8.9) (17.3) (7.6) (4.8) (3.1) (31%) (58.6%) (17.2%) (20.7%) (3.4%) (3.4%) (10.3%) (10.3%) (9.8)

1.69 .439 1.21 1.04 1.62 .274 .316 .988 3.52 .375 3.11 .439 2.36 4.75 .628 .199 2.73

.194 .508 .229 .303 .110 .814 .737 .364 .051 .540 .078 .363 .124 .029n .428 .656 .008n

25 10 22 20

(61%) (24.4%) (53.7%) (48.8%)

15 7 14 18

(51.7%) (24.1%) (48.3%) (62.1%)

.594 .001 .197 1.21

.441 .981 .657 .196

MADRS= Montgomery-Asperg Depression Ratings Scale, BDI II = Beck Depression Inventory,YMRS= Young Mania Rating Scale, ADHD= Attentional Deficit Hyperactivity Disorder a Period of 2 weeks prior testing b Data are based on self-report of patients and relatives (life time) c Valproate, Lamotrigine, Carbamazepine.

total CVLT verbal learning (r= .288, p=.016), short recall (r= .246, p=.041) and delayed recall (r= .406, p=.001) were significantly associated with number of medications.

4.

Discussion

In the present study we investigated the cognitive functioning in euthymic BP in order to delineate a deficit vs. nondeficit subgroup. The cognitive performance of our bipolar sample differed significantly from HC in psychomotor speed, divided attention, working memory, long term verbal

memory, word fluency and problem solving. Therefore, we confirmed previously reported cognitive deficits in attention, memory and executive functioning in BD. 58.6% of our patient sample showed measurable cognitive impairments, which is in line with previous studies (Bora et al., 2010; Martino et al., 2008), while 41.4% of BP were not impaired at all, which interestingly was not due to disease load (number of previous episodes, duration of illness). Also other clinical variables like bipolar type, age of onset, predominant polarity, or comorbid somatic diseases do not seem to have any impact on cognitive deficits. Therefore, our results argue in favour of the existence of cognitive


Evidence for cognitive subgroups in bipolar disorder and the influence of subclinical depression and sleep disturbances

Table 4 Regression coefficients of the regression analyses in bipolar patients. Cognitive domain Psychomotor speed TAP alertness (RT)

Predictors

Beta

p-Value

Age .35 Sleep disorder .29 Sleep disorder .48 Age .32 Sleep disorder .50 Age .31

.003 .012 .000 .002 .000 .002

.37

.002

Stroop word condition (RT) Stroop colour condition (RT) Attention TAP divided attention Sleep disorder (omissions) Memory CVLT (total verbal Sleep disorder learning) CVLT (immediate recall) Sleep disorder Age Antipsychotics CVLT (delayed recall) Sleep disorder Antipsychotics Executive functions TAP working memory — (omissions) TAP cognitive flexibility ——— (errors) Stroop interference Depression score reading Sleep Stroop interference Depression score naming Age Word fluency Antipsychotics (number of words) Tower of London Age (planning score)

.43

.000

.55 .29 .23 .55 .37

.002 .008 .026 .000 .000

—

—

—

—

.457 .28 .41 .27 .29

.001 .029 .000 .015 .016

.27

.024

BP= Bipolar patients, HCs= Healthy Controls, TAP= Test Battery of Attentional Performance, CVLT=California Verbal Learning Test, RT = Reaction Time

subgroups in BD. Interestingly, we found associations between current symptoms of BP and neuropsychological test performance: sleep disorder, sub-threshold depression, and comorbid anxiety were highly prevalent in BP with cognitive deficits, making these prime candidates in treatment to ameliorate cognitive deficits. Furthermore, antipsychotic treatment and polypharmacy were related to cognitive dysfunctions. Therefore, our results suggest that partial remission and secondary symptoms are central to cognitive impairments in BD. While the latter can easily be explained to be connected to cognitive impairment (see below), the question remains open whether there is a subgroup of patients that persistently display affective symptoms, which go along with cognitive impairment (“bipolar residual state”), or whether these patients just take a longer time to fully remit. To answer this question, familybased and longitudinal studies are needed which can be informed by the present analysis. Sleep disturbances, measured by the item “reduced sleep” of the MADRS, were highly associated with psychomotor slowing, as well as worse performance in divided attention, working memory and verbal long term memory.

7

The more sleep disturbances the patients reported, the worse they performed on cognitive tests. Sleep disorders have a high prevalence in euthymic BP. According to the literature, 15–60% of remitted BP show a delayed sleep onset and reduced sleep time, examined by self-report and actigraphy (Brill et al., 2011; Giglio et al., 2009; Sylvia et al., 2012). Moreover, recent research demonstrated that BD is highly associated with sleep and circadian rhythm abnormalities (Murray and Harvey, 2010; Scott, 2011). We could confirm these results, given that 37% of our bipolar sample reported initial or middle insomnia and a reduced sleeping time. Research about experimentally induced sleep deprivation in healthy people observed decreased response times, impairments in working memory and reduced learning (Durmer and Dinges, 2005). Furthermore, neurophysiological studies showed that sleep deprivation affects executive functioning (Drummond et al., 2001). Regarding these findings, it is plausible that BP with chronic sleep disturbances suffer from cognitive impairments. Accordingly, our results show that BP with cognitive deficits suffered significantly more often from persistent initial and middle insomnia compared to patients without deficits. Two recently published reviews described sleep disruption in BD and suggested a possible relationship to cognition (Boland and Alloy, 2013; McKenna and Eyler, 2012). To our knowledge, this is the first study which systematically examined sleep disturbances in BP and linked it to cognitive impairments. More research is needed to examine the role of sleep on cognitive functioning. Furthermore, it should be investigated to what extent the treatment of sleep disturbances improves cognitive performance in BP. On the other hand, only half of our cognitive deficit subgroup suffered from sleep disturbances, suggesting further parameters influencing cognitive performance in BD. Another variable which was also associated with cognitive deficits was sub-threshold depressive symptoms. We found that the deficit-subgroup had significantly higher depression scores compared to the non-deficit subgroup. According to our results, some authors demonstrated a relationship between residual mood symptoms and impairments in verbal memory and executive functioning (McKay et al., 1995; Palsson et al., 2013; Torrent et al., 2012). In a study by Bonnin et al. (2012) the degree of cognitive functioning was solely predicted by sub-depressive symptoms and a meta-analysis by Mann-Wrobel et al. (2011) reported that more rigorously defined euthymia goes along with smaller impairments. This result might be surprising, because traditionally BD has been described as a condition that is episodic in nature. However, recent studies revealed that patients frequently suffer from persistent residual mood symptoms. In comparison to HC, BP shows significantly higher scores in depression rating scales despite being stable for at least 6 months (Vieta et al., 2008). Accordingly, our results showed differences in current mood (measured by PANAS) in BP compared to HC. Patients reported significantly less positive affect compared to HC. Furthermore, our criteria for euthymia in MADRS were not very restrictive. While many patients of our sample subjectively perceived themselves as fully remitted, the specific questions in MADRS and BDI-II pointed to at least some persisting depressive symptoms. One explanation for that finding could be that patients are biased in the assessment of their mood after recovering from severe affective episodes. Thus, our results


8

J. Volkert et al.

indicate that remitted BP show cognitive impairments because they are not fully remitted. Unfortunately, many previous studies did not report criteria for the duration of euthymic mood in their samples, or did not control for sub-depressive symptoms (Bombin et al., 2013; Ryan et al., 2012; Thompson et al., 2005) which might well account for discrepancies in the published literature. Some investigations reported high depression scores in their samples and therefore included more symptomatic patients, which could be responsible for larger effects (Aminoff et al., 2013; Chaves et al., 2011; Levy et al., 2013). Other studies conducted neuropsychological tests after a very short period of remission, which is still a relatively unstable state (Brissos et al., 2011; Dittmann et al., 2008). Taken together, an exact definition of remission seems to be a general problem in bipolar disorder and rigorous criteria for euthymia are needed. Interestingly, only the BDIII and not the MADRS showed a significant difference between the cognitive deficit- and nondeficit subgroups. According to Svanborg and Asberg (2001), both scales are highly intercorrelated, but the self-report questionaire BDI-II rather assesses depressive cognitive attitudes while the MADRS seems to focus on core depressive symptoms and functional impairment. Therefore, the BDI-II is supposed to be more sensitive in milder forms of depression and we assume that this could be an explanation that we did not detect subthreshold depressive symptoms with the MADRS scale. The assumption that many BP show cognitive dysfunctions because they are not fully remitted could explain why cognitive deficits are prevalent even in the early course of the disorder: even after a single depressive episode mood symptoms can persist, which in turn could affect cognitive functioning. Furthermore, our results suggest that subsyndromal depression moderates the often observed relationship between illness severity (e.g. number of previous episode, psychotic symptoms) and impaired cognition in BD, as we found a significant positive correlation of MADRS depression score and number of previous episodes. The more episodes a patient underwent, the more subthreshold depressive symptoms were present after remission. Accordingly, (Benazzi, 2001) demonstrated that a history of more episodes increases the probability of persistent sub-threshold depression after remission. In addition, severity of symptoms in the past seems to influence persisting mood symptoms. Levy et al. (2013) found that BP with psychotic symptoms in their last episode had more residual mood symptoms, more impairment in memory and executive tasks, and a lower psychosocial functioning compared to remitted patients without psychotic symptoms in their last episode. However, the links between residual mood symptoms and cognitive performance are not yet fully understood. Possibly, sub-depressive symptoms worsen cognitive functioning due to low motivation or an impaired reward system, and therefore reduce sustained effort during the test (Millan et al., 2012). Another explanation could be the existence of marginal psychomotor slowing, which is frequently present during acute depressive episodes (Ryan et al., 2012). Previous studies demonstrated that subsyndromal depression has a crucial impact on the occupational and functional outcomes of BP (Burdick et al., 2010; Gilbert and Marwaha, 2013; Reinares et al., 2013). We suggest that this association is moderated by sub-depression induced cognitive dysfunctions. Accordingly, we found a significant lower psychosocial functioning in the cognitive deficit group

compared to cognitively intact patients, supporting that notion. Regardless of the precise mechanism how subthreshold depression influences cognition, these data highlight the necessity to test for and rigorously treat minor depressive symptoms. Besides residual mood symptoms and sleep disturbances, we found marginal associations between antipsychotics and some cognitive measures, which were previously reported in BD (Palsson et al., 2013; Torrent et al., 2011). Moreover, our data indicate that polypharmacy has an influence on cognitive flexibility and verbal memory. Therefore, medication-induced cognitive deficits could also play a role in our findings. Another interesting result of the present study is that comorbid anxiety is significantly more prevalent in the cognitive deficitsubgroup. 22% of BP patients with cognitive impairments suffered from comorbid anxiety disorder. Accordingly, a significant association between comorbid anxiety and reduced word fluency has been demonstrated (Palsson et al., 2013). Wu et al. (2011) compared the neurocognitive performance of euthymic BP-II patients with and without comorbid anxiety and found persistent impairments of verbal and non-verbal memory, psychomotor speed and working memory in those BP with comorbid anxiety. Levy (2013) found a relationship between higher arousal during neuropsychological tasks and worse test performance. The authors suggested that acute physiological anxiety could be responsible for worse cognitive performance in BD. However, we found no significant difference in current anxiety and agitation (PANAS scale) between the deficit and non-deficit subgroups. To our knowledge, comorbid anxiety has not been included as a moderator variable in studies of cognition in BD apart from these two studies. However, it is well known that anxiety has a negative influence on learning and memory, mediated by chronically increased stress hormones (Lupien et al., 2007). Besides the incidence of anxiety, we detected that BP with cognitive deficits more often had comorbid ADHD compared to patients without deficits, albeit this difference is not statistically significant due to the small number of co-morbid patients. Nevertheless, it is an important factor to consider because the comorbid presentation of BD and ADHD seems to be associated with a more severe course of disease, more severe mood symptoms, and lower psychosocial functioning (Klassen et al., 2010). In a study by Rucklidge (2006), BP showed marginal cognitive deficits compared to HC, but this impairment was significantly increased in the presence of comorbid ADHD. These findings suggest that comorbid anxiety and ADHD seem to have a considerable negative impact on patients' cognitive functioning. Taken together, since ADHD and anxiety are highly co-morbid with BD, future studies should address their influence on cognition by comparing BP with or without comorbid anxiety and ADHD. The strengths of the present study are the use of an extensive test battery, providing a broad neurocognitive profiling and the comparison to an age-, gender- and education-matched HC group. Furthermore, the patient sample was well characterised, and we considered numerous clinical variables which were investigated in previous studies. One limitation of the study is that the clinical variables were recorded retrospectively based on information given by patients, relatives, and clinical records. These are very subjective measures and thus entail the risk of recall biases. Furthermore, we did not assess sleep disturbances by using an evaluated questionnaire or actimetry as


Evidence for cognitive subgroups in bipolar disorder and the influence of subclinical depression and sleep disturbances the main focus of the study was not on the influence of sleep disorder on cognitive performance. In summary, we could identify a cognitive deficit vs. nondeficit subgroup in our bipolar sample, with about six out of 10 patients showing cognitive impairment during remission. Residual mood symptoms and persistent sleep disturbances seem to have a high influence on cognitive performance. Furthermore, our study revealed that comorbid anxiety disorder, antipsychotic medication and polypharmacy are more prevalent in the deficit subgroup, which could contribute to cognitive deficits. We found on the other hand no significant relationship between cognitive functioning and clinical variables indicating illness severity and diagnosis type. It is unclear however whether the latter are due to a distinct subgroup of “residual BD” or not. Further research should address this question as well as if cognitive deficits can be enhanced with optimised antidepressant treatment and adequate sleep regulation. In clinical settings, residual mood symptoms, sleep disturbances and comorbid diseases should be considered in the treatment of BP who report cognitive deficits.

Role of funding source JV was supported by a grant of the German Excellence Initiative to the Graduate School of Life Sciences, University of Wuerzburg. JK and AR received support by the DFG and Länder funds RTG 1256/2 “gk emotions”, AR by the DFG-funded study “earlyCBT” (BA 1504/ 7-1), the DFG funded study SFB TRR 58 Z02, and the Comprehensive Heart Failure Center Würzburg funded by the BMBF (project 01EO1004).

Contributors JV and AR designed the study and wrote the protocol. JV managed the literature searches and analyses, undertook the statistical analysis, and wrote the first draft of the manuscript. All authors contributed to and have approved the final manuscript.

Conflict of interest None of the authors have conflict of interest with the contents of this paper or financial ties to disclose.

Acknowledgements We acknowledge all participants who took part in the present study.

Appendix A.

Supporting information

Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/ j.euroneuro.2014.07.017.

References Altshuler, L.L., Ventura, J., van Gorp, W.G., Green, M.F., Theberge, D.C., Mintz, J., 2004. Neurocognitive function in clinically stable

9

men with bipolar I disorder or schizophrenia and normal control subjects. Biol. Psychiatry 56, 560–569. Amado-Boccara, I., Gougoulis, N., Poirier Littre, M.F., Galinowski, A., Loo, H., 1995. Effects of antidepressants on cognitive functions: a review. Neurosci. Biobehav. Rev. 19, 479–493. Aminoff, S.R., Hellvin, T., Lagerberg, T.V., Berg, A.O., Andreassen, O.A., Melle, I., 2013. Neurocognitive features in subgroups of bipolar disorder. Bipolar Disord. 15, 272–283. Ancin, I., Cabranes, J.A., Santos, J.L., Sanchez-Morla, E., Barabash, A., 2013. Executive deficits: a continuum schizophrenia-bipolar disorder or specific to schizophrenia? J. Psychiatr. Res. 47, 1564–1571. Arts, B., Jabben, N., Krabbendam, L., van Os, J., 2011. A 2-year naturalistic study on cognitive functioning in bipolar disorder. Acta Psychiatr. Scand. 123, 190–205. Aschenbrenner, S., Tucha, O., Lange K.W., 2000. Regensburger Wortflüssigkeitstest. Handanweisung. Hogrefe Verlag, Göttingen. Beck, A.T., Steer, R.A., Ball, R., Ranieri, W., 1996. Comparison of Beck Depression Inventories – IA and II in psychiatric outpatients. J. Personal. Assess. 67, 588–597. Benazzi, F., 2001. Prevalence and clinical correlates of residual depressive symptoms in bipolar II disorder. Psychother. Psychosom. 70, 232–238. Boland, E.M., Alloy, L.B., 2013. Sleep disturbance and cognitive deficits in bipolar disorder: toward an integrated examination of disorder maintenance and functional impairment. Clin. Psychol. Rev. 33, 33–44. Bombin, I., Mayoral, M., Castro-Fornieles, J., Gonzalez-Pinto, A., de la Serna, E., Rapado-Castro, M., Barbeito, S., Parellada, M., Baeza, I., Graell, M., Paya, B., Arango, C., 2013. Neuropsychological evidence for abnormal neurodevelopment associated with early-onset psychoses. Psychol. Med. 43, 757–768. Bonnin, C.M., Sanchez-Moreno, J., Martinez-Aran, A., Sole, B., Reinares, M., Rosa, A.R., Goikolea, J.M., Benabarre, A., AyusoMateos, J.L., Ferrer, M., Vieta, E., Torrent, C., 2012. Subthreshold symptoms in bipolar disorder: impact on neurocognition, quality of life and disability. J. Affect. Disord. 136, 650–659. Bora, E., Yucel, M., Pantelis, C., 2009. Cognitive endophenotypes of bipolar disorder: a meta-analysis of neuropsychological deficits in euthymic patients and their first-degree relatives. J. Affect. Disord. 113, 1–20. Bora, E., Yucel, M., Pantelis, C., 2010. Cognitive impairment in schizophrenia and affective psychoses: implications for DSM-V criteria and beyond. Schizophr. Bull. 36, 36–42. Bora, E., Yucel, M., Pantelis, C., Berk, M., 2011. Meta-analytic review of neurocognition in bipolar II disorder. Acta Psychiatr. Scand. 123, 165–174. Bourne, C., Aydemir, O., Balanza-Martinez, V., Bora, E., Brissos, S., Cavanagh, J.T., Clark, L., Cubukcuoglu, Z., Dias, V.V., Dittmann, S., Ferrier, I.N., Fleck, D.E., Frangou, S., Gallagher, P., Jones, L., Kieseppa, T., Martinez-Aran, A., Melle, I., Moore, P.B., Mur, M., Pfennig, A., Raust, A., Senturk, V., Simonsen, C., Smith, D. J., Bio, D.S., Soeiro-de-Souza, M.G., Stoddart, S.D., Sundet, K., Szoke, A., Thompson, J.M., Torrent, C., Zalla, T., Craddock, N., Andreassen, O.A., Leboyer, M., Vieta, E., Bauer, M., Worhunsky, P.D., Tzagarakis, C., Rogers, R.D., Geddes, J.R., Goodwin, G.M., 2013. Neuropsychological testing of cognitive impairment in euthymic bipolar disorder: an individual patient data metaanalysis. Acta Psychiatr. Scand., 128; , 149–162. Brill, S., Penagaluri, P., Roberts, R.J., Gao, Y., El-Mallakh, R.S., 2011. Sleep disturbances in euthymic bipolar patients. Ann. Clin. Psychiatry: Off. J. Am. Acad. Clin. Psychiatr. 23, 113–116. Brissos, S., Dias, V.V., Soeiro-de-Souza, M.G., Balanza-Martinez, V., Kapczinski, F., 2011. The impact of a history of psychotic symptoms on cognitive function in euthymic bipolar patients: a comparison with schizophrenic patients and healthy controls. Rev. Bras. Psiquiatr. 33, 353–361.


10 Burdick, K.E., Goldberg, J.F., Harrow, M., 2010. Neurocognitive dysfunction and psychosocial outcome in patients with bipolar I disorder at 15-year follow-up. Acta Psychiatr. Scand. 122, 499–506. Chaves, O.C., Lombardo, L.E., Bearden, C.E., Woolsey, M.D., Martinez, D.M., Barrett, J.A., Miller, A.L., Velligan, D.I., Glahn, D.C., 2011. Association of clinical symptoms and neurocognitive performance in bipolar disorder: a longitudinal study. Bipolar Disord. 13, 118–123. Dias, V.V., Balanza-Martinez, V., Soeiro-de-Souza, M.G., Moreno, R. A., Figueira, M.L., Machado-Vieira, R., Vieta, E., 2012. Pharmacological approaches in bipolar disorders and the impact on cognition: a critical overview. Acta Psychiatr. Scand. 126, 315–331. Diniz, B.S., Machado-Vieira, R., Forlenza, O.V., 2013. Lithium and neuroprotection: translational evidence and implications for the treatment of neuropsychiatric disorders. Neuropsychiatr. Dis. Treat. 9, 493–500. Dittmann, S., Hennig-Fast, K., Gerber, S., Seemuller, F., Riedel, M., Emanuel Severus, W., Langosch, J., Engel, R.R., Moller, H.J., Grunze, H.C., 2008. Cognitive functioning in euthymic bipolar I and bipolar II patients. Bipolar Disord. 10, 877–887. Drummond, S.P., Gillin, J.C., Brown, G.G., 2001. Increased cerebral response during a divided attention task following sleep deprivation. J. Sleep Res. 10, 85–92. Durmer, J.S., Dinges, D.F., 2005. Neurocognitive consequences of sleep deprivation. Semin. Neurol. 25, 117–129. Giglio, L.M., Andreazza, A.C., Andersen, M., Cereser, K.M., Walz, J. C., Sterz, L., Kapczinski, F., 2009. Sleep in bipolar patients. Sleep Breath.=Schlaf Atm. 13, 169–173. Gilbert, E., Marwaha, S., 2013. Predictors of employment in bipolar disorder: a systematic review. J. Affect. Disord. 145, 156–164. Goswami, U., Sharma, A., Varma, A., Gulrajani, C., Ferrier, I.N., Young, A.H., Gallagher, P., Thompson, J.M., Moore, P.B., 2009. The neurocognitive performance of drug-free and medicated euthymic bipolar patients do not differ. Acta Psychiatr. Scand. 120, 456–463. Hall, R.C., 1995. Global assessment of functioning. A modified scale. Psychosomatics 36, 267–275. Hellvin, T., Sundet, K., Simonsen, C., Aminoff, S.R., Lagerberg, T. V., Andreassen, O.A., Melle, I., 2012. Neurocognitive functioning in patients recently diagnosed with bipolar disorder. Bipolar Disord. 14, 227–238. Iverson, G.L., Brooks, B.L., Langenecker, S.A., Young, A.H., 2011. Identifying a cognitive impairment subgroup in adults with mood disorders. J. Affect. Disord. 132, 360–367. Jamrozinski, K., Gruber, O., Kemmer, C., Falkai, P., Scherk, H., 2009. Neurocognitive functions in euthymic bipolar patients. Acta Psychiatr. Scand. 119, 365–374. Kaller, C.P., Unterrainer, J. M., Kaiser, S., Weisbrod, M., Aschenbrenner, S., 2011. Tower of London- Freiburger Version. Version 21. Schuhfried GmbH, Mödling. Klassen, L.J., Katzman, M.A., Chokka, P., 2010. Adult ADHD and its comorbidities, with a focus on bipolar disorder. J. Affect. Disord. 124, 1–8. Lehrl, S., Triebig, G., Fischer, B., 1995. Multiple choice vocabulary test MWT as a valid and short test to estimate premorbid intelligence. Acta Neurol. Scand. 91, 335–345. Levy, B., 2013. Autonomic nervous system arousal and cognitive functioning in bipolar disorder. Bipolar Disord. 15, 70–79. Levy, B., Medina, A.M., Weiss, R.D., 2013. Cognitive and psychosocial functioning in bipolar disorder with and without psychosis during early remission from an acute mood episode: a comparative longitudinal study. Compr. Psychiatry 54, 618–626. Lopez-Jaramillo, C., Lopera-Vasquez, J., Gallo, A., Ospina-Duque, J., Bell, V., Torrent, C., Martinez-Aran, A., Vieta, E., 2010. Effects of recurrence on the cognitive performance of patients

J. Volkert et al. with bipolar I disorder: implications for relapse prevention and treatment adherence. Bipolar Disord. 12, 557–567. Lupien, S.J., Maheu, F., Tu, M., Fiocco, A., Schramek, T.E., 2007. The effects of stress and stress hormones on human cognition: implications for the field of brain and cognition. Brain Cogn. 65, 209–237. Mann-Wrobel, M.C., Carreno, J.T., Dickinson, D., 2011. Metaanalysis of neuropsychological functioning in euthymic bipolar disorder: an update and investigation of moderator variables. Bipolar Disord. 13, 334–342. Martinez-Aran, A., Vieta, E., Reinares, M., Colom, F., Torrent, C., Sanchez-Moreno, J., Benabarre, A., Goikolea, J.M., Comes, M., Salamero, M., 2004. Cognitive function across manic or hypomanic, depressed, and euthymic states in bipolar disorder. Am. J. Psychiatry 161, 262–270. Martino, D.J., Strejilevich, S.A., Scapola, M., Igoa, A., Marengo, E., Ais, E.D., Perinot, L., 2008. Heterogeneity in cognitive functioning among patients with bipolar disorder. J. Affect. Disord. 109, 149–156. McKay, A.P., Tarbuck, A.F., Shapleske, J., McKenna, P.J., 1995. Neuropsychological function in manic-depressive psychosis. Evidence for persistent deficits in patients with chronic, severe illness. Br. J. Psychiatry: J. Ment. Sci. 167, 51–57. McKenna, B.S., Eyler, L.T., 2012. Overlapping prefrontal systems involved in cognitive and emotional processing in euthymic bipolar disorder and following sleep deprivation: a review of functional neuroimaging studies. Clin. Psychol. Rev. 32, 650–663. Millan, M.J., Agid, Y., Brune, M., Bullmore, E.T., Carter, C.S., Clayton, N.S., Connor, R., Davis, S., Deakin, B., DeRubeis, R.J., Dubois, B., Geyer, M.A., Goodwin, G.M., Gorwood, P., Jay, T.M., Joels, M., Mansuy, I.M., Meyer-Lindenberg, A., Murphy, D., Rolls, E., Saletu, B., Spedding, M., Sweeney, J., Whittington, M., Young, L.J., 2012. Cognitive dysfunction in psychiatric disorders: characteristics, causes and the quest for improved therapy. Nature reviews. Drug Discov. 11, 141–168. Montgomery, S.A., Asberg, M., 1979. A new depression scale designed to be sensitive to change. Br. J. Psychiatry: J. Ment. Sci. 134, 382–389. Morselli, P.L., Elgie, R., Cesana, B.M., 2004. GAMIAN-Europe/BEAM survey II: cross-national analysis of unemployment, family history, treatment satisfaction and impact of the bipolar disorder on life style. Bipolar Disord. 6, 487–497. Murray, G., Harvey, A., 2010. Circadian rhythms and sleep in bipolar disorder. Bipolar Disord. 12, 459–472. Niemann, H., Sturm, W., Thöne-Otto A. I. T., Wilmes, K., 2011. California Verbal Learning Test, Version 21. Schuhfried GmbH, Mödling. Palsson, E., Figueras, C., Johansson, A.G., Ekman, C.J., Hultman, B., Ostlind, J., Landen, M., 2013. Neurocognitive function in bipolar disorder: a comparison between bipolar I and II disorder and matched controls. BMC Psychiatry 13, 165. Puhr, U., Wagner, M., 2011. Interferenztest nach Stroop. Version 24Rev. 1. Schuhfried GmbH, Mödling. Reichenberg, A., Harvey, P.D., Bowie, C.R., Mojtabai, R., Rabinowitz, J., Heaton, R.K., Bromet, E., 2009. Neuropsychological function and dysfunction in schizophrenia and psychotic affective disorders. Schizophr. Bull. 35, 1022–1029. Reinares, M., Papachristou, E., Harvey, P., Mar Bonnin, C., SanchezMoreno, J., Torrent, C., Ayuso-Mateos, J.L., Ploubidis, G.B., Vieta, E., Frangou, S., 2013. Towards a clinical staging for bipolar disorder: defining patient subtypes based on functional outcome. J. Affect. Disord. 144, 65–71. Rucklidge, J.J., 2006. Impact of ADHD on the neurocognitive functioning of adolescents with bipolar disorder. Biol. Psychiatry 60, 921–928. Ryan, K.A., Vederman, A.C., McFadden, E.M., Weldon, A.L., Kamali, M., Langenecker, S.A., McInnis, M.G., 2012. Differential


Evidence for cognitive subgroups in bipolar disorder and the influence of subclinical depression and sleep disturbances executive functioning performance by phase of bipolar disorder. Bipolar Disord. 14, 527–536. Scott, J., 2011. Clinical parameters of circadian rhythms in affective disorders. Eur. Neuropsychopharmacol.: J. Eur. Coll. Neuropsychopharmacol. 21 (Suppl 4), S671–675. Sheehan, D.V., Lecrubier, Y., Sheehan, K.H., Amorim, P., Janavs, J., Weiller, E., Hergueta, T., Baker, R., Dunbar, G.C., 1998. The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J. Clin. Psychiatry 59 (Suppl 20), 22–33 (quiz 34-57). Svanborg, P., Asberg, M., 2001. A comparison between the Beck Depression Inventory (BDI) and the self-rating version of the Montgomery Asberg Depression Rating Scale (MADRS). J. Affect. Disord. 64, 203–216. Sylvia, L.G., Dupuy, J.M., Ostacher, M.J., Cowperthwait, C.M., Hay, A.C., Sachs, G.S., Nierenberg, A.A., Perlis, R.H., 2012. Sleep disturbance in euthymic bipolar patients. J. Psychopharmacol. 26, 1108–1112. Thompson, J.M., Gallagher, P., Hughes, J.H., Watson, S., Gray, J. M., Ferrier, I.N., Young, A.H., 2005. Neurocognitive impairment in euthymic patients with bipolar affective disorder. Br. J. Psychiatry: J. Ment. Sci. 186, 32–40. Torrent, C., Martinez-Aran, A., Daban, C., Amann, B., BalanzaMartinez, V., del Mar Bonnin, C., Cruz, N., Franco, C., TabaresSeisdedos, R., Vieta, E., 2011. Effects of atypical antipsychotics on neurocognition in euthymic bipolar patients. Compr. Psychiatry 52, 613–622. Torrent, C., Martinez-Aran, A., Daban, C., Sanchez-Moreno, J., Comes, M., Goikolea, J.M., Salamero, M., Vieta, E., 2006. Cognitive impairment in bipolar II disorder. Br. J. Psychiatry: J. Ment. Sci. 189, 254–259.

11

Torrent, C., Martinez-Aran, A., del Mar Bonnin, C., Reinares, M., Daban, C., Sole, B., Rosa, A.R., Tabares-Seisdedos, R., Popovic, D., Salamero, M., Vieta, E., 2012. Long-term outcome of cognitive impairment in bipolar disorder. J. Clin. Psychiatry 73, e899–905. Vieta, E., de Arce, R., Jimenez-Arriero, M.A., Rodriguez, A., Balanza, V., Cobaleda, S., 2010. Detection of subclinical depression in bipolar disorder: a cross-sectional, 4-month prospective follow-up study at community mental health services (SINDEPRES). J. Clin. Psychiatry 71, 1465–1474. Vieta, E., Sanchez-Moreno, J., Lahuerta, J., Zaragoza, S., 2008. Subsyndromal depressive symptoms in patients with bipolar and unipolar disorder during clinical remission. J. Affect. Disord. 107, 169–174. Vohringer, P.A., Barroilhet, S.A., Amerio, A., Reale, M.L., Alvear, K., Vergne, D., Ghaemi, S.N., 2013. Cognitive impairment in bipolar disorder and schizophrenia: a systematic review. Front. Psychiatry 4, 87. Watson, D., Clark, L.A., Tellegen, A., 1988. Development and validation of brief measures of positive and negative affect: the PANAS scales. J. Personal. Soc. Psychol. 54, 1063–1070. Wu, H.I., Chang, Y.H., Lai, C.C., Wu, J.Y., Chen, S.L., Chu, C.H., Lee, I.H., Yeh, T.L., Tzeng, N.S., Huang, S.Y., Yang, Y.K., Lu, R. B., 2011. The effect of comorbid anxiety disorder on neuropsychological function in bipolar II disorder. Prog. Neuropsychopharmacol. Biol. Psychiatry 35, 1841–1845. Young, R.C., Biggs, J.T., Ziegler, V.E., Meyer, D.A., 1978. A rating scale for mania: reliability, validity and sensitivity. Br. J. Psychiatry: J. Ment. Sci. 133, 429–435. Zimmermann, P., Fimm, B., 2011. Testbattery of attentional performance (Version 2.2). Psytest, Herzogenrath.


Suicide in bipolar disorder: characteristics and subgroups.

Polygenic risk scores in bipolar disorder subgroups.

Empirical evidence for discrete neurocognitive subgroups in bipolar disorder: clinical implications.

Electrophysiological evidence of a typical cognitive distortion in bipolar disorder.

Sleep disturbances and cognitive decline in the Northern Manhattan Study.

Interventions for Sleep Disturbance in Bipolar Disorder.

Evidence for the continuous latent structure of mania and depression in out-patients with bipolar disorder: results from the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD).

Hypoxemia, Sleep Disturbances, and Depression Correlated with Fatigue in Neuromyelitis Optica Spectrum Disorder.

The role of sleep in bipolar disorder.

Depression and Mania in Bipolar Disorder.

Evidence for separate inheritance of mania and depression challenges current concepts of bipolar mood disorder.

The relationships between migraine, depression, anxiety, stress, and sleep disturbances.

Associations between sleep disturbance, cognitive functioning and work disability in Bipolar Disorder.

Sleep disturbances after cerebral infarction: role of depression and fatigue.

Correlates of depression in bipolar disorder.

Voxel-Based Meta-Analytical Evidence of Structural Disconnectivity in Major Depression and Bipolar Disorder.

Sleep disturbances and mild cognitive impairment: A review.

The Effect of Sleep Disturbances on Major Depressive Disorder.

Effects of mood state on divided attention in patients with bipolar disorder: evidence for beneficial effects of subclinical manic symptoms.

Daytime sleepiness, sleep disturbance and functioning impairment in bipolar disorder.

Cognitive functions, fatigue, depression, anxiety, and sleep disturbances: assessment of nonmotor features in young patients with essential tremor.

Familiality and clinical outcomes of sleep disturbances in major depressive and bipolar disorders.

Cognitive emotion regulation in euthymic bipolar disorder.

Cognition as a target in schizophrenia, bipolar disorder and depression.