Journal of Affective Disorders 113 (2009) 1 20

www.elsevier.com/locate/jad

Review

Cognitive endophenotypes of bipolar disorder: A meta-analysis of

neuropsychological deficits in euthymic patients and their
first-degree relatives
Emre Bora a,, Murat Yucel a,b , Christos Pantelis a
a

Melbourne Neuropsychiatry Centre, Department of Psychiatry, The University of Melbourne and Melbourne Health, VIC. Australia
b
ORYGEN Research Centre, Melbourne, VIC. Australia
Received 24 April 2008; received in revised form 10 June 2008; accepted 10 June 2008
Available online 5 August 2008

Abstract
Background: Our aim was to delineate neuropsychological deficits related to genetic susceptibility, illness process and iatrogenic
factors in bipolar disorder (BD).
Methods: Following an extensive publication search on several databases, meta-analyses were conducted for 18 cognitive variables
in studies that compared performances of euthymic BD patients (45 studies; 1423 subjects) or first-degree relatives of BD patients
(17 studies; 443 subjects) with healthy controls. The effect of demographic variables and confounding factors like age of onset,
duration of illness and medication status were analysed using the method of meta-regression.
Results: While response inhibition, set shifting, executive function, verbal memory and sustained attention deficits were common
features for both patient (medium to large effect sizes) and relative groups (small to medium effect sizes), processing speed, visual
memory and verbal fluency deficits were only observed in patients. Medication effects contributed to psychomotor slowing in BD
patients. Earlier age of onset was associated with verbal memory impairment and psychomotor slowing.
Limitation: Data related to some confounding variables was not reported in a substantial number of extracted studies.
Conclusions: Response inhibition deficit, a potential marker of ventral prefrontal dysfunction, seems to be the most prominent
endophenotype of BD. The cognitive endophenotype of BD also appears to involve fronto-temporal and fronto-limbic related
cognitive impairments. Processing speed impairment is related, at least partly, to medication effects indicating the influence of
confounding factors rather than genetic susceptibility. Patterns of sustained attention and processing speed impairments differ from
schizophrenia. Future work in this area should differentiate cognitive deficits associated with disease genotype from impairments
related to other confounding factors.
2008 Elsevier B.V. All rights reserved.
Keywords: Bipolar disorder; Cognitive; Endophenotype; Memory; Executive function

Corresponding author. Melbourne Neuropsychiatry Centre, University of Melbourne, Alan Gilbert Building, NNF level 3, Carlton, VIC, 3053,
Australia.
E-mail address: emrebora@hotmail.com (E. Bora).
0165-0327/$ - see front matter 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.jad.2008.06.009

E. Bora et al. / Journal of Affective Disorders 113 (2009) 120

Contents
1.
2.

Introduction . . . . . . . . . . . . . . . . .
Method . . . . . . . . . . . . . . . . . . .
2.1. Neuropsychological variables . . . .
2.1.1. Verbal learning and memory
2.1.2. Visual memory . . . . . . .
2.1.3. Sustained attention . . . . .
2.1.4. Processing speed . . . . . .
2.1.5. Verbal fluency . . . . . . .
2.1.6. Set shifting . . . . . . . . .
2.1.7. Working memory . . . . . .
2.1.8. Response inhibition . . . . .
2.1.9. Visuospatial abilities . . . .
2.1.10. General intelligence . . . .
2.2. Statistical analyses . . . . . . . . . .
3. Results . . . . . . . . . . . . . . . . . . .
3.1. Remission . . . . . . . . . . . . . .
3.2. Relatives . . . . . . . . . . . . . . .
4. Discussion . . . . . . . . . . . . . . . . .
Role of funding source . . . . . . . . . . . . . .
Conflict of interest . . . . . . . . . . . . . . . .
Acknowledgements. . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . .

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

1. Introduction
Endophenotypes are intermediate phenotypes that
are considered a more promising index of underlying
genetic liability than the illness itself. To be accepted
as an endophenotype, intermediate phenotypes must
meet several criteria proposed by Gottesman and
Gould (2003). Endophenotypes should be associated
with illness, they should be heritable and they should
co-segregate within families with illness. There are
two additional conditions needed to meet criteria for
an endophenotype: (a) the endophenotype must be
state independent, it must be demonstrable in remitted
patients. (b) The endophenotypes should be more
frequent in unaffected relatives of patients compared
to the general population. In this context, while there
is convincing evidence regarding the value of
cognitive deficits as putative endophenotypes of
schizophrenia (Gur et al., 2007; Pantelis et al., in
press; Snitz et al., 2006), the value of such markers as
endophenotypes of bipolar disorder (BD) is a largely
understudied subject.
With respect to the first criterion (that markers are
state independent and observable in remitted patients),
the most consistent cognitive findings that may
represent potential endophenotypes within euthymic

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

2
3
3
3
3
8
8
8
8
8
8
8
8
11
12
12
15
15
17
17
17
18

patients with BD are verbal memory, executive function

and sustained attention deficits. Recently, three metaanalytic reports (Arts et al., 2008; Robinson et al., 2006;
Torres et al., 2007) provided further evidence for
cognitive impairment in BD. However, these findings
should be interpreted cautiously, as it is quite likely that
confounding factors such as medication, chronicity and
subthreshold affective symptoms are also contributing
to the observed findings. Furthermore, it is still not
known whether these findings are signs of multiple
independent cognitive impairments or whether they are
reflections of an underlying a single more basic
cognitive abnormality (for example, psychomotor
speed or working memory).
Regarding the second criterion (that markers are more
frequently observed in unaffected relatives of patients in
comparison to the general population), only a handful of
studies have investigated cognitive deficits of unaffected
relatives of affected patients. The findings of these studies
have been less consistent than those conducted in affected
patients themselves. For example, while several studies
suggest that verbal memory deficits are the most prominent
findings in relatives of BD patients (Gourovitch et al., 1999;
Keri et al., 2001), other studies do not support this notion
(Ferrier et al., 2004; Clark et al., 2005a,b). The evidence
regarding executive dysfunction appears to be similarly

E. Bora et al. / Journal of Affective Disorders 113 (2009) 120

inconsistent. For example, Frangou et al. (2005a,b)

suggested that executive functions attributable to the
ventral prefrontal cortex (VPFC) but not dorsal prefrontal
cortex (DPFC) are associated with genetic risk for BD,
however some other studies (Clark et al., 2005a,b) do not
support this proposal. One important limitation of relative
studies is sample size wherein studies are typically
characterised by small numbers.
Overall, while verbal memory, executive function
and sustained attention deficits are frequently reported,
the nature and magnitude of such impairments, as well
as their consistency can vary markedly across studies
due to differences in sample characteristics and
research methodologies. In this context, meta-analysis
is a useful tool for systematically combining all
research in this area to identify cognitive deficits
showing the most robust changes in BD. It is also a
useful methodology to workout the effect of confounding factors. In this way, we may be able to better
understand the pervasive cognitive disturbances that
can't be explained by the effects of iatrogenic factors or
by the neurotoxic effects of recurrent episodes, as well
as their neural underpinnings in bipolar disorder. To
date, only one meta-analytic study has analysed the
studies in first-degree relatives of BD. To our knowledge, the effects of clinical and iatrogenic confounders
have not been studied by meta-analytic methods
previously. Our aim was to investigate the possibility
that there exist cognitive endophenotypes of BD. To do
this we used published data in euthymic patients and
first-degree relatives.
2. Method
The relevant articles were searched using Pubmed,
Medline Web of Science and Psychinfo with the
following search terms: bipolar disorder (or manic
depress) and cognit, neuropsych, attention, memory, learning, executive. The search was limited to
studies published in peer-reviewed journals in English
available between 1995 and October 2007. Inclusion
criteria for studies were that they: (1) included
neuropsychological data pertaining to a remitted adult
BD patient group or first-degree relatives of patients
with BD; (2) included a healthy control group (3)
reported mean test scores and standard deviations (or
standard errors) of neuropsychological measures for
healthy controls and BD patients or their unaffected
relatives: (4) included at least one cognitive measure
that was studied in at least three studies in both BD
patients and unaffected relatives of BD patients.
Following initial publication search, the titles and

abstracts of articles were assessed for potential

suitability and references of these articles were also
crosschecked for further relevant articles. This step
identified 185 studies. When these articles were
examined further, only 68 of them met all four inclusion
criteria. Another 12 studies were excluded since they
were based on the same sample with other included
studies. Finally 56 studies that compared cognitive
performances of patients with BD (45 studies) or
relatives of BD patients (17 studies) with healthy
controls were included in the current meta-analysis
(Tables 1 and 2) (Altshuler et al., 2004; Antila et al.,
2007; Balanza-Martinez et al., 2005; Bora et al., 2007;
Bora et al., 2008; Brambilla et al., 2007; Cavanagh
et al., 2002; Christensen et al., 2006; Clark et al., 2002;
Clark et al., 2005a,b; Clark et al., 2005b; Deckersbach
et al., 2004a; Deckersbach et al., 2004b; Dittmann et al.,
2007; Dixon et al., 2004; El-Badri et al., 2001; Ferrier
et al., 1999; Ferrier et al., 2004; Fleck et al., 2003;
Frangou et al., 2005a; Frangou et al., 2005b; Goswami
et al., 2006; Gourovitch et al., 1999; Harmer et al.,
2002; Hawkins et al., 1997; Jones et al., 1994; Kaya
et al., 2007; Keri et al., 2001; Kerr et al., 2005;
Kieseppa et al., 2005; Klimes-Dougan et al., 2006;
Kolur et al., 2006; Krabbendam et al., 2000; Kremen
et al., 1998; Martinez-Aran et al., 2007; McIntosh et al.,
2005; Mur et al., 2007; Nehra et al., 2006; Paradiso
et al., 1997; Pirkola et al., 2005; Rocca et al., 2008;
Rossi et al., 2000; Schouws et al., 2007; Senturk et al.,
2007; Smith et al., 2006; Sobczak et al., 2003; Stoddart
et al., 2007; Swann et al., 2003; Szoke et al., 2006;
Thompson et al., 2005; Thompson et al., 2007; Van
Gorp et al., 1998; Van Gorp et al., 1999; Varga et al.,
2006; Zalla et al., 2004; Zubieta et al., 2001).
2.1. Neuropsychological variables
2.1.1. Verbal learning and memory
Effect sizes of 4 different measures of verbal memory
were included in the meta-analysis (Learning, immediate recall, delayed recall and recognition). These scores
were derived from the following tests: Rey Auditory
Verbal Learning Test (RAVLT) (Rey, 1964), California
Verbal Learning Test (CVLT) (Delis et al., 1987), Visual
Verbal Learning Test (VVT) (Lezak, 2005).
2.1.2. Visual memory
Rey Osterreich Complex Figure (ROCF) (Rey, 1941)
and WMS-R (Wechsler Memory Scale-Revised) Visual
memory (Wechsler, 1987) were used to assess visual
memory skills. For both tests, only delayed recall scores
were extracted.

E. Bora et al. / Journal of Affective Disorders 113 (2009) 120

Table 1
Studies with bipolar patients included in the meta-analysis
Study

Groups

Matched

Cognitive tests

da

Jones et al. (1994) (13)

26 BD
16 HC

Gender

Paradiso et al. (1997) (14)

11 BD
19 HC

Gender

Hawkins et al. (1997) (15)

22 BD
22 HC

Age, edu, gender

Van Gorp et al. (1998) (16)

13 BD
22 HC

Age, edu, IQ

Ferrier et al., (1999) (17)

41 BD
20 HC

Premorbid IQ
Age

Van Gorp et al. (1999) (18)

18 BD
20 HC
22 BD
22 HC

Age, edu, IQ

DSST
Stroop
CPT commission
CPT commission
Visual memory recall
DSST
Stroop
TMT-A
TMT-B
DSST
TMT-A
TMT-B
Fluency
WCST cat
WCST per
Stroop
TMT-A
TMT-B
Verbal learning
Immediate recall
Delayed recall
Visual copy
Visual memory recall
Fluency
DSST
TMT-A
TMT-B
CPT commission
Rey Learning
Digit span forwards
Digit Span backwards
Visual copy
Visual memory recall
CVLT recognition

0.47
0.46
0.50
0.55
0.47
0.65
0.55
0.69
0.19
0.82
0.54
0.78
0.11
1.01
0.95
0.08
0.32
0.24
0.96
0.70
0.52
0.09
0.25
0.67
0.59
0.55
0.86
0.21
0.68
0.21
0.70
0.53
0.77
0.17

66 HC
40 BD
29 BD
26 HC

Zubieta et al. (2001) (22)

15 BD

Age, education
Ethnicity, IQ
All patients has a history of psychotic episodes

Cavanagh et al. (2002) (23)

20 BD
20 HC

Age, gender, premorbid IQ

Fluency
Verbal learning
Delayed recall
Verbal recognition
Stroop
DSST
WCST cat
WCST per
Fluency
TMT-B
DSST
Fluency
Stroop
CPT commission
CPT commission
WCST cat
WCST per
Visual memory recall
Fluency
Stroop
Verbal learning
Delayed recall
Verbal recognition

0.54
0.94
0.94
0.50
0.67
1.12
0.82
0.69
0.42
0.80
0.73
0.77
1.12
0.97
1.41
0.84
1.52
0.42
0.31
0.61
1.06
0.96
0.62

Krabbendam et al. (2000) (19)

Rossi et al., 2000 (20)

El-Badri et al. (2001) (21)

Education, age

Age, IQ

E. Bora et al. / Journal of Affective Disorders 113 (2009) 120

Table 1 (continued)
Study

Groups

Matched

Cognitive tests

da

Clark et al. (2002) (24)

30 BD
30 HC

Gender, edu, IQ, age

Harmer et al. (2002) (25)

19 BD
19 HC
14 BD
40 HC

Age, edu, premorbid IQ

22 BD
35 HC
40 BD
22 HC

Deckersbach et al. (2004a,b) (29)

30 BD
30 HC

Education, age
Gender

Deckersbach et al. (2004a,b) (30)

25 BD
25 HC
15 BD
30 HC
37 BD
20 HC

Education, age
Gender
Gender, age

Balanza-Martinez et al. (2005) (33)

15 BD
26 HC

Gender, age

Clark et al. (2005a,b) (34)

15 BD
15 HC
44 BD
44 HC

IQ, age, gender

Verbal learning
Immediate recall
Delayed recall
Verbal recognition
CPT commission
CPT commission
CANTAB ED errors
CPT commission
CPT commission
Verbal learning
Immediate recall
Delayed recall
Verbal recognition
CPT commission
CPT commission
Fluency
WCST cat
WCST per
TMT-A
TMT-B
Stroop,
Verbal learning
Immediate recall
Delayed recall
Verbal recognition
Visual copy
Visual memory recall
Verbal learning
Immediate recall
Delayed recall
Verbal recognition
Visual copy
Visual memory recall
Fluency
Stroop
Stroop
TMT-A
TMT-B
WCST cat
WCST per
Fluency
DSST
Stroop
TMT-A
TMT-B
WCST cat
WCST per
CPT commission

0.74
0.44
0.16
0.29
0.96
0.18
0.71
1.01
0.04
1.25
1.01
0.77
0.00
0.71
0.22
0.16
0.89
0.77
0.39
0.40
0.41
0.91
0.75
0.78
0.22
0.30
0.57
2.03
1.40
1.67
0.64
0.06
0.70
0.17
0.72
1.17
0.61
0.83
0.44
0.72
1.28
1.05
1.62
0.68
0.89
1.48
1.67
1.00

15 BD
18 HC
26 BD
114 HC

Gender, premorbid IQ

Fluency
WCST cat
WCST per
Stroop
Visual memory recall
Stroop

0.88
0.25
0.38
0.57
0.60
1.78

Verbal learning
Delayed recall

0.43
0.60

Fleck et al., (2003) (26)

Swann et al. (2003) (27)

Altshuler et al. (2004) (28)

Dixon et al. (2004) (31)

Zalla et al. (2004) (32)

Frangou et al. (2005a,b) (35)

Kerr et al. (2005) (36)

Kieseppa et al. (2005) (37)

Gender, age, edu

Education, age
Gender

Gender, age

Gender, age premorbid IQ

Estimated IQ, age

(continued on next page)

E. Bora et al. / Journal of Affective Disorders 113 (2009) 120

Table 1 (continued)
Study

Groups

Matched

Kieseppa et al. (2005) (37)

McIntosh et al. (2005) (38)

47 BD
50 HC
22 BD
100 HC
63 BD
63 HC

Premorbid IQ

Goswami et al. (2006) (41)

37 BD
37 HC

Age, gender, education

Kolur et al. (2006) (42)

30 BD
30 HC

Education, age
Gender

Nehra et al. (2006) (43)

46 BD
20 HC

Gender,

Smith et al. (2006) (44)

21 BD
33 HC

Age, gender, epre IQ

Szoke et al. (2006) (45)

95 BD
48 HC

Age

Varga et al. (2006) (46)

19 BD
31 HC

Edu, gender

Pirkola et al. (2005) (39)

Thompson et al. (2005) (40)

Gender, edu, premorb IQ

Age, gender, IQ, edu

Cognitive tests

da

Digit span Backwards

DSST
Visual memory recall
Fluency
DSST
Digit span forwards
Digit span backwards
Fluency
TMT-A
TMT-B
DSST
Stroop
Digit span backwards
Digit span forwards
Verbal learning
Immediate recall
Delayed recall
Verbal recognition
CPT commission
CPT commission
TMT-A
TMT-B
Digit span backwards
Digit span forwards
Verbal learning
Immediate recall
Delayed recall
DSST
WCST cat
WCST per
Stroop
TMT-A
TMT-B
CPT commission
CPT commission
Fluency
TMT-A
TMT-B
WCST cat
WCST per
Verbal learning
Immediate recall
Delayed recall
Recog
Stroop
TMT-A
TMT-B
TMT-A
TMT-B
WCST per
TMT-A
TMT-B
Verbal learning
Immediate recall
Delayed recall
Stroop
WCST cat
WCST per
DSST

0.37
0.64
1.08
0.78
1.39
0.36
0.41
0.36
0.47
0.23
0.91
0.58
0.37
0.05
0.59
0.53
0.53
0.56
0.63
0.33
0.54
1.99
2.28
0.50
0.69
0.30
0.41
0.19
2.59
1.96
1.86
1.31
1.93
1.35
0.40
0.85
0.76
0.81
0.09
0.08
1.09
0.95
0.84
0.90
0.81
1.56
1.52
0.60
0.61
0.34
0.53
1.18
1.02
1.53
1.04
0.71
0.15
0.55
0.54

E. Bora et al. / Journal of Affective Disorders 113 (2009) 120

Table 1 (continued)
Study

Groups

Matched

Cognitive tests

da

Bora et al. (2007) (47)

65 BD
30 HC

Gender, age, edu

Brambilla et al. (2007) (48)

15 BD
26 HC
55 BD
17 HC
43 BD
22 HC

Gender, age

Martinez-Aran et al. (2007) (51)

77 BD
35 HC

Age, education, gender

Mur et al. (2007) (52)

44 BD
46 HC

Age, gender

Rocca et al. (2008) (53)

25 BD
31 BD

Age

Schouws et al. (2007) (54)

15 BD
15 HC

Edu, gender, age

Senturk et al. (2007) (55)

28 BD
29 HC

Edu, gender

Fluency
Stroop
TMT-A
TMT-B
CPT commission
CPT commission
WCST cat
WCST per
Verbal learning
Immediate recall
Delayed recall
Verbal recognition
CPT commission
CPT commission
TMT-A
TMT-B
AVLT learning
AVLT delayed
AVLT recog
Stroop
Fluency
WCST cat
WCST per
Verbal learning
Immediate recall
Delayed recall
Verbal recognition
Stroop
TMT-A
TMT-B
Digit Span forwards
Digit span backwards
Fluency
WCST cat
WCST per
Digit span forward
Digit span backwards
Stroop
Verbal learning
Immediate recall
Delayed recall
Verbal recognition
Visual memory recall
TMT-A
TMT-B
Fluency
WCST cat
WCST per
Stroop
Fluency
TMT-A
TMT-B
Stroop
Digit span forwards
Digit span backwards
Verbal learning
WCST cat
WCST per

0.78
0.73
0.68
0.84
0.67
0.58
0.62
0.57
0.57
0.73
0.63
0.54
1.14
0.33
0.45
0.50
1.01
1.30
0.19
0.27
0.39
0.28
0.56
0.67
0.55
0.88
0.56
0.57
0.90
0.60
0.77
0.94
0.91
0.95
0.73
0.78
0.97
0.90
0.49
0.43
0.65
0.48
0.23
0.97
1.05
0.87
0.08
0.04
0.01
1.28
0.64
0.65
1.08
0.31
0.52
1.30
0.58
0.67

Dittmann et al. (2007) (49)

Kaya et al. (2007) (50)

Age, edu, gender

Gender, age, edu

(continued on next page)

E. Bora et al. / Journal of Affective Disorders 113 (2009) 120

Table 1 (continued)
Study
Senturk et al. (2007) (55)
Stoddart et al. (2007) (56)

Thompson et al. (2007) (57)

Groups

Matched

Cognitive tests

22 BD
40 HC

Gender

50 BD
57 HC

IQ, edu, gender, age

DSST
Stroop
TMT-A
TMT-B
Fluency
Stroop
Digit span backwards
Digit span forwards

da
0.65
1.29
0.90
1.23
0.35
0.58
0.40
0.05

= Cohen d.

2.1.3. Sustained attention

To assess sustained attention continuous performance
tests (CPT) (Clark and Goodwin, 2004) were used.
Omission error and commission error scores of CPT
tasks were included in this study. Target sensitivity
indexes that are dependent on both omission and
commission errors were not included. Reaction time
measures were also not included since there was not
enough data for the relatives of BD patients.
2.1.4. Processing speed
Two different effect sizes were calculated to analyse
processing speed abilities.
Time to complete the part A of the Trail Making Test
(TMT-A) (Reitan, 1958) and the Digit Symbol Substitution test and symbol digit modalities test (DSST).
2.1.5. Verbal fluency
A measure of phonetic fluency (FAS) was included in
the current meta-analysis (Lezak, 1995). Category
fluency tasks were not included since there was no
sufficient study for the relatives of the patients with BD.
2.1.6. Set shifting
Set shifting is the ability to change the cognitive
strategies in response to change in the environment. To
assess the impairment in the set shifting abilities two
different tests was included into the current metaanalysis:
Trail Making TestPart B (TMT-B) (Reitan, 1958):
This test is a measure of set shifting and processing
speed.
Wisconsin Cart Sorting Test (WCST) (Heaton,
1981): Perseverative errors scores of this test were
used as a measure of set shifting. A measure of
Cambridge Neuropsychological Test Automated Battery
(CANTAB) (Downes et al., 1989) extradimensional/
intradimensional task (extradimensional shifting errors)

was also involved with this task. Number of categories

achieved scores of WCST was involved as a measure of
rule discovery.
2.1.7. Working memory
As a measure of working memory Backwards digit
span of the WAIS-R Digit Span (Wechsler, 1987) was
used.
2.1.8. Response inhibition
Response inhibition refers to the suppression of
actions that are inappropriate in a given context. In this
meta-analysis, interference score (time to completion) of
the Stroop Colour-Word test (Lezak, 1995) was used to
assess response inhibition deficits.
2.1.9. Visuospatial abilities
ROCF copy score (Rey, 1941) was included for
analysing visuospatial abilities.
2.1.10. General intelligence
WAIS-R (Wechsler Adult Intelligence ScaleRevised) (Wechsler, 1981) full Scale IQ and its shorter
versions were used to assess current IQ abilities. For
analysing premorbid IQ, effect sizes of the NART
(National Adult Reading Test) (Nelson, 1982) and the
WAIS Vocabulary subtask (premorbid IQ) were
included.
For the purpose of the study, tasks measuring similar
constructs were assessed together. For example, we
combined the following: (a) RAVLT, CVLT and VVT,
(b) WCST perseverative errors and CANTAB extradimensional error scores, (c) Digit Symbol Substitution
test and symbol digit modalities test, (d) ROCF and
WMS visual memory. Identical scores (omission and
commission error scores) from various different versions of sustained attention tasks were also included
together. Since different studies reported different

E. Bora et al. / Journal of Affective Disorders 113 (2009) 120

Table 2
Studies with relatives of bipolar patients included in the meta-analysis
Study

Groups

Matched

Cognitive tests

da

Kremen et al. (1998)

15 BD
44 HC

Gender, age
edu

Gourovitch et al. (1999)

7 BD
15 HC

Keri et al. (2001)

20 BD
20 HC

Sobczak et al. (2003)

22 BD
15 HC

Ferrier et al. (2004)

17 BD
17 HC

Gender, age, edu

Premorbid IQ

Zalla et al. (2004)

33 BD
20 HC

Gender, age

Clark et al. (2005a)

27 BD
46 HC

Gender, age, edu

Clark et al. (2005b)

27 BD
47 HC
15 BD
43 HC
19 BD

Gender, age, edu

DSST
Stroop
TMT-A
TMT-B
WCST cat
WCST per
Visual copy
Visual memory recall
Fluency
TMT-A
TMT-B
Verbal learning
Immediate recall
Delayed recall
Verbal recognition
WCST cat
WCST per
Dig span forwards
Digit Span backwards
CPT commission
Visual copy
Visual memory recall
Fluency
WCST cat
WCST per
Digit span forwards
Digit span backwards
Fluency
Stroop
Verbal learning
Delayed recall
Verbal recognition
Fluency
DSST
Stroop
TMT-A
TMT-B
Verbal learning
Digit span forwards
Digit span backwards
Verbal recognition
CPT commission
CPT commission
Stroop
TMT-A
TMT-B
WCST cat
WCST per
Verbal learning
Immediate recall
Delayed recall
CANTAB IDED
CPT commission

0.05
0.42
0.28
0.11
0.45
0.09
0.24
0.34
0.28
0.10
0.01
0.73
0.32
1.15
0.92
0.00
0.52
1.16
0.97
0.32
0.04
0.68
0.12
0.11
0.10
0.33
0.18
0.25
0.46
0.25
0.34
0.09
0.12
0.24
0.00
0.07
0.32
0.18
0.40
0.99
0.29
0.44
0.06
0.96
0.31
0.60
0.12
0.57
0.43
0.20
0.12
0.77
0.38

WCST cat
WCST per
DSST

0.53
0.40
0.12

Frangou et al. (2005a)

Kieseppa et al. (2005)

Gender, age, edu

IQ

(continued on next page)

10

E. Bora et al. / Journal of Affective Disorders 113 (2009) 120

Table 2 (continued)
Study

Groups

Kieseppa et al. (2005)

114 HC

McIntosh et al. (2005)

24 BD
50 HC
16 BD
100 HC
40 BD
55 HC

Gender, age

21
88
43
50

BD
HC
BD
HC

Age

Szoke et al. (2006)

63 BD
48 HC

Age

Bora et al. (2008)

34 BD
25 HC

Gender, age, edu

Premorbid IQ

Pirkola et al. (2005)

Antila et al. (2007)

Christensen et al. (2006)

Klimes-Dougan et al. (2006)

Matched

Age ??
Gender, age
Premorbid IQ

Age

Cognitive tests

da

Verbal learning
Delayed recall
Visual memory recall
Visual copy
Digit span backwards
Fluency
DSST
Digit span forwards
Digit span backwards
DSST
TMT-A
TMT-B
Verbal learning
Immediate recall
Delayed recall
Verbal recognition
Digit span forwards
Digit span backwards
Stroop
TMT-A
TMT-A
TMT-B
Verbal learning
Immediate recall
Delayed recall
WCST cat
WCST per
CPT commission
CPT commission
TMT-A
TMT-B
WCST per
Stroop
TMT-A
TMT-B
Verbal learning
Immediate recall
Delayed recall
Verbal recognition
WCST cat
WCST per
Digit span forwards
Digit span backwards
CPT commission
CPT commission

0.13
0.08
0.24
0.04
0.18
0.58
0.50
0.80
0.31
0.44
0.27
0.26
0.13
0.19
0.09
0.29
0.05
0.17
0.40
0.00
0.24
0.35
0.39
0.62
0.62
0.60
0.56
0.24
0.14
0.32
0.50
0.22
0.72
0.15
0.72
0.26
0.27
0.04
0.33
0.68
0.69
0.37
0.56
0.50
0.39

= Cohen d.

scoring systems for the Stroop task, identical scores that

are sensitive to response inhibition were included
together.
In some studies, means and standard deviations
(SDs) of more than one group with euthymic BD
(Ferrier et al., 1999; Nehra et al., 2006; Senturk et al.,
2007) or unaffected BD relatives (Christensen et al.,
2006) were reported. In these studies, the mean values

and SDs are combined. However, in another study that

reported scores from two different groups (Van Gorp
et al., 1998), only patients without comorbid alcohol
dependency were included in the current meta-analysis.
If there were more than one publication from the
common samples, only the data from the study with the
larger sample was included, unless results were reported
for different cognitive tasks.

E. Bora et al. / Journal of Affective Disorders 113 (2009) 120

The current study reports the results of meta-analyses

for seventeen neurocognitive variables in 45 euthymic
BD (1423 BD patients1524 healthy controls) and 17
BD relative studies (443 relatives, 797 healthy controls)
Mean effect sizes for current IQ, premorbid IQ and
education were also calculated since these variables can
significantly influence the magnitude of group
differences.
2.2. Statistical analyses
Meta-analyses were conducted with MIX software
(Bax et al., 2006). We used the standardised mean
difference method with Hedge's correction for bias in
small samples. Whenever BD patients and their relatives
performed poorer than controls, we reported betweengroup differences by positive effect sizes. Therefore, the
effect sizes for the relevant variables were multiplied by
minus one. Homogeneity of the resulting mean
weighted effect sizes was tested with Q test. Since
there was heterogeneity for many of the analyses, we
used a random effects model rather than a fixed effects
model for the meta-analyses.
Meta-analytic methods accept published studies as a
representative of all valid studies undertaken. However,
direction of results may influence the chance of
submission and publication of the studies and this fact
can be a source of bias in results of meta-analyses
(publication bias). Studies with negative outcomes
(especially when the sample size is small) are less

11

likely to be published. In the current meta-analysis,

publication bias was tested with funnel plot and Egger's
test. However, Egger test may give false positive results,
especially when effect sizes distributed heterogeneously.
To reduce the risk of false positive results and to further
investigate the source of funnel plot asymmetry, tasks
with a significant asymmetry (Egger's test, p b 0.05)
were further analysed. The individual characteristics of
the studies were further investigated, a Fail Safe number
(number of negative studies necessary to make the
group difference insignificant) was calculated and trim
and fill method was used to estimate the actual effect
size. A significance level of p b 0.05 was used for the
random effects model, homogeneity and publication
bias analyses.
The effects of demographic variables, medication
(percentage of patients using antipsychotics, antidepressants, and lithium), clinical variables (age of onset and
duration of illness, number of manic and depressive
episodes, Hamilton depression score), between-group
differences of IQ and other cognitive skills were
analysed with meta-regression. One difficulty in performing meta-regression analyses was the limited data
for clinical and treatment variables. Therefore, to
increase the number of studies three combined scores
for psychomotor speed (TMT-A, DSST), executive
function (WCST perseverations, Stroop Interference
score, TMT-B) and memory recall (delayed verbal
memory, ROCF delayed) were also calculated and used
for meta-regression analyses. Meta-regression analyses

Table 3
Mean weighted effect sizes for individual tasks and education for patient-control differences
Test

Study

Bipolar

Control

95% CI

Q-test p

Bias

TMT-B
Verbal learning
CPT ommission
Delayed recall
Stroop
DSST
Digit span backwards
Immediate recall
WCST per
TMT-A
WCST Cat
FAS
Visual memory recall
Verbal recognition
Current IQ
Digit span forwards
CPT commission
Visual copy
IQ premorbid
Education

21
18
10
17
24
13
9
12
17
20
15
19
9
13
7
8
9
4
23
32

793
619
303
578
746
381
375
453
663
768
538
681
274
488
239
349
288
119
714
1017

626
632
279
612
707
479
487
419
543
600
465
594
424
411
218
373
264
89
792
1046

0.86
0.85
0.83
0.77
0.76
0.75
0.75
0.73
0.70
0.69
0.66
0.60
0.59
0.44
0.40
0.37
0.36
0.23
0.17
0.01

0.651.06
0.681.01
0.661.00
0.610.93
0.590.93
0.570.94
0.411.01
0.530.93
0.490.91
0.570.82
0.360.96
0.450.74
0.400.78
0.310.58
0.010.80
0.150.60
0.130.59
0.050.51
0.020.36
0.130.16

8.20
10.1
9.42
9.34
8.68
7.98
4.29
7.15
6.54
11.09
4.33
7.95
6.02
6.33
1.95
3.21
3.09
1.61
1.73
0.14

b0.0001
b0.0001
b0.0001
b0.0001
b0.0001
b0.0001
b0.0001
b0.0001
b0.0001
b0.0001
b0.0001
b0.0001
b0.0001
b0.0001
0.05
0.001
0.002
0.11
0.08
0.89

b0.001
0.03
0.55
0.06
0.0004
0.1
b0.001
0.04
0.0001
0.31
b0.0001
0.07
0.31
0.46
0.0003
0.06
0.1
0.49
b0.0001
b0.0001

0.13
0.0004
0.10
0.07
0.07
0.75
0.21
0.02
0.02
0.59
0.15
0.50
0.66
0.13
0.79
0.62
0.43
0.32
0.20
0.04

12

E. Bora et al. / Journal of Affective Disorders 113 (2009) 120

Fig. 1.

were conducted in SPSS 11.0 by using the macros

written by David B. Wilson. This procedure allows the
performance of weighted generalized least squares
regression. Meta-regression analyses were performed
with the random effects model using restricted-information maximum likelihood method with a significance
level of p b 0.05.
3. Results
3.1. Remission
The meta-analysis for euthymic BD patients included
45 studies. These studies compared cognitive performance of a total of 1446 patients and 1524 healthy
controls. There were no significant differences for age
(reported in 44 studies) and gender composition
(reported in 43 studies) between patients (mean
age = 38.8, percentage of males = 48.8%) and controls
(mean age = 38.3, percentage of males = 49.9%). There

were no significant between-group differences for

education and premorbid IQ (Table 3). Current IQ had
a tendency to be lower in patients. There was a
significant level of heterogeneity for current and
premorbid IQ analyses.
In 17 of 18 meta-analyses conducted for each
cognitive test, BD patients performed significantly
worse than control subjects (Table 3). Medium or
large effect sizes were noted in most measures of
executive functions, verbal memory, sustained attention
and psychomotor speed. However, effect sizes for visual
memory, verbal recognition memory, CPT commission
errors and digits forward were small. There was no
between-group difference on the visual copying task.
Five of the 18 analyses reported a significant degree
of heterogeneity (Trails B, Digit Span-backwards,
Stroop, WCST category, WCST perseveration). Most
of the heterogeneity in these studies was explained by
several studies. The studies of Goswami et al. (2006),
Kolur et al. (2006), Smith et al. (2006) were responsible

E. Bora et al. / Journal of Affective Disorders 113 (2009) 120

13

Fig. 2.

for heterogeneity in the meta-analysis of TMT-B. The

study of Goswami et al. (2006) was also the cause of the
heterogeneity of Digit Span-backwards. In the case of
the Stroop test, (Fig. 1) extreme positive effect sizes of
Kerr et al. (2005), Balanza-Martinez et al. (2005), Kolur
et al. (2006) and negative effect sizes of Rocca et al.

(2008) were responsible for the heterogeneity. Finally,

the heterogeneity in the analysis of WCST category and
perseveration scores (Fig. 2) was mostly due to the
studies of Balanza-Martinez et al. (2005) and Kolur
et al. (2006). After excluding all of these studies that
caused the heterogeneity the findings, Q-tests for all of

Table 4
Mean weighted effect sizes for individual tasks and education for relative-control differences
Test

Study

Bipolar relatives

Control relatives

95% CI

Q-test p

Bias

Stroop
TMT-B
WCST per
CPT ommission
Immediate recall
Learning
FAS
Delayed recall
WCST cat
DSST
Digit Span Backwards
Memory recognition
Current IQ
CPT commission
Edu
TMT-A
Visual memory recall
Digit span forwards
Premorbid IQ
Visual copy

6
8
9
5
5
8
5
7
7
5
7
5
7
3
11
9
3
6
8
3

142
252
257
128
151
209
90
192
167
115
153
120
157
94
269
277
41
134
165
41

209
274
312
153
192
338
117
321
217
280
346
127
242
92
576
358
173
232
441
173

0.51
0.38
0.36
0.36
0.33
0.28
0.27
0.27
0.24
0.22
0.22
0.20
0.20
0.18
0.18
0.17
0.13
0.08
0.03
0.1

0.270.76
0.200.55
0.200.54
0.12060
0.110.55
0.090.46
0.010.55
0.040.50
0.080.56
0.040.49
0.140.57
0.110.51
0.240.63
0.110.47
0.050.42
00.33
0.390.65
0.380.54
0.290.23
0.440.25

4.1
4.15
4.18
2.93
2.94
2.97
1.85
2.27
1.48
1.69
1.21
1.27
0.89
1.21
1.52
1.97
0.48
0.32
0.23
0.56

b0.0001
b0.0001
b0.0001
0.003
0.003
0.003
0.06
0.02
0.14
0.09
0.23
0.20
0.37
0.23
0.13
0.05
0.63
0.75
0.83
0.58

0.37
0.52
0.08
0.95
0.62
0.92
0.56
0.21
0.05
0.28
0.02
0.25
0.0006
0.55
0.02
0.82
0.14
0.003
0.07
0.84

0.60
0.52
0.71
0.53
0.86
0.38
0.34
0.32
0.22
0.52
0.24
0.96
0.36
0.83
0.11
0.07
0.74
0.44
0.45
0.84

14

E. Bora et al. / Journal of Affective Disorders 113 (2009) 120

Fig. 3.

number of omission errors (B = 0.029, SE = 0.01,

p = 0.02). IQ difference between-groups was only
associated with the effect size of the Stroop (B = 0.45,
SE = 0.16, p = 0.007, 10 studies). Younger age of illness
onset was associated with larger effect sizes for verbal
learning (B = 0.05, SE = 0.02, p = 0.027, 15 studies) and
TMT-A (B = 0.07, SE = 0.02, p = 0.0014, 14 studies).
Medication was associated with the magnitude of
impairment for psychomotor speed and sustained
attention. Studies that had reported a higher percentage
of antipsychotic usage found larger effect size impairments for psychomotor speed (B = 0.05, SE = 0.02,
p = 0.04, 24 studies) and omission errors (B = 0.011,
SE = 0.005, p = 0.037, 9 studies). Antidepressant use was
also associated with psychomotor speed (B = 0.0107,
SE = 0.004, p = 0.0039, 17 studies) and TMT-A performance (B = 0.01, SE = 0.004, p = 0.01, 12 studies).
Psychomotor slowness also increased the effect size
of the impairment for WCST (B = 1.1, SE = 0.43,
p = 0.009, 13 studies) and Stroop (B = 0.66, SE = 0.27,
p = 0.015, 17 studies). Low performance on TMT-A
was also associated with larger effect sizes for the

these five tasks were non-significant (p N 0.15). Some

differences in the characteristics of these studies may
explain these results. In the study of Balanza-Martinez
et al. (2005), the patients had lower education and
premorbid IQ, while in the study of Rocca et al. (2008),
the patients had a significantly higher IQ. Further, the
study of Smith et al. (2006) was characterized by early
onset (b 15 years of age).
Egger's test for the meta-analyses for three tasks
showed a significant publication bias (Verbal learning
and verbal memory early recall and WCST perseveration). The publication bias was especially significant for
verbal learning score. Fail Safe number for the verbal
learning was 836 studies and trim and fill method
predicted a medium effect size of D = 0.66 (CI = 0.480.85) instead of a large effect size suggested by the prior
analysis. Fail-safe numbers for verbal memory early
recall and WCST perseveration score were 297 and 513,
respectively. Trim and fill method did not predict a
different effect size for these tasks.
Meta-regression analyses revealed that mean age of
BD patients was negatively associated with the increased

Fig. 4.

E. Bora et al. / Journal of Affective Disorders 113 (2009) 120

Stroop (B = 0.66, SE = 0.30, p = 0.025, 14 studies),

omission errors (B = 0.88, SE = 0.4, p = 0.03) and FAS
(B = 0.98, SE = 0.45, p = 0.03, 10 studies).
There was no association between verbal memory
and executive function impairments. Patients who made
more omission errors performed more poorly on the
Stroop task (B = 1.69, SE = 0.41, p b 0.001). Working
memory impairment (reverse digit span) was also
associated with the magnitude of executive dysfunction
(B = 0.76, SE = 0.15, p b 0.001, 7 studies).
3.2. Relatives
Meta-analyses of relatives' studies included 17
studies (443 relatives of BD patients and 797 healthy
controls). The mean age (15 studies) and gender
compositions (16 studies) of relatives (38.5 years,
37.7% male) and healthy controls (41.4 years, 43.6%
male) were comparable. There were no significant group
differences for education, current and premorbid IQ
between groups (Table 4).
Q-test revealed a significant heterogeneity for current
IQ. In 6 of 18 cognitive measures, relatives of BD
patients performed significantly poorer than controls
(see Table 4). The greatest impairment was found on the
Stroop task (medium effect size) (Fig. 3). The effect
sizes for the impairments in TMT-B, WCST perseveration (Fig. 4), CPT omission, verbal learning and
immediate recall were small.
There was a significant heterogeneity for only one
task (Digit Span-forwards). Positive effect size in the
study of Gourovitch et al. (1999) and good performance
of the relatives in the study of Pirkola et al. (2005) were
responsible for this heterogeneity. The study of
Gourovitch had an extremely small sample size and
included only monozygotic twins. None of the analyses
in relatives showed a significant publication bias.
Meta-regression analyses revealed effects of
between-group IQ differences and the mean age of
first-degree relatives on some cognitive tasks. Age had a
significant effect on relative-control differences of
psychomotor speed (B = 0.022, SE = 0.007, p = 0.003,
10 studies) and verbal memory delayed recall (B =
0.016, SE = 0.008, p = 0.03, 5 studies). Thus, the
studies with older samples reported smaller effect
sizes. Studies that reported lower IQ scores in relatives
(compared to controls) also found larger effect sizes for
executive function (B = 0.57, SE = 0.24, p = 0.022, 8
studies), psychomotor speed (B = 0.559, SE = 0.218,
p = 0.01, 8 studies), verbal delayed recall (B = 0.68,
SE = 0.30, p = 0.02, 5 studies), Digit Span-backwards
(B = 1.41, SE = 0.47, p = 0.003, 6 studies) and Digit

15

Span-forwards (B = 1.71, SE = 0.56, p = 0.0025, 5

studies).
4. Discussion
This meta-analytic study demonstrated that impaired
response inhibition might be the most prominent
cognitive endophenotype of BD. Another executive
measure, set shifting and two other cognitive domains,
verbal memory and sustained attention also met the
criteria as potential endophenotypes of BD. While
impairments in processing speed, verbal working
memory and visual memory are related to the clinical
expression of BD, they were not observed in relatives
and therefore do not seem to be associated with genetic
susceptibility to BD. Processing speed impairments may
be partly secondary to medication and can also
contribute to other cognitive impairments found in
euthymic patients with BD. Early onset of illness may
be associated with more severe verbal memory impairment and psychomotor slowing in BD. The observed
pattern of sustained attention impairment and prominence of response inhibition deficit and lack of
impairment in processing speed in relatives of patients
with BD partly contrast with reported findings of studies
in first-degree relatives of schizophrenia.
Response inhibition seems to be the most significant
endophenotype of BD. In previous studies in BD, in
addition to the Stroop test, impaired response inhibition
was also reported with the Hayling Sentence Completion
Task both in euthymic patients and relatives (Frangou
et al., 2005a). However, we did not include this task in our
analyses, since fewer than three published studies have
reported this measure in relatives of patients with BD. Our
results are partly consistent with Frangou et al. (2005a,b)
who suggested that only VPFC related functions are
endophenotypes of BD. While it may be oversimplistic to
equate response inhibition with VPFC and Cingulate
function, brain imaging studies provided evidence
regarding differential role of VPFC and dorsal prefrontal
cortex for response inhibition (Blumberg et al., 2003) and
set shifting (Monchi et al., 2001) respectively. Anterior
Cingulate gyrus and VPFC abnormalities may have a role
in the aetiology of BD. However, unlike Frangou et al.
(2005a,b), current results also suggest a role for dorsal
prefrontal cortex related set shifting abilities as cognitive
endophenotypes of BD. We found a small but significant
impairment for TMT-B and WCST perseverative errors in
relatives of patients with BD. In a recent meta-analysis,
Arts et al. (2008) found impairments in TMT-B but not in
WCST perseverative errors in relatives of BD patients.
This difference may be related to their lower sample size

16

E. Bora et al. / Journal of Affective Disorders 113 (2009) 120

for this aspect of their meta-analysis. The current study

provides evidence for a selective role for executive
functions as endophenotypes of BD. Thus, while ability
on tasks of set shifting and response inhibition seemed to
be more related to genetic risk for BD, other EF functions
like working memory and verbal fluency were not.
In our study, verbal memory was also impaired both
in euthymic patients and their relatives. While a
relatively large effect size for verbal memory was
found in euthymic patients, the effect sizes for verbal
memory in relatives were modest. This result partly
contradicts the findings of Arts et al. (2008) who
reported that relatives had the largest impairment in
verbal memory. Publication bias seems to exaggerate the
actual impairment for verbal memory, especially verbal
learning, in euthymic patients with BD. Originally, we
also included visual memory skills in our analyses. The
results of the study do not suggest a role for nonverbal
learning abilities as endophenotypes of BD.
Originally, the current meta-analysis also provided
support for the potential role of sustained attention as an
endophenotype of BD. As far as we know, this is the first
meta-analysis that has examined sustained attention in
relatives of BD patients. Both the euthymic BD patients
and relatives made more omission errors on CPT tasks.
Failure to detect targets seems to be a possible trait
marker for BD. We found larger effect size impairment
for CPT in euthymic patients compared to meta-analyses
of Robinson et al. (2006) and Arts et al. (2008). The
different outcome seems to be related to the measures of
sustained attention examined by these other authors.
These studies analysed the measures of sensitivity index
of sustained attention and latency. Sensitivity is a derived
score from correct target detection percentage and false
alarm rates. This measure depends on not only omission
errors but also commission errors that do not seem to be
increased in euthymia (Bora et al., 2006). The selective
impairment of target detection in BD differs from the
pattern observed in schizophrenia (see below).
While the effect sizes for impairment on psychomotor
tasks were relatively large in patients with BD, psychomotor processing seems to be intact in first-degree
relatives of BD. These results suggest that other
confounding factors rather than genetic susceptibility
may be the source of psychomotor slowness of BD
patients. According to our results, treatment effects may
be partly responsible for this finding in euthymic patients.
Antipsychotic use was associated with psychomotor
slowness. There is some previous evidence regarding
negative impact of typical and atypical antipsychotics on
psychomotor abilities (Hughes et al., 1999; Morrens et al.,
2007). Antipsychotics were also associated with increased

magnitude of impairment for sustained attention. Since

psychomotor slowness was related to larger effect sizes
for verbal fluency, sustained attention and WCST
perseverative errors, antipsychotics may also have an
indirect impact on other cognitive functions.
The proposed cognitive endophenotypes of bipolar
disorder partly differ from schizophrenia. While metaanalytic studies in relatives of schizophrenia patients
revealed psychomotor slowing and verbal fluency as an
important endophenotype of schizophrenia (Sitskoorn
et al., 2004; Snitz et al., 2006; Szke et al., 2005), this
was not the case for BD in the current study. Metaanalyses of the Stroop test in relatives of schizophrenia
patients (Sitskoorn et al., 2004; Snitz et al., 2006)
reported a milder deficit than in BD patients in the
current study, despite the fact that they reported more
pronounced general intellectual deficits compared to BD
relative studies. Unlike in BD, response inhibition
deficit is not the most pronounced impairment in
relatives of patients with schizophrenia. The observed
pattern of sustained attention is also different in
schizophrenia and BD. Meta-analyses in relatives of
schizophrenia patients provided evidence for all aspects
of sustained attention but especially for false alarming
and target sensitivity (Sitskoorn et al., 2004). In contrast,
target detection impairment rather than false alarming
has a role as an endophenotype of BD. While response
inhibition and a selective type of sustained attention
deficit are more specific endophenotypes of BD,
processing speed and general intelligence impairments
may be endophenotypes of schizophrenia. However,
there is also evidence for shared endophenotypes in BD
and schizophrenia. Verbal memory and set shifting
impairments are observed in relatives of both patient
groups. This finding may be compatible with brain
imaging findings, which suggest there are shared frontolimbic and fronto-subcortical deficits in schizophrenia
as well as BD (McIntosh et al., 2006). Heterogeneity of
schizophrenia and BD may also contribute to shared and
different endophenotypes of BD. While verbal memory
and set-shifting abnormalities may be trait markers of
only BD patients with a history of psychosis, response
inhibition deficits may be an endophenotype for all
patients with BD (Bora et al., 2005; Bora et al., 2007;
Martinez-Aran et al., 2008). This may also explain the
larger effect sizes for response inhibition in relatives of
patients with BD. Verbal memory and set shifting
impairments may be endophenotypes of psychosis
independent of diagnosis.
Not all of the cognitive impairments in euthymic
patients with BD are true endophenotypes, even though
they are not secondary to iatrogenic effects or

E. Bora et al. / Journal of Affective Disorders 113 (2009) 120

subsyndromal symptoms. Since endophenotypes must be

stable over time, progressive impairments related to
disease progression may contribute to the cognitive
profile of established BD patients. While longitudinal
studies are very rare in BD, there is some evidence of
progression of cognitive impairments in schizophrenia
studies. Late maturational changes that may start before
the onset of illness and continue after the first episode may
contribute to observed neurocognitive pattern in major
psychoses (Pantelis et al., 2005). Wood and colleagues
recently examined progressive changes in cognitive
function over the transition to psychosis as part of the
Melbourne UHR studies (Wood et al., 2007). While
performance on most tests was stable or improved,
visuospatial memory, verbal fluency and attention switching showed significant decline over the transition to
psychosis. These progressive impairments were not seen
in the non-psychotic UHR group. These data would seem
consistent with progressive brain structural changes over
transition to psychosis (Pantelis et al., 2005; Pantelis et al.,
2007). Unfortunately, there is very scarce data regarding
longitudinal studies in BD. However, it is interesting that
working memory and verbal fluency are among the
impairments which are only observed in patients with BD.
Considering the overlaps between schizophrenia and BD,
it is likely that late-maturational changes can contribute to
the cognitive profile of BD. Another proposed mechanism
for illness related impairments in BD is the potential
neurotoxic effects of repeated illness episodes on limbic
structures. Thus, there is some evidence for the association between the number of manic episodes, duration of
illness and cognitive impairment in BD (Robinson and
Ferrier, 2006). While our meta-regression analyses failed
to support evidence for this association, some methodological factors including the limitations of the metaregression approach, and factors related to sample
selection in published studies may explain this outcome.
Analysis of direct correlations from the individual studies
would be a better option, however insufficiencies of
published data prevent us from performing a correlational
meta-analysis. Long term follow-up studies that investigate cognitive functions in high-risk groups and patients
with established diagnosis are necessary to tackle the preonset and post-onset cognitive changes in BD.
Differences between underlying disease-severities of
patients included in different studies may be another
confounding factor. Consistent with this idea, metaregression analysis demonstrated an association between
young onset, verbal memory and psychomotor slowness.
A subgroup of BD patients may present with earlier onset
and more pronounced impairments in verbal memory
and processing speed. A similar pattern was previously

17

observed in schizophrenia studies, with more severe

impairment for verbal memory reported for early-onset
schizophrenia (Tuulio-Henriksson et al., 2004). Residual
mood symptoms also can have an impact on heterogeneity of analyses and can increase the magnitude of
effect sizes in euthymic patients. This issue was simply
not investigated in many studies and in others very
different measures were used to assess residual symptoms. We were only able to undertake an analysis on a
sample of studies with regard to the effect of Ham-D
scores on cognition and failed to show any impact on any
cognitive measures. However, the relationship between
antidepressant use and lower processing speed may be a
sign of an impact of residual symptoms on cognition.
This issue deserves further investigation. One other
potential confounding factor that can have an impact on
the magnitude of impairment in first-degree relatives of
BD could be the type of family members included
(siblings, offspring, twins). Since the number of
published relative studies was restricted, it was not
possible to investigate this issue further.
This meta-analytic study has several strengths and
original points. It investigates the cognitive deficits both
in euthymic patients and relatives of patients with BD.
Regarding sample size and cognitive domains involved,
it the most comprehensive meta-analytic study to date.
To our knowledge, it is the first meta-analytical study
that attempts to address the impact of clinical and
treatment confounders on cognitive phenotypes of BD.
In conclusion, response inhibition, set shifting, verbal
memory and target detection impairments are potential
candidate endophenotypes for BD. Some of the cognitive
impairments (including psychomotor slowness) observed
in euthymic patients could be related to the effects of
medication and illness-related factors. Future work should
carefully try to differentiate cognitive deficits associated
with disease genotype from impairments related to other
confounding factors. Longitudinal studies, studies investigating heritability of cognitive impairment in BD and its
relation with brain connectivity and genetics would be
especially useful.
Role of funding source
No funding source contributed to this paper.
Conflict of interest
Authors report no conflict of interest.

Acknowledgements
Murat Ycel was supported by a National Health &
Medical Research Council (NH&MRC) Clinical Career
Development Award (I.D. 509345).

18

E. Bora et al. / Journal of Affective Disorders 113 (2009) 120

References
Altshuler, L.L., Ventura, J., Van Gorp, W.G., Green, M.F., Theberge,
D.C., Mintz, J., 2004. Neurocognitive function in clinically stable
men with bipolar disorder or schizophrenia and normal control
subjects. Biol. Psychiatry 56, 560569.
Antila, A., Tuulio-Henriksson, A., Kieseppa, T., Eerola, M., Partonen,
T., Lonnqvist, J., 2007. Cognitive functioning in patients with
familial bipolar I disorder and their unaffected relatives. Psychol.
Med. 37, 679687.
Arts, B., Jabben, N., Krabbendam, L., van Os, J., 2008. Meta-analyses
of cognitive functioning in euthymic bipolar patients and theirfirstdegree relatives. Psychol Med 38, 771785.
Balanza-Martinez, V., Tabares-Seisdedos, R., Selva-Vera, G., MartinezAran, A., Torrent, C., Salazar-Fraile, J., Leal-Cercs, C., Vieta, E.,
Gmez-Beneyto, M., 2005. Persistent cognitive dysfunctions in
bipolar I disorder and schizophrenic patients: a 3-year follow-up
study. Psychother. Psychosom. 74, 113119.
Bax, L., Yu, L.M., Ikeda, N., Tsuruta, H., Moons, K.G.M., 2006.
Development and validation of MIX: comprehensive free software
for meta-analysis of causal research data. BMC Med. Res.
Methodol. 6, 50.
Blumberg, H.P., Leung, H.C., Skudlarski, P., Lacadie, C.M., Fredericks,
S.A., Harris, B.C., 2003. A functional magnetic resonance imaging
study of bipolar disorder: state and trait related dysfunctions in
ventral prefrontal cortices. Arch. Gen. Psychiatry 60, 601609.
Bora, E., Vahip, S., Gonul, A.S., Akdeniz, F., Alkan, M., Ogut, M.,
Eryavuz, A., 2005. Evidence for theory of mind deficits in euthymic
patients with bipolar disorder. Acta Psychiatr. Scand. 112, 110116.
Bora, E., Vahip, S., Akdeniz, F., 2006. Sustained attention deficits in
manic and euthymic patients with bipolar disorder. Prog.
Neuropsychopharmacol. Biol. Psychiatry 30, 10971102.
Bora, E., Vahip, S., Akdeniz, F., Gonul, A.S., Eryavuz, A., Ogut, M.,
Alkan, M., 2007. The effect of previous psychotic mood episodes
on cognitive impairment in euthymic bipolar patients. Bipolar
Disord. 9, 468477.
Bora, E., Vahip, S., Akdeniz, F., lerisoy, H., Aldemir, E., Alkan, M.,
2008. Executive and verbal working memory dysfunction in firstdegree relatives of patients with bipolar disorder. Psychiatry Res.
161, 318324.
Brambilla, P., MacDonald, A.W., Sassi, R.B., Johnson, M.K., Mallinger,
A.G., Carter, S., Soares, J.C., 2007. Context processing performance
in bipolar disorder patients. Bipolar Disord. 9, 230237.
Cavanagh, J.T., Van Beck, M., Muir, W., Blackwood, D.H.R., 2002. Case
control study of neurocognitive function in euthymic bipolar
disorder: an association with mania. Br. J. Psychiatry 180, 320326.
Christensen, M.V., Kyvik, K.O., Kessing, L.V., 2006. Cognitive
function in unaffected twins discordant for affective disorder.
Psychol. Med. 36, 11191129.
Clark, L., Goodwin, G.M., 2004. State- and trait-related deficits in
sustained attention in bipolar disorder. Eur. Arch. Psychiatry Clin.
Neurosci. 254, 6168.
Clark, L., Iversen, S.D., Goodwin, G.M., 2002. Sustained attention
deficit in bipolar disorder. Br. J. Psychiatry 180, 313319.
Clark, L., Sarna, A., Goodwin, G.M., 2005a. Impairment of executive
function but not memory in first-degree relatives of patients with
bipolar I disorder and in euthymic patients with unipolar
depression. Am. J. Psychiatry 162, 19801982.
Clark, L., Kempton, M.J., Scarna, A., Grasby, P.M., Goodwin, G.M.,
2005b. Sustained attention deficit confirmed in euthymic bipolar
disorder but not in first degree relatives of bipolar patients or
euthymic unipolar depression. Biol. Psychiatry 57, 183187.

Deckersbach, T., Savage, C.R., Reilly-Harrington, N., Clark, L.,

Sachs, G., Rauch, S.L., 2004a. Episodic memory impairment in
bipolar disorder and obsessive compulsive disorder: the role of
memory strategies. Bipolar Disord. 6, 233244.
Deckersbach, T., McMurrich, S., Ogutha, J., Savage, C.R., Sachs, G.,
Rauch, S.L., 2004b. Characteristics of nonverbal memory impairment in bipolar disorder: the role of encoding strategies. Psychol.
Med. 34, 823832.
Delis, D.C., Kramer, J.H., Kaplan, E., Ober, B.A., 1987. California
Verbal Learning Test: Adult Version. The Psychological Corporation, San Antonio, TX.
Dittmann, S., Seemller, F., Schwarz, M.J., Kleindienst, N., Stampfer, R.,
Zach, J., Born, C., Bernhard, B., Fast, K., Grunze, H., Engel, R.R.,
Severus, E., 2007. Association of cognitive deficits with elevated
homocysteine levels in euthymic bipolar patients and its impact on
psychosocial functioning: preliminary results. Bipolar Disord. 9,
6370.
Dixon, T., Kravariti, E., Frith, C., Murray, R.M., McGuire, P.K., 2004.
Effects of symptoms on executive function in bipolar illness.
Psychol. Med. 34, 811821.
Downes, J.J., Roberts, A.C., Sahakian, B.J., Evenden, J.L., Robins,
T.W., 1989. Impaired extradimensional shift performance
in medicated and unmedicated Parkinson's disease: evidence
for a specific attentional dysfunction. Neuropsychologia 27,
13291344.
El-Badri, S.M., Ashton, C.H., Moore, P.B., Marsh, V.R., Ferrier, I.N.,
2001. Electrophysiological and cognitive function in young
euthymic patients with bipolar affective disorder. Bipolar Disord.
3, 7987.
Ferrier, I.N., Stanton, B.R., Kelly, T.P., Scott, J., 1999. Neuropsychological function in euthymic patients with bipolar disorder. Br. J.
Psychiatry 175, 246251.
Ferrier, I.N., Chowdury, R., Thompson, J.M., Watson, S., Young, A.H.,
2004. Neurocognitive function in unaffected first-degree relatives
of patients with bipolar disorder: a preliminary report. Bipolar
Disord. 6, 319322.
Fleck, D.E., Shear, P.K., Zimmerman, M.E., Getz, G.E., Corey, K.B.,
Jak, A., Lebowitz, B.K., Strakowski, S.M., 2003. Verbal memory
in mania: effects of clinical state and task requirements. Bipolar
Disord. 5, 375380.
Frangou, S., Haldane, M., Roddy, M., Kumari, V., 2005a. Evidence for
deficits in tasks of ventral, but not dorsal, prefrontal executive
function as an endophenotypic marker for bipolar disorder. Biol.
Psychiatry 58, 838839.
Frangou, S., Donaldson, S., Hadjulis, M., Landau, S., Goldstein, L.H.,
2005b. The Maudsley bipolar disorder project: executive dysfunction in bipolar disorder I and its clinical correlates. Biol. Psychiatry
58, 859864.
Goswami, U., Sharma, A., Khastigir, U., Nicol, I., Ferrier, A.H.,
Gallagher, P., Thompson, J.M., Moore, P.B., 2006. Neuropsychological dysfunction, soft neurological signs and social disability in
euthymic patients with bipolar disorder. Br. J. Psychiatry 188,
366373.
Gottesman, I.I., Gould, T.D., 2003. The endophenotype concept in
psychiatry: etymology and strategic intentions. Am. J. Psychiatry
160, 636645.
Gourovitch, M.L., Torrey, E.F., Gold, J.M., Randolph, C., Weinberger,
D.R., Goldberg, T.E., 1999. Neuropsychological performance of
monozygotic twins discordant for bipolar disorder. Biol. Psychiatry 45, 639646.
Gur, R.E., Nimgaonkar, V.L., Almasy, L., Calkins, M.E., Ragland, J.D.,
Pogue-Geile, M.F., Kanes, S., Blangero, J., Gur, R.C., 2007.

E. Bora et al. / Journal of Affective Disorders 113 (2009) 120

Neurocognitive endophenotypes in a multiplex multigenerational
family study of schizophrenia. Am. J. Psychiatry 164, 813819.
Harmer, C.J., Clark, L., Grayson, L., Goodwin, G.M., 2002. Sustained
attention deficit in bipolar disorder is not a working memory
impairment in disguise. Neuropsychologia 40, 15861590.
Hawkins, K.A., Hoffman, R.E., Quinlan, D.M., Rakfeldt, J., Docherty,
N.M., Sledge, W.H., 1997. Cognition negative symptoms and
diagnosis: a comparison of schizophrenic, bipolar and control
samples. J. Neuropsychiatr. Clin. Neurosci. 9, 8189.
Heaton, R.K., 1981. Wisconsin Cart Sorting Test Manual: Odessa, FL,
Psychological Assessment Resources, Inc.
Hughes, A.M., Lynch, P., Rhodes, J., Ervine, C.M., Yates, R.A., 1999.
Electroencephalographic and psychomotor effects of chlorpromazine and risperidone relative to placebo in normal healthy
volunteers. Br. J. Clin. Pharmacol. 48, 323330.
Jones, P.B., Duncan, C.C., Mirsky, A.F., Post, R.M., Theodore, W.H.,
1994. Neuropsychological profiles in bipolar affective disorder
and complex partial seizure disorder. Neuropsychology 8, 5564.
Kaya, E., Aydemir, O., Selcuki, D., 2007. Residual symptoms in
bipolar disorder. The effect of last episode after remission. Progress
Neurophramacol. Biol. Psychiatr. 31, 13871392.
Keri, S., Kelemen, O., Benedek, G., Janka, Z., 2001. Different trait
markers for schizophrenia and bipolar disorder: a neurocognitive
approach. Psychol. Med. 31, 915922.
Kerr, N., Scott, J., Phillips, M.L., 2005. Patterns of attentional deficits
and emotional bias in bipolar and major depressive disorder. Br. J.
Clin. Psychol. 44, 343356.
Kieseppa, T., Tuulio-Herniksson, A., Haukka, J., Van Erp, T., Glahn,
D., Cannon, T.D., Partonen, T., Kaprio, J., Lnnqvist, J., 2005.
Memory and verbal learning functions in twins with bipolar-I
disorder and the role of information processing speed. Psychol.
Med. 35, 205215.
Klimes-Dougan, B., Ronsaville, D., Wiggs, E.A., Martinez, P.E., 2006.
Neuropsychological functioning in adolescent children of mothers
with a history of bipolar or major depressive disorders. Biol.
Psychiatry 60, 957965.
Kolur, U.S., Reddy, Y.C.J., John, P., Kandavel, T., Jain, S., 2006.
Sustained attention and executive functions in euthymic young
people with bipolar disorder. Br. J. Psychiatry 189, 453458.
Krabbendam, L., Honig, A., Wiersman, J., Vuurman, E.F.P.M.,
Hofman, P.A.M., Derix, M.M.A., Nolen, W.A., Jolles, J., 2000.
Cognitive dysfunction and white matter lesions in patients with
bipolar disorder in remission. Acta Psychiatr. Scand. 101, 274280.
Kremen, W.S., Faraone, S.V., Seidman, L.J., Pepple, J.R., Tsuang, M.T.,
1998. Neuropsychological risk indicators for schizophrenia: a
preliminary study of female relatives of schizophrenic and bipolar
probands. Psychiatry Res. 79, 227240.
Lezak, M.D., 1995. Neuropsychological Assessment. New York,
Oxford University Press.
Martinez-Aran, A., Vieta, E., Torrent, C., Sanchez-Moreno, J.,
Goikolea, J.M., Salamero, M., Malhi, G.S., Gonzalez-Pinto, A.,
Daban, C., Alvarez-Grandi, S., Fountoulakis, K., Kaprinis, G.,
Tabares-Seisdedos, R., Ayuso-Mateos, J.L., 2007. Functional
outcome in bipolar disorder: the role of clinical and cognitive
factors. Bipolar Disord. 9, 103113.
Martinez-Aran, A., Torrent, C., Tabares-Seisdedos, R., Salamero, M.,
Daban, C., Balanza-Martinez, V., Sanchez-Moreno, J., Manuel
Goikolea, J., Benabarre, A., Colom, F., Vieta, E., 2008.
Neurocognitive impairment in bipolar patients with and without
history of psychosis. J. Clin. Psychiatry 69, 233239.
McIntosh, A.M., Harrison, L.K., Forrester, K., Lawrie, S.M.,
Johnstone, E.C., 2005. Neuropsychological impairments in people

19

with schizophrenia or bipolar disorder and their unaffected

relatives. Br. J. Psychiatry 185, 378385.
McIntosh, A.M., Job, D.E., Moorhead, W.J., Harrison, L.K., Whalley,
H.C., Johnstone, E.C., Lawrie, S.M., 2006. Genetic liability to
schizophrenia or bipolar disorder and its relationship to brain
structure. Am. J. Med. Genet. B Neuropsychiatr. Genet. 141,
7683.
Monchi, O., Petrides, M., Petre, V., Worsley, K., Dagher, A., 2001.
Wisconsin Card Sorting revisited: distinct neural circuits participating in different stages of the task identified by event related
functional magnetic resonance imaging. J. Neurosci. 21, 7773-7741.
Morrens, M., Wezenberg, E., Verkes, R.J., Hulstijn, W., Ruigt, G.S.,
Sabbe, B.G., 2007. Psychomotor and memory effects of
haloperidol, olanzapine, and paroxetine in healthy subjects after
short-term administration. J. Clin. Psychopharmacol. 27, 1521.
Mur, M., Portella, M.J., Martinez-Aran, A., Pifarre, J., Vieta, E., 2007.
Persistent neuropsychological deficit in euthymic bipolar patients:
executive function as a core deficit. J. Clin. Psychiatry 68,
10781086.
Nehra, R., Chakrabarti, S., Pradhan, B.K., Khehra, N., 2006.
Comparison of cognitive functions between first- and multiepisode bipolar affective disorders. J. Affect. Disord. 93, 185192.
Nelson, H.E., 1982. National Adult Reading Test. NFER-Nelson,
Windsor, UK.
Pantelis, C., Yucel, M., Wood, S.J., Velakoulis, D., Sun, D., Berger, G.,
Stuart, G.W., Yung, A., Phillips, L., McGorry, P.D., 2005.
Structural brain imaging evidence for multiple pathological
processes at different stages of brain development in schizophrenia. Schizophrenia Bull. 31, 672696.
Pantelis, C., Velakoulis, D., Wood, S.J., Yucel, M., Yung, A.R.,
Phillips, L.J., Sun, D.Q., McGorry, P.D., 2007. Neuroimaging and
emerging psychotic disorders: the Melbourne ultra-high risk
studies. Int. Rev. Psychiatry 19, 371379.
Pantelis, C., Yucel, M., Wood, S.J., Brewer, W.J., Fornito, A., Berger,
G., Cannon, T., Velakoulis, D., in press. Recognition and
management of early psychosis: a preventive approach, 2nd
edition. Cambridge, Cambridge University Press.
Paradiso, S., Lamberty, G.J., Garvey, M.J., Robinson, R.G., 1997.
Cognitive impairment in the euthymic phase of chronic unipolar
depression. J. Nerv. Ment. Dis. 185, 748754.
Pirkola, T., Tuulio-Henriksson, A., Glahn, D., Kiesepp, T., Haukka,
J., Kaprio, J., Lnnqvist, J., Cannon, T.D., 2005. Spatial working
memory function in twins with schizophrenia and bipolar disorder.
Biol. Psychiatry 58, 930936.
Reitan, R.M., 1958. Validity of trail making test as an indication of
organic brain disease. Percept. Mot. Skills 8, 271276.
Rey, A., 1941. Psychological examination of traumatic encephalopathy. Archieves de Psychologic 28, 286340.
Rey, A., 1964. L'examen clinique en psychologie. Presses Universitaires de France, Paris.
Robinson, L.J., Ferrier, I.N., 2006. Evolution of cognitive impairments
in bipolar disorder: a systemic review of cross-sectional evidence.
Bipolar disord.. 8, 103116.
Robinson, L.J., Thompson, J.M., Gallagher, P., Goswami, U., Young,
A.H., Ferrier, I.N., Moore, P.B., 2006. A metaanalysis of cognitive
deficits in euthymic patients with bipolar disorder. J. Affect.
Disord. 93, 105115.
Rocca, C.C., Macedo-Soares, M.B., Gorenstein, C., Tamada, R.S.,
Isller, C.K., Dias, R.S., Almeida, K.M., Schwartzmann, A.M.,
Amaral, J.A., Lafer, B., 2008. Verbal fluency dysfunction in
euthymic bipolar patients: a controlled study. J. Affect. Disord.
107, 187192.

20

E. Bora et al. / Journal of Affective Disorders 113 (2009) 120

Rossi, A., Arduini, A., Daneluzzo, E., Bustini, M., Prosperini, P.,
Stratta, P., 2000. Cognitive function in euthymic bipolar patients,
stabilized schizophrenic patients and healthy controls. J. Psychiatr.
Res. 34, 333339.
Schouws, S.N.T.M., Zoeterman, J.B., Comijs, H.C., Stek, M.L.,
Beekman, A.T.F., 2007. Cognitive functioning in elderly patients
with early onset bipolar disorder. I. J. Geriatric Psychiatr. 22,
856861.
Senturk, V., Goker, C., Bilgic, A., Olmez, S., Tugcu, H., Oncu, B.,
Atbasoglu, E.C., 2007. Impaired verbal memory and otherwise
spared cognition in remitted bipolar patients on monotherapy with
lithium and valproate. Bipolar Disord. 9, 136144.
Sitskoorn, M.M., Aleman, A., Ebisch, S.J., Appels, M.C., Kahn, R.S.,
2004. Cognitive deficits in relatives of patients with schizophrenia:
a meta-analysis. Schizophr. Res. 71, 285295.
Smith, D.J., Muir, W.J., Blackwood, D.H.R., 2006. Neurocognitive
impairment in euthymic young adults with bipolar spectrum
disorder and recurrent major depressive disorder. Bipolar Disord.
8, 4046.
Snitz, B.E., Macdonald 3rd, A.W., Carter, C.S., 2006. Cognitive
deficits in unaffected first-degree relatives of schizophrenia
patients: a meta-analytic review of putative endophenotypes.
Schizophr. Bull. 32, 179194.
Sobczak, S., Honig, A., Schmitt, J.A., Riedel, W.J., 2003. Pronounced
cognitive deficits following an intravenous L-tryptophan challenge
in first-degree relatives of bipolar patients compared to healthy
controls. Neuropsychopharmacol 28, 711719.
Stoddart, S.D.R., Craddock, N.J., Jones, L.A., 2007. Differentiation of
executive and attention impairments in affective illness. Psychol.
Med. 37, 16131623.
Szoke, A., Schuroff, F., Golmard, J.L., Alter, C., Roy, I., Meary, A.,
Etain, B., Bellivier, F., Leboyer, M., 2006. Familial resemblance of
executive functions in families of schizophrenic and bipolar
patients. Psychiatry Res. 14, 131138.
Swann, A.C., Pazzaglia, P., Nicholls, A., Dougherty, D.M., Moeller, F.G.,
2003. Impulsivity and phase of illness in bipolar disorder. J. Affect.
Disord. 73 (1-2), 105111.
Szke, A., Schrhoff, F., Mathieu, F., Meary, A., Ionescu, S., Leboyer,
M., 2005. Tests of executive functions in first-degree relatives of
schizophrenic patients: a meta-analysis. Psychol. Med. 35,
771782.

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 disorder. Br. J. Psychiatry 186,
3240.
Thompson, J.M., Gray, J.M., Hughes, J.H., Watson, S., Young, A.H.,
Ferrier, I.N., 2007. Impaired working memory monitoring in
euthymic bipolar patients. Bipolar Disord. 9, 478489.
Torres, I.J., Boudreau, V.G., Yatham, L.N., 2007. Neuropsychological
functioning in euthymic bipolar disorder: a meta-analysis. Acta
Psychiatr. Scand. Suppl. 434, 1726.
Tuulio-Henriksson, A., Partonen, T., Suvisaari, J., Haukka, J.,
Lnnqvist, J., 2004. Age at onset and cognitive functioning in
schizophrenia. Br. J. Psychiatry 185, 215219.
Van Gorp, W.G., Altshuler, L., Theberge, D.C., Wilkins, J., Dixon, W.,
1998. Cognitive impairment in euthymic patients with and without
prior alcohol dependence. Arch. Gen. Psychiatry 55, 4146.
Van Gorp, W.G., Altshuller, L., Theberge, D.C., Mintz, J., 1999.
Declarative and procedural memory in bipolar disorder. Biol.
Psychiatry 46, 525531.
Varga, M., Magnusson, A., Flekkoy, K., Ronneberg, U., Opjordsmoen,
S., 2006. Insight, symptoms and neurocognition in bipolar I
patients. J. Affect. Disord. 91, 19.
Wechsler, D., 1981. Wechsler Adult Intelligence Scale Revised.
Psychological Corporation, New York.
Wechsler, D., 1987. Wechsler Memory Scale Revised Manual. The
Psychological Corporation, New York.
Wood, S.J., Brewer, W.J., Koutsouradis, P., Phillips, L.J., Francey, S.M.,
Proffitt, T.M., Yung, A.R., Jackson, H.J., McGorry, P.D., Pantelis, C.,
2007. Cognitive decline following psychosis onset: data from the
PACE clinic. Br. J. Psychiatry 51, 5257.
Zalla, T., Joyce, C., Szoke, A., Schurhoff, F., Pillon, B., Komano, O.,
Perez-Diaz, F., Bellivier, F., Alter, C., Dubois, B., Rouillon, F.,
Houde, O., Leboyer, M., 2004. Executive dysfunctions as potential
markers of familial vulnerability to bipolar disorder and schizophrenia. Psychiatry Res. 121, 207217.
Zubieta, J.K., Huguelet, P., O, ' Neil, R.L., Giordani, B.J., 2001.
Cognitive function in euthymic bipolar I disorder. Psychiatry Res.
(102), 920 2001.