You are on page 1of 8

Available online at www.sciencedirect.

com

Progress in Neuro-Psychopharmacology & Biological Psychiatry 32 (2008) 1022 1029


www.elsevier.com/locate/pnpbp

A relationship between bipolar II disorder and borderline


personality disorder?
Franco Benazzi
University of California at San Diego, United States
Psychiatry Research Center at Forli, Italy
Department of Psychiatry, University of Szeged, Szeged, Hungary
Department of Psychiatry, National Health Service, Forli, Italy
Received 30 December 2007; received in revised form 19 January 2008; accepted 21 January 2008
Available online 4 February 2008

Abstract
Background: The relationship between DSM-IV-TR borderline personality disorder (BPD) and bipolar disorders, especially bipolar II disorder
(BP-II), is still unclear. Many recent reviews on this topic have come to opposite or different conclusions.
Study aim: The aim was to test the association between hypomania symptoms and BPD traits, as hypomania is the defining feature of BP-II in
DSM-IV-TR.
Methods: During follow-up visits in a private practice, consecutive 138 remitted BP-II outpatients were re-diagnosed by a mood disorder specialist
psychiatrist, using the Structured Clinical Interview for DSM-IV (as modified by Benazzi and Akiskal for better probing hypomania). Soon after,
patients self-assessed (blind to interviewer) the SCID-II Personality Questionnaire for BPD.
Associations and confounding were tested by logistic regression, between each criteria symptom of hypomania (apart from racing thoughts
and distractibility, not assessed as probing focused mainly on behavioral, observable signs), and the entire set of BPD traits. Multivariate
regression was also used to jointly regress the entire set of hypomanic symptoms on the entire set of BPD traits.
Results: Mean (SD) age was 39.0 (9.8) years, females were 76.3%. Frequency of BPD traits ranged between 17% and 66% (e.g. impulsivity trait
41%, affective instability trait 63%), mean (SD) number of traits was 4.2 (2.3). The most common episodic hypomanic symptoms were elevated
mood (91%) and overactivity (93%); frequency of excessive risky, impulsive activities (impulsivity) was 62%. By logistic regression the only
significant association was between the episodic impulsivity of hypomania and the trait impulsivity of BPD. Multivariate regression of the entire
set of hypomanic symptoms jointly regressed on the entire set of BPD traits was not statistically significant.
Discussion: The core feature of BP-II, i.e. hypomania, does not seem to have a close relationship with BDP traits in the study setting, partly
running against a strong association between BPD and BP-II and a bipolar spectrum nature of BPD.
2008 Elsevier Inc. All rights reserved.
Keywords: Bipolar II disorder; Borderline personality disorder; Hypomania; Relationship; Traits

Abbreviations: BPD, borderline personality disorder; borderline personality disorder traits: BPD1, frantic efforts to avoid abandonment; BPD2, unstable
interpersonal relationships; BPD3, unstable self-image; BPD4, impulsivity; BPD5, suicidality; BPD6, affective instability; BPD7, chronic emptiness; BPD8, anger;
BPD9, paranoid ideation; BP-II, bipolar II disorder; DSM-IV-TR, diagnostic and statistical manual of mental disorders, forth edition, text revision; GAF, global
assessment of functioning scale; hypomania symptoms: H1, elevated mood; H2, irritable mood; H3, inflated self-esteem; H4, decreased need for sleep; H5, more
talkative; H6, racing thoughts; H7, distractibility; H8, increase in goal-directed activity; H9, excessive risky, impulsive activities; MDD, major depressive disorder;
MDE, major depressive episode; NA, not available; N, number; NC, not calculable; OR, odds ratio; 95% CI, 95% confidence intervals; SCID-II, Structured Clinical
Interview for DSM-IV Axis II personality disorders; SCID-CV, Structured Clinical Interview for DSM-IV Axis I Disorders-Clinician Version; SD, standard deviation;
95% CI, 95% confidence intervals; t, t-test statistics; p, p value.
Via Pozzetto 17, 48015 Castiglione di Cervia RA, Italy. Tel.: +39 335 6191 852; fax: +39 054 330 069.
E-mail address: FrancoBenazzi@FBenazzi.it.
0278-5846/$ - see front matter 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.pnpbp.2008.01.015

F. Benazzi / Progress in Neuro-Psychopharmacology & Biological Psychiatry 32 (2008) 10221029

1. Introduction
The relationship between DSM-IV-TR (American Psychiatric Association, 2000) borderline personality disorder (BPD)
and bipolar disorders (the bipolar spectrum), especially bipolar
II disorder (BP-II), is still unclear. According to DSM-IV-TR,
the features of a personality disorder should be enduring,
stable and of long duration (American Psychiatric Association, 2000). Compared to other personality disorders, BPD is
described as less stable in DSM-IV-TR. Many recent reports and
reviews on this topic have come to opposite or different conclusions about the BPD and BP-II relationship, i.e. that BPD is
part of the bipolar spectrum (e.g. Magill, 2004; Akiskal, 2004;
Smith et al., 2004; Mackinnon and Pies, 2006; Joyce et al.,
2007), or that there is no (or only a modest) relationship
between these disorders (e.g. Koenigsberg et al., 2002; George
et al., 2003; Gunderson et al., 2006; Paris, 2007; Paris et al.,
2007; Gunderson, 2007; Wilson et al., 2007). From the bipolar
spectrum ventage, there is an overlap of symptoms (mainly
cross-sectional) between BP-II and BPD, related to several BPD
affective traits. This overlap is more evident (or confounded?)
in tertiary-care settings (Hantouche et al., 1998; Perugi et al.,
2003; Akiskal et al., 2003; Mckinnon and Pies, 2006) than in
non-tertiary care (e.g. Benazzi, 2000). This BPDBP-II overlap
is also related, among other things, to more or less strict criteria
for diagnosing BP-II, to the methods used for probing hypomania (Dunner and Tay, 1993; Angst et al., 2003, 2005;
Benazzi, 2007a; Bader and Dunner, 2007), and to the interviewers' qualifications and training (leaving aside possible biases). An overlap of symptoms between BPD and BP-II is
especially likely in the depressive and hypomanic mixed states
(defined by opposite polarity symptoms concurrently present in
the same episode), which are a common feature of BP-II (reviewed by Goodwin and Jamison (2007) and by Benazzi
(2007a,b,c, 2008)). The Collaborative Longitudinal Personality
Disorders Study (CLPS) has shown that meaningful improvements are not uncommon in personality disorders (Skodol et al.,
2005a). Dimensional personality traits appear to be the foundation of behaviors described by many personality disorders
criteria (Skodol et al., 2005a). On the basis of the CLPS findings, personality disorders have been reconceptualized (differently from DSM-IV-TR) as hybrids of stable personality traits
and intermittently expressed symptomatic behaviors (Skodol
et al., 2005a). About BPD, its course has been found to be
remitting or a partly remitting (Zanarini et al., 2004, 2007;
Tragesser et al., 2007). According to Akiskal (2004), a remitting
course in BPD could shift it into the bipolar spectrum (because
it has several affective symptoms/traits), as the core feature of
bipolar disorders is thought to be a highly recurrent course
(Kraepelin, 1921; Goodwin and Jamison, 2007). Probably, a
core feature of bipolar disorders could also be a young age at
onset (Benazzi and Akiskal, 2007), which is also a core feature
of personality disorders (American Psychiatric Association,
2000). Furthermore, the high frequency of inter-episode residual symptoms in bipolar disorders (Judd et al., 2003; Paykel
et al., 2006) increases the similarities between BPD and bipolar
disorders (especially BP-II), on the basis of an unstable (re-

1023

mitting/partly remitting) course of BPD, making it difficult


to distinguish BPD from BP-II both cross-sectionally and
longitudinally.
As the defining, core feature of BP-II (recurrent major
depressive episodes with hypomanic episodes) is hypomania
(American Psychiatric Association, 2000; Bader and Dunner,
2007), if BP-II and BPD were related there should be a close
relationship between hypomania and BPD traits.
1.1. Study aim
The aim was to test the association between symptoms of
hypomania and BPD traits. This is the first study, to my knowledge, approaching the topic of the relationship between BPD and
BP-II by this method. A Medline search (access 26 Dec 2007),
key words borderline personality and hypomania, found only
15 papers added in any date, of which only three reported on a
relationship between hypomania and BPD (without detailed analyses on the relationship between hypomanic symptoms and
BPD traits) in small samples or subsamples.
2. Methods
Detailed study methods can be found in recent, related reports published in this Journal (Benazzi, 2006a, 2007d).
2.1. Study setting
An outpatient psychiatry private practice in northern Italy.
This setting is the first or second (after general practitioners)
line of treatment of mood disorders in this region, which most
individuals can afford (fee-for-service, reducing a possible income bias). It is thus more representative of non-psychotic
outpatient mood disorders than tertiary care, which is biased
toward the most severe cases (Perugi et al., 2003).
2.2. Interviewer
A clinical (25 years in practice) and research mood disorder
specialist psychiatrist (FB), whose inter-rater reliability k for
BP-II diagnosis, previously tested, was 0.73 (Benazzi, 2003a).
2.3. Patient population
Consecutive 138 remitted BP-II outpatients (i.e. GAF score
N80 for at least one month, no/very few symptoms, normal/near
normal functioning), who had previously presented voluntarily
for treatment of a major depressive episode (MDE), were assessed
in 2005. As suggested by DSM-IV-TR (American Psychiatric
Association, 2000), interviewing remitted BP-II patients is more
likely to reduce any recall bias and any bias related to the
prevalent mood of an episode, which can impact assessment of
past hypomania and of personality disorders (Peselow et al.,
1995). Concurrent interview of key informants (often present)
further increased the validity of assessment of past hypomania and
BPD (Peselow et al., 1995). Patients showing clinically significant general medical illnesses and cognitive disorders were

1024

F. Benazzi / Progress in Neuro-Psychopharmacology & Biological Psychiatry 32 (2008) 10221029

not included. This remitted group of patients was a parallel group


of a larger, growing group of acute BP-II patients described in
details in previous reports (e.g. Benazzi and Akiskal, 2003;
Akiskal and Benazzi, 2005), different because BPD was not an
exclusion criterion as in the acute sample. Only a small number of
variables were recorded, due to the time-limits of this busy private
practice. Written informed consent was taken, and study was
approved by the local ethical committee.
2.4. Assessment instruments
During follow-up visits, patients were re-diagnosed by the
Structured Clinical Interview for DSM-IV Axis I DisordersClinician Version (SCID-CV, reported inter-rater reliability
k = 0.701.0) (First et al., 1997a), as modified and validated by
Benazzi and Akiskal (2003) and Akiskal and Benazzi (2005) to
improve the probing for history of hypomania. Patients were kept
blind to the diagnostic interview results. The SCID-CV is partly
semi-structured (i.e. wording of the questions can be changed to
improve and check understanding by the interviewee if necessary)
and is based on clinical evaluation. Semi-structured interviewing
has been shown to outperform structured interviewing by reducing the false-negative BP-II (Simpson et al., 2002; Benazzi,
2003b; Kessler et al., 2006; Bader and Dunner, 2007).
In order to increase reliability and validity, the interview on
past hypomania did not include the symptoms racing thoughts
and distractibility, and it was mainly focused on observable
behaviors (such as overactivity) because behaviors are more
likely to be remembered by patients and key informants
(Dunner and Tay, 1993; Akiskal, 1996; Angst et al., 2003, 2005;
Benazzi and Akiskal, 2003; Benazzi, 2007a; Bader and Dunner,
2007). Mood change was however always required to diagnose
hypomania (and thus BP-II). It is easier to remember mood
change after remembering overactivity (Benazzi and Akiskal,
2003; Angst et al., 2003, 2005), without seeing it as normal
mood fluctuations, which is a major cause of misdiagnosis in
the SCID-CV, where mood change is the stem (and skip-out)
question (Dunner and Tay, 1993). Patients were asked which
were the most common hypomanic symptoms during the episodes. The Global Assessment of Functioning scale (GAF, in
the SCID-CV) was used to assess remission, as it assesses both
symptoms and functioning (as functional remission may not
follow symptoms remission). Often, family members or close
friends supplemented clinical information during the interviews. BPD traits were self-assessed (blind to the interviewer)
soon after the follow-up visit in the waiting room, by the SCIDII Personality Questionnaire (First et al., 1997b). A diagnosis of
BPD cannot be made by this Questionnaire alone, because
diagnosing BPD requires clinical evaluation of the severity of
its traits. Thus, the self-scored BPD traits were a mixture of
clinical and subclinical traits. In a previous study (Benazzi,
2000) in the same setting by the same interviewer, by using the
clinician version of the SCID-II (First et al., 1997b), frequency
of BPD in BP-II was found to be low (12%). Instead, by the selfassessed SCID-II Personality Questionnaire in the present BP-II
sample, frequency of BPD traits was much higher. The severity
of BPD traits should not have much impacted the results, as the

Table 1
Sample features
Variables: mean (SD), %

n = 138

Age, years
Females
Episodic hypomanic symptoms
H1 Elevated mood
H2 Irritable mood
H3 Inflated self-esteem
H4 Decreased need for sleep
H5 More talkative
H6 Racing thoughts
H7 Distractibility
H8 Increase in goal-directed activity
H9 Excessive impulsive activities
BPD traits
BPD1 Frantic efforts to avoid abandonment
BPD2 Unstable interpersonal relationships
BPD3 Unstable self-image
BPD4 Impulsivity
BPD5 Suicidality
BPD6 Affective instability
BPD7 Chronic emptiness
BPD8 Anger
BPD9 Paranoid ideation
N traits

39.0 (9.8)
67.3
90.5
67.6
77.2
49.6
67.1
NA
NA
93.4
62.0
65.9
47.4
44.4
41.1
16.7
62.7
59.8
39.4
49.6
4.2 (2.3)

Borderline personality disorder = BPD; H = hypomania; NA = not available


because not assessed.

same hypomania lies along a continuum of severity (Kraepelin,


1921; Akiskal, 1996; Judd and Akiskal, 2003; Judd et al., 2003;
Angst et al., 2003, 2005; Benazzi and Akiskal, 2006; Akiskal
et al., 2006b; Merikangas et al., 2007; Angst, 2007), as well as
BPD (Skodol et al., 2005a,b). A dimensional approach to personality disorders seems more supported than the current,
DSM-IV-TR, categorical one (Jablensky, 2002; Skodol et al.,
2005a,b; Benazzi, 2006b; Walters et al., 2007), a view, however,
not shared by others (e.g. Paris, 2007).
2.5. Testing study aim
In order to assess the relationship between BPD and BP-II,
associations were tested between each criteria symptom/sign of
hypomania (as hypomania is the defining feature of BP-II in
DSM-IV-TR) and the entire set of traits defining BPD in DSMIV-TR. Not having recorded the hypomanic criteria symptoms
racing thoughts and distractibility should not have impacted
results significantly, as these symptoms are not a stem criterion
(nor very specific symptoms, too) of DSM-IV-TR hypomania
and of its suggested, alternative definitions (requiring behavioral overactivity instead of mood change, or both) (Akiskal,
1996; Angst et al., 2003; Akiskal and Benazzi, 2005, 2007a).
2.6. Statistics
Multivariable (multiple) logistic regression (testing the
associations between one categorical dependent variable and
a set of independent variables) was used to test associations and
to control for confounding. Multivariate regression (by which a
set of dependent variables can be jointly regressed on a set of

F. Benazzi / Progress in Neuro-Psychopharmacology & Biological Psychiatry 32 (2008) 10221029

1025

Table 2
Multivariable logistic regression of each hypomanic symptom (dependent variable) versus all set of borderline personality disorder traits (independent variables)

BPD1
BPD2
BPD3
BPD4
BPD5
BPD6
BPD7
BPD8
BPD9

H1

H2

H3

1.0 (0.24.3)
0.8 (0.13.9)
7.8 (0.877.5)
0.5 (0.12.7)
1.1 (0.013.5)
1.3 (0.28.2)
2.7 (0.417.2)
0.5 (0.12.4)
0.8 (0.13.9)

2.1 (0.94.9)
0.9 (0.32.2)
0.7 (0.21.9)
1.9 (0.84.9)
0.6 (0.12.1)
2.3 (0.86.9)
0.5 (0.21.6)
2.2 (0.95.6)
1.5 (0.63.9)

1.1
1.8
1.7
0.8
0.4
0.3
1.7
1.4
0.8

(0.42.9)
(0.74.6)
(0.64.7)
(0.32.0)
(0.11.7)
(0.11.1)
(0.55.3)
(0.53.8)
(0.32.0)

H4

H5

H6

H7

H8

H9

0.4 (0.21.0)
0.8 (0.31.7)
1.8 (0.74.4)
1.2 (0.52.7)
0.8 (0.22.4)
0.8 (0.32.1)
0.5 (0.21.3)
1.8 (0.84.0)
0.8 (0.41.9)

0.8 (0.32.0)
2.2 (0.95.4)
3.1 (1.28.2)
2.1 (0.85.2)
1.3 (0.34.6)
0.3 (0.11.0)
1.3 (0.43.9)
0.7 (0.31.8)
0.6 (0.21.5)

NA
NA
NA
NA
NA
NA
NA
NA
NA

NA
NA
NA
NA
NA
NA
NA
NA
NA

2.8 (0.612.5)
1.2 (0.26.8)
1.5 (0.211.3)
0.3 (0.01.8)
NC
1.1 (0.17.1)
1.2 (0.27.0)
4.7 (0.543.9)
1.9 (0.311.2)

0.9 (0.42.1)
1.2 (0.52.9)
0.9 (0.32.2)
2.8 (1.26.5)
1.2 (0.43.9)
0.4 (0.11.3)
1.5 (0.54.2)
1.3 (0.53.0)
0.7 (0.31.7)

Odds ratios and 95% confidence intervals are presented.


= p b 0.05; = p b 0.01; NA = not available, because not assessed; NC = not calculable.
DSM-IV-TR hypomania symptoms: H1 = elevated mood, H2 = irritable mood, H3 = inflated self-esteem, H4 = decreased need for sleep, H5 = more talkative, H6 =
racing thoughts, H7 = distractibility, H8 = increase in goal-directed activity, H9 = excessive risky, impulsive activities.
DSM-IV-TR borderline personality disorder traits: BPD1 = frantic efforts to avoid abandonment, BPD2 = unstable interpersonal relationships, BPD3 = unstable selfimage, BPD4 = impulsivity, BPD5 = suicidality, BPD6 = affective instability, BPD7 = chronic emptiness, BPD8 = anger, BPD9 = paranoid ideation.

independent variables) was also used to check, by a different


statistics, the results of logistic regression, and also as an
alternative analysis. ROC (Receiver-Operating-Characteristic)
analysis was used to study the distribution of BPD traits.
STATA statistical software, release 9.2, was used (StataCorp.
2006, College Station, TX, USA). P values were two-tailed,
and alpha level was set at 0.05.
3. Results
Sample features are presented in Table 1. BPD traits were
frequent in BP-II. The mean (SD) number of BPD traits was 4.2
(2.3), and all but one (suicidality) BPD traits had a frequency
ranging between 39% and 66%. The three most common BPD
traits were frantic efforts to avoid abandonment (65.9%),
affective instability (62.7%), and chronic emptiness (59.8%).
Impulsivity had a frequency of 41.1%. The distribution of
BPD traits among the BP-II patients was the following: 97.0%
had N= 1 trait, 83.9% had N=2 traits, 71.5% had N=3 traits,
63.5% had N= 4 traits, 45.9% had N=5 traits, 33.5% had N=6
traits, 19.7% had N=7 traits, 8.7% had N= 8 traits, and 0.7% had 9
traits.
Logistic regression of each hypomanic symptom/sign (the
categorical dependent variable) versus the entire set of BPD
traits (the independent variables, in order to control for confounding), is presented in Table 2. Only two statistically significant independent associations were found between BPD
traits and hypomania (H) symptoms: 1) BPD3 unstable selfimage and H5 more talkative, and 2) BPD4 impulsivity
and H9 excessive risky, impulsive activities.
Then, by multivariate regression, the entire set of hypomanic
symptoms (dependent variables) was jointly regressed on the
entire set of BPD traits. The test of the joint significance showed
F = 1.27, p = 0.1339. Looking at the individual regressions of
each hypomanic symptom/sign on the set of BPD traits, none
resulted statistically significant, apart from the association
between BPD4 impulsivity trait and H9 excessive risky,
impulsive activities (coefficient = 0.21, t = 2.27, p = 0.025, 95%
CI = 0.02 to 0.40).

4. Discussion
4.1. An association between BPD and BP-II?
BPD traits (both clinical and subclinical) were common in
the present study BP-II sample. The impulsivity trait was
present in 41%, the affective instability trait in 63%, and the
mean (SD) number of BPD traits was 4.2 (2.3) (DSM-IV-TR
BPD criteria require at least 5 traits impairing functioning). Four
or more BPD traits were present in 63.5% of BP-II, and 5 or
more were present in 45.9% of BP-II. Such a high co-occurrence could suggest a link between these disorders. However,
when the core feature of BP-II, i.e. hypomania (Bader and
Dunner, 2007), was tested (each symptom/sign independently)
versus the entire set of BPD traits (by multivariable logistic
regression), it resulted that only two hypomanic signs were
significantly associated with only two BPD traits. Of these two
hypomanic signs, only one, i.e. excessive risky, impulsive
activities (e.g. buying sprees, sexual indiscretions, reckless
driving, foolish business investments, substance abuse, according to DSM-IV-TR), had a meaningful association, as it was
associated with the BPD impulsivity trait (OR = 2.8). Using a
different statistics, i.e. multivariate regression, by which the
entire set of hypomanic symptoms was jointly regressed on
the entire set of BPD traits, no joint significance was found (i.e.
no significant relationship between the two sets of symptoms
and traits was found). This second analysis is more conservative. By looking at its individual regressions of each hypomanic
symptom/sign on the set of BPD traits (overcome by the joint
regression, however, and which has a similarity to multivariable
logistic regression in regressing one dependent variable on a set
of variables), none resulted statistically significant, apart from
the association between BPD impulsivity trait and the excessive risky, impulsive activities of hypomania. The results of
this second type of analysis suggest no link between hypomania
and BPD, and only a possible link between the impulsivity of
BPD and BP-II (replicating the first analysis, which found a
significant association between the impulsivity of hypomania
and that of BPD).

1026

F. Benazzi / Progress in Neuro-Psychopharmacology & Biological Psychiatry 32 (2008) 10221029

A relationship was found between the hypomanic/hyperthymic personality and traits of BPD (Kwapil et al., 2000). The
present study tested instead the relationship between symptoms
of hypomanic episodes and BPD traits. The hypomanic/hyperthymic personality (which may look like a low-grade, persistent
hypomania) is clinically different from BP-II, which is defined
by recurrent episodes of hypomania and depression (American
Psychiatric Association, 2000). Only a minority of BP-II (15
20%) have a concurrent hyperthymic personality (Hantouche
et al., 1998; Akiskal et al., 2006a,b). Frequency of BPD in BP-II
was found to be high in tertiary-care settings (Perugi et al.,
2003; Akiskal et al., 2006a; Joyce et al., 2007), but associations
among symptoms of hypomania and BPD traits were not tested,
as in the present study. Childhood and adolescent hypomania
alone (not BP-II) was found to be a risk factor for adult BPD
(Joyce et al., 2003), but its frequency in BPD was low (Levitt
et al., 1990).
4.2. Impulsivity in BPD and BP-II
BPD impulsivity trait and affective instability trait are
among the most common and persistent features of BPD (more
often so for affective instability) (Links et al., 1999;
McGlashan et al., 2005; Zanarini et al., 2007; Tragesser et al.,
2007; Grilo et al., 2007). Thus, BPD impulsivity may represent one of its most important domains. The impulsivity of
hypomanic episodes has been shown to lie along a continuum
with inter-episode trait impulsivity, a trait found to be common
in BP-II (Fergus et al., 2003; Swann et al., 2007; Benazzi,
2007e; Peluso et al., 2007). Therefore, the present study association found between the impulsivity trait of BPD and the
impulsivity sign of hypomania may simply result from a trait
impulsivity common to both disorders. Furthermore, by studying impulsivity in more details, it has been shown that the
impulsivity profile of BPD may have distinct features compared
to the impulsivity profile of BP-II (i.e. more severity, less
attentional impulsiveness, more non-planning and motor impulsiveness, more cognitive and anxiety symptoms) (Wilson
et al., 2007). Also the affective instability of BPD and BP-II
seem to have distinguishing features when assessed by detailed
scales (Henry et al., 2001).
In our previous factor analysis study of the same sample
(Benazzi, 2006a) BPD traits showed two factors: one including
affective instability, unstable interpersonal relationships,
unstable self-image, chronic emptiness, and anger; and
one including impulsivity, suicidal behavior, avoidance of
abandonment, and paranoid ideation. The second factor including the impulsivity trait was not associated with a diagnosis of BP-II (which was instead the case for the affective
instability trait), but both factors had the same weak OR versus
BP-II (i.e. OR = 1.2), with minor differences in the 95% confidence intervals. Compared to this study, the present study has
several important differences: symptoms of hypomania and not
the diagnosis of BP-II were tested; more advanced statistical
methods were used, taking into account confounding among the
sets of symptoms of hypomania and of BPD traits; the impulsivity trait was tested (and not a factor including several

other traits apart from impulsivity); the entire sets of symptoms (the symptoms of hypomania, the traits of BPD) were
tested by multivariate analyses. The present study methods
tested the relationship between BPD and BP-II more in-depth,
and its results should have more validity compared to the former
study.
4.3. The definition of BPD
Much of the current unclear status of BPD (a personality
disorder or a bipolar spectrum disorder?) results also from the
DSM-IV-TR set of BPD traits, which include many traits more or
less related to mood instability and to bipolar disorders (i.e.
affective instability, unstable interpersonal relationships,
unstable self-image, impulsivity, suicidality, chronic emptiness, and anger) (Akiskal, 2004). Which domain(s) best
defines BPD is another current hot topic (affective instability?,
impulsivity?, disturbed relationships? or cognitive defects?) (Links et al., 1999; McGlashan et al., 2005; Paris,
2007; Paris et al., 2007; Tragesser et al., 2007; Zanarini et al.,
2007; Grilo et al., 2007; Gunderson, 2007).
Course may be the best validator to study BPD (Zanarini
et al., 2004; Skodol et al., 2005a; Gunderson et al., 2006;
Tragesser et al., 2007). However, it has been shown that BPD
traits have different degrees of long-term stability, varying according to the studies (Links et al., 1999; McGlashan et al.,
2005; Zanarini et al., 2004, 2007; Gunderson et al., 2006;
Tragesser et al., 2007; Grilo et al., 2007).
4.4. Some features of BP-II blurring the boundaries with BPD
On the other side, BP-II has been shown to be not only a
highly recurrent disorder (i.e. a very unstable one), but also to
have frequently mixed features, residual symptoms, trait impulsivity, and trait mood instability (Judd et al., 2003; Paykel
et al., 2006; Akiskal et al., 1995, 2003; Angst et al., 2003;
Perugi et al., 2003; Suppes et al., 2005; Benazzi, 2007b,e;
Swann et al., 2007; Peluso et al., 2007; Rihmer et al., 2007). All
these features lead to a stable instability in BP-II overlapping
with BPD traits, making it difficult to set a boundary between
(i.e. to disentangle) these disorders.
The present study, by focusing only on the hypomanic episodes, the defining feature of BP-II, may have partly overcome
some of these problems. It has been suggested that the affective
instability trait included in DSM-IV-TR BPD definition can be
the main cause of its misdiagnosis as a bipolar spectrum disorder, and that the core feature of BPD could be its trait impulsivity (Gunderson et al., 1996). The results of the present
study seem to support this view, but study limitations have to be
taken into account.
4.5. Any treatment impact from the study findings?
Placing BPD among personality disorders instead of placing it
among bipolar disorders could have significant impacts on treatment, e.g. more psychological treatment than antibipolar (mood
stabilising agents) pharmacology. However, clear evidence of the

F. Benazzi / Progress in Neuro-Psychopharmacology & Biological Psychiatry 32 (2008) 10221029

efficacy of mood stabilising agents on BPD, apart from the impact


on some traits, has still to be shown (e.g. Hollander et al., 2005).
Combination of these treatment approaches sounds rational (and
necessary), and is part of the daily care of BPD patients in the
usual clinical practice of the present study country.
4.6. Conclusion
The core feature of BP-II, i.e. hypomania, does not seem to
have a close relationship with BPDs entire set of traits in the
study setting, partly running against a strong association between BPD and BP-II, and a bipolar spectrum nature of BPD.
4.7. Limitations
This population of patients may not be the best to answer the
primary research question under investigation. Obviously, a
parallel study which assessed hypomanic features in BPD patients would also be of interest in assessing the overlap between
these two disorders.
One limitation of the self-assessment questionnaire used in
this study (i.e. the SCID-II Personality Questionnaire) is its lack
of grading of severity of traits, assessed only by yes/no questions, and of clinical evaluation, which can have inflated the
frequency of traits by including an unknown proportion of nonclinically, or only marginally clinically significant traits. However, it is likely that BPD traits are graded in the population as
for other DSM-IV-TR personality disorders (Jablensky, 2002;
Skodol et al., 2005a,b; Benazzi, 2006b; Walters et al., 2007),
ranging from subclinical levels to clinical levels of different
severity. This likely grading of the severity of the BPD traits
matches the same grading of severity of the hypomanic symptoms, ranging from symptoms which significantly improve
functioning to symptoms which instead reduce functioning (e.g.
Angst et al., 2003; Akiskal et al., 2003, 2006a,b; Benazzi and
Akiskal, 2006; Bader and Dunner, 2007). An advantage of selfassessment was that it overcame any possible interviewer's bias,
and self-assessment questionnaires are more and more used in
personality disorders research (e.g. Kwapil et al., 2000; Akiskal
et al., 2003; Peluso et al., 2007; Swann et al., 2007). Assessment
of BPD traits could have been better accomplished by using
detailed scales assessing distinct dimensions of BPD, such as
impulsivity and affective instability scales (e.g. Henry et al.,
2001; Wilson et al., 2007). Instead, the SCID-II Personality
Questionnaire and the SCID-CV do not assess the domains of
BPD traits, but assess broad indicators of these traits. However,
the SCID-II and SCID-CV are commonly used to diagnose Axis
II and Axis I disorders in research settings. In the present study
setting, a busy private practice in which research and clinical
work are carried out side by side, time is a strong limiting factor
for research. Study findings must therefore take into account
these limitations. However, in the study setting, using these
SCID-II broad indicators of affective instability and impulsivity traits (Benazzi, 2007e) led to findings close to those of
studies which used detailed instruments to assess temperamental instability in BP-II (e.g. Akiskal et al., 2003). It is
recommended to replicate the findings with a design in which

1027

the clinician version of the SCID-II is used and it is administered by an independent assessor.
The present study setting may be more clinically representative than tertiary care, but it might be biased towards some
disorders or complex patients because of the author's local
reputation (e.g. bipolar spectrum disorders, especially BP-II,
treatment-resistant depressions). However, the patients included
in the studies carried out in this setting must present off recent
psychopharmacotherapy (which is usually not the case in the
severe tertiary-care patients) and must not abuse substances;
complex patients (a minority in this setting) are not included.
Interviews were conducted by the same interviewer.
However, the interviewer had been studying BP-II for many
years, and his inter-rater reliability for BP-II diagnosis was
acceptable (Benazzi, 2003a). Key informants were often concurrently interviewed. Diagnosing BP-II followed current best
practice, based on semi-structured interviews by bipolar-trained
clinicians (Dunner and Tay, 1993; Simpson et al., 2002; Bader
and Dunner, 2007). The minimum duration of hypomania was
2 days instead of the 4 days required by DSM-IV-TR, but
this manual's cutoff was consensus-based, while the present
study cutoff is evidence-based (for more details: Judd et al.,
2003; Angst et al., 2003; Bader and Dunner, 2007; Benazzi,
2007b,c,d).
References
Akiskal HS. The prevalent clinical spectrum of bipolar disorders: beyond DSMIV. J Clin Psychopharmacol 1996;16(Suppl 1):4S14S.
Akiskal HS. Demystifying borderline personality: critique of the concept and
unorthodox reflections on its natural kinship with the bipolar spectrum. Acta
Psychiatr Scand 2004;110:4017.
Akiskal HS, Benazzi F. Optimizing the detection of bipolar II disorder in
outpatient private practice: toward a systematization of clinical diagnostic
wisdom. J Clin Psychiatry 2005;66:91421.
Akiskal HS, Maser JD, Zeller PJ, Endicott J, Coryell W, Keller M, et al.
Switching from unipolar to bipolar II. An 11-year prospective study of
clinical and temperamental predictors in 559 patients. Arch Gen Psychiatry
1995;52:11423.
Akiskal HS, Hantouche EG, Allilaire JF. Bipolar II with and without cyclothymic
temperament: dark and sunny expressions of soft bipolarity. J Affect
Disord 2003;73:4957.
Akiskal HS, Akiskal KK, Perugi G, Toni C, Ruffolo G, Tusini G. Bipolar II and
anxious reactive comorbidity: toward better phenotypic characterization
suitable for genotyping. J Affect Disord 2006a;96:23947.
Akiskal HS, Akiskal KK, Lancrenon S, Hantouche EG, Fraud JP, Gury C, et al.
Validating the bipolar spectrum in the French National EPIDEP Study:
overview of the phenomenology and relative prevalence of its clinical
prototypes. J Affect Disord 2006b;96:197205.
American Psychiatric Association. Diagnostic and statistical manual of mental
disorders, 4th ed, text revision (DSM-IV-TR). Washington, DC: American
Psychiatric Association; 2000.
Angst J. The bipolar spectrum. Br J Psychiatry 2007;190:18991.
Angst J, Gamma A, Benazzi F, Ajdacic V, Eich D, Rossler W. Toward a redefinition of subthreshold bipolarity: epidemiology and proposed criteria for
bipolar-II, minor bipolar disorders and hypomania. J Affect Disord
2003;73:13346.
Angst J, Adolfsson R, Benazzi F, Gamma A, Hantouche E, Meyer TD, et al. The
HCL-32: towards a self-assessment tool for hypomanic symptoms in
outpatients. J Affect Disord 2005;88:21733.
Bader CD, Dunner DL. Bipolar disorder not otherwise specified in relation to
the bipolar spectrum. Bipolar disorder not otherwise specified in relation to
the bipolar spectrum. Bipolar Disord 2007;9:8607.

1028

F. Benazzi / Progress in Neuro-Psychopharmacology & Biological Psychiatry 32 (2008) 10221029

Benazzi F. Borderline personality disorder and bipolar II disorder in private


practice depressed outpatients. Compr Psychiatry 2000;41:10610.
Benazzi F. Depression with racing thoughts. Psychiatry Res 2003a;120:27382.
Benazzi F. Diagnosis of bipolar II disorder: a comparison of structured versus
semistructured interviews. Prog Neuro-psychopharmacol Biol Psychiatry
2003b;27:98591.
Benazzi F. Borderline personalitybipolar spectrum relationship. Prog Neuropsychopharmacol Biol Psychiatry 2006a;30:6874.
Benazzi F. Does temperamental instability support a continuity between bipolar
II disorder and major depressive disorder? Eur Psychiatr 2006b;21:2749.
Benazzi F. Is overactivity the core feature of hypomania in bipolar II disorder?
Psychopathology 2007a;40:5460.
Benazzi F. Bipolar disorderfocus on bipolar II disorder and mixed depression.
Lancet 2007b;369:93545.
Benazzi F. Bipolar II disorder: epidemiology, diagnosis and management. CNS
Drugs 2007c;21:72740.
Benazzi F. Delineation of the clinical picture of dysphoric/mixed hypomania.
Prog Neuro-psychopharmacol Biol Psychiatry 2007d;31:94451.
Benazzi F. Impulsivity in bipolar-II disorder: trait, state, or both? Eur Psychiatry
2007e;22:4728.
Benazzi F. Reviewing the diagnostic validity and utility of mixed depression
(depressive mixed states). Eur Psychiatry 2008 Jan;23(1):408.
Benazzi F, Akiskal HS. Refining the evaluation of bipolar II: beyond the strict
SCID-CV guidelines for hypomania. J Affect Disord 2003;73:338.
Benazzi F, Akiskal H. The duration of hypomania in bipolar-II disorder in private
practice: methodology and validation. J Affect Disord 2006;96:18996.
Benazzi F, Akiskal HS. How best to identify a bipolar-related subtype among
major depressive patients without spontaneous hypomania: superiority of
age at onset criterion over recurrence and polarity? J Affect Disord 2007.
doi:10.1016/j.jad.2007.07.032 [Electronic publication ahead of print].
Dunner DL, Tay KL. Diagnostic reliability of the history of hypomania in
bipolar II patients and patients with major depression. Compr Psychiatry
1993;34:3037.
Fergus EL, Miller RB, Luckenbaugh DA, Leverich GS, Findling RL, Speer AM,
et al. Is there progression from irritability/dyscontrol to major depressive and
manic symptoms? A retrospective community survey of parents of bipolar
children. J Affect Disord 2003;77:718.
First MB, Spitzer RL, Gibbon M, Williams JBW. Structured Clinical Interview
for DSM-IV axis I Disorders-Clinician Version (SCID-CV). Washington,
DC: American Psychiatric Press; 1997a.
First MB, Gibbon M, Spitzer RL, Williams JBW, Benjamin LS. Structured
Clinical Interview for DSM-IV Axis II Personality Disorders (SCID-II).
Washington, DC: American Psychiatric Press; 1997b.
George EL, Miklowitz DJ, Richards JA, Simoneau TL, Taylor DO. The
comorbidity of bipolar disorder and axis II personality disorders: prevalence
and clinical correlates. Bipolar Disord 2003;5:11522.
Goodwin FK, Jamison KR. Manic-depressive illness, second edition. New York,
USA: Oxford University Press; 2007.
Grilo CM, Sanislow CA, Skodol AE, Gunderson JG, Stout RL, Bender DS, et al.
Longitudinal diagnostic efficiency of DSM-IV criteria for borderline personality disorder: a 2-year prospective study. Can J Psychiatry 2007;52:35762.
Gunderson JG. Disturbed relationships as a phenotype for borderline personality
disorder. Am J Psychiatry 2007;164:163740.
Gunderson J, Zanarini MC, Kisiel CL. Borderline personality disorder. In:
Widiger TA, Frances AJ, Pincus HL, Ross R, First MB, Davis WW, editors.
DSM-IV Sourcebook, vol. 2. Washington, DC: American Psychiatric
Association; 1996. p. 71731.
Gunderson JG, Weinberg I, Daversa MT, Kueppenbender KD, Zanarini MC,
Shea MT, et al. Descriptive and longitudinal observations on the relationship
of borderline personality disorder and bipolar disorder. Am J Psychiatry
2006;163:11738.
Hantouche EG, Akiskal HS, Lancrenon S, Allilaire JF, Sechter D, Azorin JM,
et al. Systematic clinical methodology for validating bipolar-II disorder:
data in mid-stream from a French national multi-site study (EPIDEP). J
Affect Disord 1998;50:16373.
Henry C, Mitropoulou V, New AS, Koenigsberg HW, Silverman J, Siever LJ.
Affective instability and impulsivity in borderline personality and bipolar II
disorders: similarities and differences. J Psychiatr Res 2001;35:30712.

Hollander E, Swann AC, Coccaro EF, Jiang P, Smith TB. Impact of trait
impulsivity and state aggression on divalproex versus placebo response in
borderline personality disorder. Am J Psychiatry 2005;162:6214.
Jablensky A. The classification of personality disorders: critical review and need
for rethinking. Psychopathology 2002;35:1126.
Joyce PR, McKenzie JM, Luty SE, Mulder RT, Carter JD, Sullivan PF, et al.
Temperament, childhood environment and psychopathology as risk factors
for avoidant and borderline personality disorders. Aust N Z J Psychiatry
2003;37:75664.
Joyce PR, Light KJ, Rowe SL, Kennedy MA. Bipolar disorder not otherwise
specified: comparison with bipolar disorder I/II and major depression. Aust
N Z J Psychiatry 2007;41:8439.
Judd LL, Akiskal HS. The prevalence and disability of bipolar spectrum
disorders in the US population: re-analysis of the ECA database taking into
account subthreshold cases. J Affect Disord 2003;73:12331.
Judd LL, Akiskal HS, Schettler PJ, Coryell W, Endicott J, Maser JD, et al. A
prospective investigation of the natural history of the long-term weekly
symptomatic status of bipolar II disorder. Arch Gen Psychiatry 2003;60:
2619.
Kessler RC, Akiskal HS, Angst J, Guyer M, Hirschfeld RM, Merikangas KR,
et al. Validity of the assessment of bipolar spectrum disorders in the WHO
CIDI 3.0. J Affect Disord 2006;69:25969.
Koenigsberg HW, Harvey PD, Mitropoulou V, Schmeidler J, New AS,
Goodman M, et al. Characterizing affective instability in borderline
personality disorder. Am J Psychiatry 2002;159:7848.
Kraepelin E. Manic-depressive insanity and paranoia. Edinburgh: Livingstone E
& S; 1921.
Kwapil TR, Miller MB, Zinser MC, Chapman LJ, Chapman J, Eckblad M. A
longitudinal study of high scorers on the hypomanic personality scale.
J Abnorm Psychology 2000;109:2226.
Levitt AJ, Joffe RT, Ennis J, MacDonald C, Kutcher SP. The prevalence of
cyclothymia in borderline personality disorder. J Clin Psychiatry 1990;51:
3359.
Links PS, Heslegrave R, van Reekum R. Impulsivity: core aspect of borderline
personality disorder. J Pers Disord 1999;13:19.
Mackinnon DF, Pies R. Affective instability as rapid cycling: theoretical and
clinical implications for borderline personality and bipolar spectrum
disorders. Bipolar Disord 2006;8:114.
Magill CA. The boundary between borderline personality disorder and bipolar
disorder: current concepts and challenges. Can J Psychiatry 2004;49:5516.
McGlashan TH, Grilo CM, Sanislow CA, Ralevski E, Morey LC, Gunderson
JG, et al. Two-year prevalence and stability of individual DSM-IV criteria
for schizotypal, borderline, avoidant, and obsessivecompulsive personality
disorders: toward a hybrid model of axis II disorders. Am J Psychiatry
2005;162:8839.
Merikangas KR, Akiskal HS, Angst J, Greenberg PE, Hirschfeld RM,
Petukhova M, et al. Lifetime and 12-month prevalence of bipolar spectrum
disorder in the National Comorbidity Survey replication. Arch Gen
Psychiatry 2007;64:54352.
Paykel ES, Abbott R, Morriss R, Hayhurst H, Scott J. Sub-syndromal and
syndromal symptoms in the longitudinal course of bipolar disorder. Br J
Psychiatry 2006;189:11823.
Paris J. The nature of borderline personality disorder: multiple dimensions,
multiple symptoms, but one category. J Pers Disord 2007;21:45773.
Paris J, Gunderson J, Weinberg I. The interface between borderline personality
disorder and bipolar spectrum disorders. Compr Psychiatry 2007;48:14554.
Peluso MA, Hatch JP, Glahn DC, Monkul ES, Sanches M, Najt P, et al. Trait
impulsivity in patients with mood disorders. J Affect Disord 2007;100:
22731.
Perugi G, Toni C, Travierso MC, Akiskal HS. The role of cyclothymia in
atypical depression: toward a data-based reconceptualization of the borderlinebipolar II connection. J Affect Disord 2003;73:8798.
Peselow ED, Sanfilipo MP, Fieve RR. Relationship between hypomania and
personality disorders before and after successful treatment. Am J Psychiatry
1995;152:2328.
Rihmer Z, Gonda X, Akiskal KK, Akiskal HS. Affective temperament: a
mediating variable between environment and clinical depression? Arch Gen
Psychiatry 2007;64:10967.

F. Benazzi / Progress in Neuro-Psychopharmacology & Biological Psychiatry 32 (2008) 10221029


Simpson SG, McMahon FJ, McInnis MG, MacKinnon DF, Edwin D, Folstein
SE, et al. Diagnostic reliability of bipolar II diagnosis. Arch Gen Psychiatry
2002;59:73640.
Skodol AE, Gunderson JG, Shea MT, McGlashan TH, Morey LC, Sanislow CA,
et al. The Collaborative Longitudinal Personality Disorders Study (CLPS):
overview and implications. J Pers Disord 2005a;19:487504.
Skodol AE, Oldham JM, Bender DS, Dyck IR, Stout RL, Morey LC, et al.
Dimensional representations of DSM-IV personality disorders: relationships
to functional impairment. Am J Psychiatry 2005b;162:191925.
Smith DJ, Muir WJ, Blackwood DH. Is borderline personality disorder part of
the bipolar spectrum? Harv Rev Psychiatry 2004;12:1339.
Suppes T, Mintz J, McElroy SL, Altshuler LL, Kupka RW, Frye MA, et al.
Mixed hypomania in 908 patients with bipolar disorder evaluated prospectively in the Stanley Foundation Bipolar Treatment Network: a sexspecific phenomenon. Arch Gen Psychiatry 2005;62:108996.
Swann AC, Gerard Moeller F, Steinberg JL, Schneider L, Barratt ES, Dougherty
DM. Manic symptoms and impulsivity during bipolar depressive episodes.
Bipolar Disord 2007;9:20612.
Tragesser SL, Solhan M, Schwartz-Mette R, Trull TJ. The role of affective
instability and impulsivity in predicting future BPD features. J Pers Disord
2007;21:60314.

1029

Walters GD, Gray NS, Jackson RL, Sewell KW, Rogers R, Taylor J, et al. A
taxometric analysis of the Psychopathy Checklist: Screening Version (PCL:
SV): further evidence of dimensionality. Psychol Assess 2007;19:3309.
doi:10.1037/1040-3590.19.3.330.
Wilson ST, Stanley B, Oquendo MA, Goldberg P, Zalsman G, Mann JJ.
Comparing impulsiveness, hostility, and depression in borderline personality
disorder and bipolar II disorder. J Clin Psychiatry 2007;68:15339.
Zanarini MC, Frankenburg FR, Vujanovic AA, Hennen J, Reich DB, Silk KR.
Axis II comorbidity of borderline personality disorder: description of 6-year
course and prediction to time-to-remission. Acta Psychiatr Scand 2004;110:
41620.
Zanarini MC, Frankenburg FR, Reich DB, Silk KR, Hudson JI, McSweeney LB.
The subsyndromal phenomenology of borderline personality disorder: a 10year follow-up study. Am J Psychiatry 2007;164:92935.

You might also like