<<

. 17
( 68 .)



>>


Sex 2.03 NS
Male 17 45 5 23
Female 21 55 17 77
Marital status NA
Single 22 58 8 36
Married 11 29 12 55
Separated 4 10 0 0
Widowed 1 3 2 9
First episode 0.13 NS
Manic/hypomanic 17 45 8 36
Depressive 21 55 14 64
Rapid cycling 0.35 NS
Present 5 13 5 23
Absent 33 87 17 77
Seasonal pattern 0.21 NS
Present 17 45 12 55
Absent 21 55 10 45
Psychotic symptoms 11.64 <0.001
Yes 34 90 10 45
No 4 10 12 55
Suicide attempts 0.00 NS
Yes 12 32 6 27
No 26 68 16 73
Familial history NA
Bipolar I 5 13 2 9
Bipolar II 1 3 4 18
Unipolar 8 24 6 27
Other 10 26 4 18
None 13 34 6 27
Familial history of suicide 0.85 NS
Yes 8 22 2 9
No 28 78 20 91

NS, not significant; NA, not applicable.
Reproduced with permission from Vieta et al. (1997a).
93 Bipolar I and bipolar II: a dichotomy?


we analyze all these data, it is essential to consider the influence of treatment
because many patients, before being diagnosed as bipolar II, could have been
misdiagnosed as personality disorders or unipolar depressive disorder, which
would have precluded the use of mood stabilizers that facilitated treatment with
tryciclic antidepressants, with a high risk of drug-induced switching (Vieta, 1999).
Figure 4.1 illustrates the natural course of some bipolar I and bipolar II patients
when antidepressant use was strictly controlled and all patients received mood
stabilizers.
Coryell et al. (1987, 1995) reported that bipolar II disorder seemed to be
phenomenologically stable, with a relatively low percentage of patients who
became bipolar I during follow-up. This would support the long-term validity of
the bipolar II category and the hypothesis of a true dichotomy. Akiskal et al. (1995)
performed an 11-year follow-up of a sample who presented a major depressive
episode. The results showed that 3.9% developed a full-blown manic episode,
becoming bipolar type I, and 8.6% presented with at least one episode of hypo-
mania, confirming that they belonged to bipolar II disorder. Unipolar patients
who switched to bipolar II disorder were characterized by early age at onset,
recurrent depression, high rates of divorce or separation, high rates of scholastic
and/or job maladjustment, isolated antisocial acts, drug abuse, and a broad
´
melange of atypical depressive symptoms with borderline taint. Mood lability
was the most specific predictor of which depressions would prospectively switch
to bipolar II disorder. The study testifies to the fact that bipolar II disorder is a
complex affective disorder with biographical instability deriving from an intense
temperamental dysregulation (Akiskal, 2002a, b). Other authors have suggested
that the depressive phase of bipolar II disorder features more atypical symptoms
than unipolar depression (Benazzi, 2000). Akiskal et al. (1983) proposed eight
criteria that are believed to be predictive of ˜˜bipolarization™™ from depression:
(1) treatment-induced hypomania; (2) family history of bipolar disorder; (3) strong
inheritability; (4) depression with hypersomnia and motor slowing-down;


100
80
Patients in 60
remission
40
(%)
20
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Bipolar I Bipolar II
Fig. 4.1 Course of bipolar I and bipolar II disorders. Reproduced with permission from Vieta (1999).
94 E. Vieta et al.


(5) psychotic depression; (6) continuous multigenerational familial transmission;
(7) postpartum onset; and (8) early onset. They suggested that these episodes should
be called ˜˜pseudounipolar depression™™ and that the condition would show its
bipolar nature afterwards. Bourgeois et al. (1996) confirmed the significantly higher
frequency of these markers on bipolar disorder. Goldberg et al. (2001) found that
depressed patients with psychosis at the index depressive episode were significantly
more likely than non-psychotic patients to demonstrate mania or hypomania at
follow-up.
With regard to outcome, Tsuang et al. (1979) found that, when marital,
residential, occupational, and psychiatric symptoms were combined, outcome
was good in 64% of bipolar I patients, intermediate in 14%, and poor in 22%.
Endicott et al. (1985) described a more chronic course in bipolar II than in bipolar
I disorder; this difference not only accounted for affective symptoms but also for
the presence of other psychiatric problems between episodes. Coryell et al. (1989)
found comparable degrees of psychosocial disability over time in bipolar II and
bipolar I patients, although bipolar II patients were particularly likely to report
work impairment at the end of a 5-year follow-up. Other authors have found that
bipolar I disorder was more incapacitating than bipolar II disorder (Vieta et al.,
2002a). As far as social impairment and functioning between episodes are con-
cerned, more than one-third of bipolar patients seem to have some chronic
symptoms; some of this pathology is sequelae of the episodes themselves, and
much of it reflects the absence of prophylactic or poor treatment (Goodwin and
Jamison, 1990). Judd et al. (2002) studied the weekly symptomatic status of
patients with bipolar I disorder during a prospective long-term follow-up; the
results showed that symptomatic structure was primarily depressive rather than
manic, and subsyndromal and minor affective symptoms predominated. A study
by Tohen et al. (2000) with 219 cases of first episode of major affective disorder
with psychotic features (159 manic or mixed bipolar patients) showed that
syndromal recovery was attained by most patients (98.6% of bipolar or mixed
patients) soon after hospitalization, but only one-third were functionally recovered
by 24 months (40.4% of bipolar or mixed patients); functional recovery was
associated with older age at onset and shorter hospitalization. The psychosocial
impairment related to relapse persists for years in a great number of bipolar
patients (Coryell et al., 1993; Keck et al., 1998).
Rapid cycling generally represents a transient phase in the course of bipolar
disorder, with a prevalence rate lower than 20% in most studies (Akiskal et al.,
2000). It is more likely to arise from a bipolar II base (Coryell et al., 1992;
Baldessarini et al., 2000), thus alternating at least four depressive or hypomanic
episodes per year.
95 Bipolar I and bipolar II: a dichotomy?


Several studies have reported a higher risk of suicide in bipolar II patients
(Dunner et al., 1976; Stallone et al., 1980; Dunner, 1983; Goldring and Fieve,
´
1984; Arato et al., 1988). Rihmer and Pestality (1999) reviewed the rate of lifetime
history of suicide attempts and found it was attempted by 17% of bipolar I
patients, 24% of bipolar II patients, and 12% of unipolar major depressive
patients. In addition, bipolar II patients were relatively overrepresented among
suicide victims. It was not only the personal history of suicide attempts, but also
the family history of completed suicide in first-degree relatives that was signific-
antly higher in bipolar II patients as compared to unipolar major depressives
(Rihmer, 2002). Interpersonal conflicts, marital instability, and/or family break-
down were found to be particularly more frequent among bipolar II patients with
respect to bipolar I and unipolar depressive patients, which, in the opinion of the
authors, might contribute to the high suicidality. Kupfer et al. (1988) found higher
past suicide attempts in a group of bipolar II patients than in a group of unipolar
depressive patients.
Other authors have shown that bipolar II patients with personality disorders
had a higher suicidal risk (Vieta et al., 1999). The presence of comorbidity seemed
to have no relevant impact on the clinical course of bipolar II patients except for
suicidality (Vieta et al., 2000). Previous studies did not show significant differences
between bipolar I and bipolar II patients with respect to suicidal behavior (Coryell
et al., 1989; Vieta et al., 1997b). These discrepancies may be due to different
definitions of bipolar II disorder, biases derived from the setting, and comorbidity.
In fact, it has been suggested that the exclusion of personality traits and substance
abuse might eliminate some of the differences between bipolar I and bipolar II
patients (Cooke et al., 1995). There is some evidence that bipolar II disorder is
more likely than unipolar disorder or bipolar I disorder to occur with other
psychiatric diagnoses (Endicott et al., 1985).


Pathophysiology
Unfortunately, most of the studies do not compare bipolar I and bipolar II
disorders. We will comment on some of the studies that take into account bipolar
subtypes.

Family studies and genetics
Family studies are useful for an understanding of the pathophysiology of bipolar
disorders and then to evaluate the dimensional model and the category model.
Some studies have concluded that bipolar II patients seem to have more bipolar II
and unipolar relatives and fewer bipolar I relatives than bipolar I patients do
(Coryell et al., 1984; Fieve et al., 1984). Coryell et al. (1984) not only found that
96 E. Vieta et al.


bipolar II probands were significantly more likely to have bipolar II relatives than
were non-bipolar or bipolar I probands, but also that bipolar II probands were
slightly more likely than non-bipolar probands and slightly less likely than bipolar I
probands to have relatives with bipolar I illness. A higher morbid risk of depression
among relatives of bipolar II patients with regard to the group of unipolar
depressive patients has been described (Kupfer et al., 1988). Gershon et al.
(1982) noted that bipolar I and bipolar II disorders were more frequent in relatives
of bipolar I and II patients than in relatives of unipolar depressives and controls.
A study from Andreasen et al. (1987) showed higher familial loads of bipolar I
disorders in bipolar I subjects and an increase of bipolar II illness in the relatives of
bipolar II probands. In contrast with the results of Andreasen et al. (1987), Heun
and Maier (1993) found that morbid risks for bipolar I disorder were equivalent in
relatives of bipolar I and bipolar II patients and lower in relatives of unipolar
subjects. However, the familial load for bipolar II disorder was higher in bipolar II
than in bipolar I subjects, and lower in unipolar depressives and in controls.
Bipolar II disorder has been reported to be the most prevalent affected phenotype
in both bipolar I and bipolar II families and the only expressed phenotype in half of
the bipolar II families (Simpson et al., 1993).
With regard to genetics, a recent study from McMahon et al. (2001) found that
paternal allele sharing on 18q21 was greatest in pairs where both siblings had
bipolar II“disorder. Prospective analysis confirmed the finding that bipolar
II“bipolar II siblings pairs showed significantly greater paternal allele sharing.
Paternal allele sharing across 18q21“23 was also significantly higher in families
with at least one bipolar II“bipolar II sibling pair. In these families, multipoint
affected sibling pair linkage analysis produced a peak paternal LOD score of 4.67
versus 1.53 in all families. Therefore, affected sibling pairs with bipolar II dis-
criminated between families who showed evidence linkage to 18q and families who
did not. Families with a bipolar II sibling pair produced a increased lod score and
improved linkage resolution.

Neuroimage
Kato et al. (1994) studied brain phosphorus metabolism in patients with bipolar I
and bipolar II disorder using magnetic resonance spectroscopy. Phosphocreatine
levels were lower in patients with bipolar II in the three states (euthymic, hypo-
manic, and depressed) compared to controls. High values of phosphomonoester
were found in bipolar II patients in the hypomanic and depressive states, but
phosphomonoester values in the euthymic state did not differ from controls.
Intracellular pH of bipolar II patients in all phases was similar to control values,
whereas euthymic bipolar I patients had lower pH values. The authors suggested
that brain high-energy phosphate metabolism might be impaired in bipolar II
97 Bipolar I and bipolar II: a dichotomy?


patients and that there might be pathophysiological differences between bipolar I
and bipolar II.
Differences have been found between bipolar I and II disorders on magnetic
resonance imaging scanning and on the presence of vascular abnormalities,
including Raynaud™s phenomenon, migraine, and migraine equivalents (Ferrier
et al., 2001). Endicott (1989) found a positive correlation between bipolarity and
a cluster of disturbances, including classic migraine headache, the peripheral
vascular disturbance of Raynaud™s disease, enuresis, vague episodic phenomena
similar to migraine prodrome, fingernail biting, and learning disorders; most of
the psychophysiological conditions showed a higher incidence in bipolar II
patients with respect to bipolar I patients.
By contrast, periventricular hyperintensities have been reported to be more
common in bipolar I patients than in bipolar II patients and normal comparison
subjects (Altshuler et al., 1995). It would be necessary to increase the number of
brain-imaging studies that separate bipolar I from bipolar II patients to know
whether a pathophysiologic brain marker could distinguish between bipolar I and
bipolar II disorders.
There were no quantitative magnetic resonance imaging studies of bipolar II
patients until Hauser et al. (2000) measured temporal lobe and ventricular struc-
tures in bipolar I, bipolar II, and control subjects. Their results showed that there
were no differences in temporal lobe or hippocampal volume estimates in the third
ventricle area and lateral ventricle-to-cerebrum area ratio among diagnostic
groups. The lateral ventricle area and the lateral ventricle-to-cerebrum area ratio
were significantly larger in bipolar I patients than either bipolar II patients or
control subjects only in the left hemisphere. Furthermore, these measures were
approximately twice as large in the bipolar I patients as in the other groups. This
study concluded that bipolar I disorder, particularly in males, might show differ-
ent neurobiological alterations compared to bipolar II or control subjects.
Positron emission tomography was used in a recent study by Berns et al. (2002) to
determine whether patients with bipolar II disorder had altered regional brain
responses to novel motor sequences with respect to healthy subjects. The results
showed that, in the comparison subjects, a spatial attention circuit in the superior
parietal lobe and supplementary motor area was activated in response to the intro-
duction of the new sequence. Bipolar II patients did not display this activation pattern;
instead, a widespread limbic network was activated in response to the new sequence.
Future neuroimaging research should study comparative functional neuroimag-
ing with single-photon emission computed tomography and positron emission
tomography of bipolar I disorder and bipolar II disorder. Neuroimaging could be
useful in validating diagnostic subtypes, although the results to date are too
unspecific (Benabarre et al., 2002).
98 E. Vieta et al.


Neurochemical studies
Siah et al. (1999) showed a trend towards higher platelet 5-hydroxytryptamine
(5-HT) levels in bipolar I and II depressions when compared to normal controls,
whereas there was no difference in platelet 5-HT levels between bipolar I and II
depressed patients. The finding of increased platelet 5-HT levels in bipolar
depressed patients compared to normal controls is consistent with the results of
previous studies (Wirz-Justice and Puhringer, 1978; Stahl et al., 1983), and might
suggest an increase in presynaptic 5-HT reuptake, presumably resulting from
diminished synaptic 5-HT availability in this condition. When bipolar I and II
patients were pooled, there was a trend toward a weak positive correlation between
platelet 5-HT and 21-item Hamilton Depression Rating Scale scores in the patient
groups, suggesting that the presumed deficiency of serotoninergic function might

<<

. 17
( 68 .)



>>